DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submissions, filed on August 21st, 2026, have been entered.
Status of Claims
Claims 1-20 are pending, claims 1, 8-10, 15-16 and 20 have been amended. Claims 1-20 remains rejected.
Response to Argument(s)
Regarding the 101 rejection:
The examiner finds the amendment to follow the examiner’s suggestion which was given during the conducted interview stated in the interview summary mailed on July 20th, 2026. The claimed invention now reflects a practical application wherein, the processing result is to identify a user demand for an item from an online platform, according to a determination of the processing result and determination of a similarity met a threshold, which indicates a specific conditioning for the practical application. Therefore, the examiner finds the amendment to narrow down the scope of the claims, reflect a practical application and improvement, the 101 rejection has been withdrawn.
In view of the Amendments to independent claims 1, 10 and 16 the previously applied prior art rejections are withdrawn. Applicants’ arguments are rendered moot in view of the new grounds of rejection set forth below.
The examiner recognizes the amendment has introduced new features into the claims hence, has narrowed down the scope of the claims and which requires new ground of rejections; though, regarding the Applicants’ arguments on some of the features of the claims that have been maintained, the examiner finds them to be non-persuasive, for the following features of the independent claims 1, 10 and 16:
“identifying…a query them based on the aggregated query image feature fata for the plurality of query images”
The Applicants state these features are not taught or suggested by the proposed prior art Ghadekar (previously used for the rejections of the independent claims 1, 10 and 16).
In support of the above argument, the Applicants state that, according to the rejection, Ghadekar’s equation (1) was mapped to the recited query theme. However, the Applicants find a transaction list of the items included in one transaction or multiple transactions are not equivalent to a query theme. A transaction involves buying or selling one or more items, that is the items are actually bought and not queried, which is in contrast with the claimed limitation.
Examiner’s reply:
The examiner respectfully disagrees with the Applicants’ argument. The Applicants are reminded that the claims are construed based on BRI (broadest reasonable interpretation) in light of the specification. Therefore, the examiner finds the recited “query theme” to still fall within the scope of the reference Ghadekar, particularly, the examiner finds the recited “query theme” to indicate a collection of data representation that relates to a query/search, hence, Ghadakar’s transaction list of equation (1) in its section V includes a data representation/transaction list which indicates a theme, moreover, since Ghadekar’s transaction, in its section V, 1st 2 Pars. state “detect patterns in large databases. The number of patterns detected depends on the size of the database…example {pencil, eraser => {notebook}. The above association states that if a person buys a pencil and an eraser, then he is most likely to buy a notebook….the recommendations” which is an association or a rule A => B, which is further stated in, section V, 4th Par., of “the rule A => B holes in the transaction set ‘D’ with support ‘s’, where ‘s’ is the percentage of transactions in ‘D’”, wherein the transaction set include transactions that are used to find association between items when being purchased would go together, therefore, the transaction set is the base dataset used to train the neural network to make such prediction for a user’s search/query, therefore, the transaction set including the transactions represent a theme in purchasing trend, or in other words, a query theme. Therefore, any of the transaction of the transaction set is part of the purchase representation, or data of a query theme.
Therefore, the examiner finds this claimed limitation is still taught by the reference Ghadekar.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitation(s) that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that use the word “means” or “step” but are nonetheless not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph because the claim limitation(s) recite(s) sufficient structure, materials, or acts to entirely perform the recited function.
Claims 1, 6, 10 and 11-15, recite(s) limitation(s) that use words like “means” (or “step”) or similar terms with functional language and do invoke 35 U.S.C. 112(f):
1) claim 1 recites, “receiving, by an image tagging module….,” “generating, by the image tagging module,….” [line 3, line 5];
2) claims 10 and 12 recite, “an image tagging module configured to generate….” [line 2] and [line 1];
furthermore, additional claim limitations evoke 112(f) are:
3) claims 1 and 6 recite, “analyzing, by an image content analyzer module…..” [line 7] and [line 3];
4) claims 10 and 13 recite “an image content analyzer module configured to….” [line 5] and [line 3];
5) claim 1 recites “identifying, a theme identification module….” as recited in claim 1, [line 13];
6) claims 10 and 14 recite “a theme identification module configured to….” [line 9] and [line 1];
7) claim 1 recites “generating, by a demand analysis module….” [line 15];
8) claims 10-11 and 15 recite “a demand detection module configured to….” [line 13] and [lines 1-2] and [lines 1-2].
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
After a careful analysis, as disclosed above, and a careful review of the specification the following limitations in claims 1, 6, 10 and 11-15;
(i) “an image tagging module” the instant specification, filed on October 02nd, 2023, provides sufficient structure, material and/or act for the recited image tagging module to perform the recited function, wherein Par. [0054], and Figure 8, discloses “the computing system executes program code that configures the processor to perform one or more of the operations…examples of the program code include, in various embodiments, the application executed an image tagging module”, thus have sufficient structure or material/algorithm, act wherein is a program being implemented by a processor.).
(ii) “an image content analyzer module” the instant specification, filed on October 02nd, 2023, provides sufficient structure, material and/or act for the recited image tagging module to perform the recited function, wherein Par. [0054], and Figure 8, discloses “the computing system executes program code that configures the processor to perform one or more of the operations…examples of the program code include, in various embodiments, the application executed…an image content analyzer module”, thus have sufficient structure or material/algorithm, act wherein is a program being implemented by a processor.).
(iii) “a theme identification module” the instant specification, filed on October 02nd, 2023, provides sufficient structure, material and/or act for the recited image tagging module to perform the recited function, wherein Par. [0054], and Figure 8, discloses “the computing system executes program code that configures the processor to perform one or more of the operations…examples of the program code include, in various embodiments, the application executed…a theme identification module”, thus have sufficient structure or material/algorithm, act wherein is a program being implemented by a processor.).
(iv) “a demand analysis module” there is no sufficient structure, material and/or act for the recited demand analysis module to perform the recited function, the closest disclosure can be found in Par. [0054], and Figure 8, of “the computing system executes program code that configures the processor to perform one or more of the operations…examples of the program code include, in various embodiments, the application executed…a demand detection module” however, stating a program implemented by a processor for a demand detection module and not explicitly stating that it’s the same as for the demand analysis module, thus have insufficient structure, material and/or act.
(v) “a demand detection module” the instant specification, filed on October 02nd, 2023, provides sufficient structure, material and/or act for the recited image tagging module to perform the recited function, wherein Par. [0054], and Figure 8, discloses “the computing system executes program code that configures the processor to perform one or more of the operations…examples of the program code include, in various embodiments, the application executed…a demand detection module”, thus have sufficient structure or material/algorithm, act wherein is a program being implemented by a processor.).
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 1 along with their dependent claims are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor, or for pre-AIA the applicant regards as the invention.
Claim 1’s limitations:
“generating, by a demand analysis module….” as recited in claim 1, line 15.
Claim 1 respectively invokes 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. However, the written description fails to disclose the corresponding structure, material, or acts for performing the entire claimed function and to clearly link the structure, material, or acts to the function. The specification is devoid of adequate structure to perform the claimed functions. The specification does not provide sufficient details such that one of the ordinary skill in the art would understand which structure performed(s) the claimed function.
Therefore, the claim is indefinite and is rejected under 35 U.S.C. 112(b) or pre-AIA 35 U.S.C. 112, second paragraph.
Applicant may:
(a) Amend the claim so that the claim limitation will no longer be interpreted as a limitation under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph;
(b) Amend the written description of the specification such that it expressly recites what structure, material, or acts perform the entire claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(c) Amend the written description of the specification such that it clearly links the structure, material, or acts disclosed therein to the function recited in the claim, without introducing any new matter (35 U.S.C. 132(a)).
If applicant is of the opinion that the written description of the specification already implicitly or inherently discloses the corresponding structure, material, or acts and clearly links them to the function so that one of ordinary skill in the art would recognize what structure, material, or acts perform the claimed function, applicant should clarify the record by either:
(a) Amending the written description of the specification such that it expressly recites the corresponding structure, material, or acts for performing the claimed function and clearly links or associates the structure, material, or acts to the claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(b) Stating on the record what the corresponding structure, material, or acts, which are implicitly or inherently set forth in the written description of the specification, perform the claimed function. For more information, see 37 CFR 1.75(d) and MPEP §§ 608.01(o) and 2181.
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claim 1 along with their dependent claims are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for pre-AIA the inventor(s), at the time the application was filed, had possession of the claimed invention. As described above, the disclosure does not provide adequate structure to perform the claimed function in the recited limitation.
Claim 1’s limitations:
“generating, by a demand analysis module….” as recited in claim 1, line 15.
The specification does not demonstrate that applicant has made an invention that achieves the claimed function because the invention is not described with sufficient detail such that one of ordinary skill in the art can reasonably conclude that the inventor had possession of the claimed invention.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 3-5, 7-8, 10-12, 14-18 and 20 rejected under 35 U.S.C. 103 as being unpatentable over Premanand Ghadekar et. al. (“Image-Based Product Recommendations Using Market Basket Analysis, Sept. 2019, 2019 5th International Conference on Computing, Communication, Control And Automation” hereinafter as “Ghadekar”) further in view of Biao Lu et. al. (“US 2024/0265309 A1” hereinafter as “Lu”) and Dinesh Kumar et. al. (“US 2023/0214696 A1” hereinafter as “Kumar”).
Regarding claim 1, Ghadekar teaches a method performed by one or more processing devices (Abstract discloses the invention uses neural network which can be understood to include the use of processing devices such as computer), comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time (Section VII, 1st par., discloses a scenario where the user upload an image of a product to the smart engine and the smart engine recommend related products, the uploaded image here is analogous to the query image as claimed and the smart engine is analogous to the image tagging module as claimed; moreover, in section VI’s “Collaborative Filtering User-User Similarity” section, it discloses the system identifies another user who is similar and has same interests as that of the current user include comparison involves liking of different users, indicating that the system has collected information regarding other user’s performing the same querying steps including a plurality of images have been updated for search during a period of time); generating, by the image tagging module, multiple image tags for each query image of the plurality of query images (Section VII, 2nd par., discloses identifying distinct products in the image by a CNN, wherein the identified information include contents of the product [analogous to multiple image tags as claimed, such as the labels given to each product in the image as shown in FIG. 1, indicating that when an image includes multiple products, would include multiple labels such as an image of a pencil and an eraser together such as shown in Section V, 1st par.], as part of the smart engine moreover, in section VI’s “Collaborative Filtering User-User Similarity” section, it discloses the system identifies another user who is similar and has same interests as that of the current user include comparison involves liking of different users, this also indicates that the system has collected information regarding other user’s performing the same querying steps, it can be understood that the same image tags have been generated for these images); analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph (Section V, 1st 3 pars., discloses when the image labels are identified, the association rule mining is used to state association between the products in the image with another product likely to be bought or be interested in by the user, the association rule here can be understood as a knowledge graph as claimed) associated with an online platform (Section VII, 1st Par. discloses “the proposed model provides a smart engine for online websites”) to create query image feature data for each query image (Section V, 1st 3 pars., discloses when the image labels are identified, the association rule mining is used to state association between the products in the image with another product likely to be bought or be interested in by the user, the association rule here can be understood as a knowledge graph as claimed, and the association here can be understood to be the recited query image feature data as claimed; and the machine learning for the associations rule mining here is analogous to the recited image content analyzer module as claimed; the same feature data has been analyzed for other users’ steps); aggregating the query image feature data for each query image of the plurality of query images (Section IX, 1st Par., discloses “associations are generated based on the most frequent products bought by similar users” indicating different users perform the query [each query image of the plurality of query images as claimed], Section V, 2nd par., discloses “the above association states that if a person buys a pencil and an eraser, the he is most likely to buy a notebook. However, in the proposed solution, image will be provided as an input, and the recommendations generated will also be imaged”, and the associations/recommendations are generated through showing a list of products [section VII, 1st Par., discloses “the system will identify the product and would then recommend the related products to the user…the user is shown the list of products”], wherein recommending the list of products indicating generating an aggregated list of products through the associations/recommendations) to obtain aggregated query image feature data for the plurality of query images (Section V, 2nd par., discloses “the above association states that if a person buys a pencil and an eraser, the he is most likely to buy a notebook. However, in the proposed solution, image will be provided as an input, and the recommendations generated will also be imaged”, and the associations/recommendations are generated through showing a list of products [section VII, 1st Par., discloses “the system will identify the product and would then recommend the related products to the user…the user is shown the list of products”], wherein recommending the list of products indicating generating an aggregated list of products through the associations/recommendations); identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images (The association as discussed previously, as disclosed in section V, 1st 2 paragraphs, to provide transaction list of what to be recommended to the user such as the list of equation (1) in section V, the list here can be understood to be a query theme as claimed, based on aggregated query image feature data, since each item in the list if obtained from an association, the list includes aggregated list of associated items to be recommended to the user; the processor programed to execute this step is analogous to the theme identification module as claimed; the same identifying step can be understood to be performed for other users’ performance of the system); and identifying, by a demand analysis module, a user demand for an item (Section VI discloses the output of the processing is to provide the user with recommendation of items based on the input image of what other users have had purchased in the past to follow interest rule of what items the current user may be interested in, it can be understood as a demand data has been generated of what users would be interested in of which products are being in demand according to the input image, moreover, the recommendation here is based on the output of equation 1 of section V; the processor with program to perform this steps can be understood to be analogous to the recited demand analysis module as claimed; any user demand would be indicating a level of demand such as being analogous to the recited “indicating a level of user demand”) from the plurality of users that the online platform has not been met (Section VI., “Collaborative Filtering User-User Similarity” section, discloses “products are recommended to the user based on similar user liking. The system identifies another user who is similar and has same interests as that of the current user and recommends the products which the current user lacks. Hence the comparison involves liking of different users with the same taste” wherein, “the other users’ similar likings” indicating “from the plurality of users” moreover, “recommending the products which the current user lacks” indicating “that the online platform has not been met” or the products that the current user has not thought of or met using the current online platform) based on comparing the query theme to catalog data of the online platform (The demand data is based on comparing the input image to the images of what other users purchased in the past of catalog data of an online platform, such as finding the similarity between user-user according to section VI, since determining of similarity indicates a comparison).
However, Ghadekar does not explicitly teach the knowledge graph comprising descriptions of items provided by the online platform.
Lu teaches the knowledge graph (Pars [0014-0015] disclose “the target recommendation model further indicates an association relationship between at least two sample items…a relationship between items and a relationship between objects are established, so that information hidden in a graph structure can be further used to effectively use the other information in the entire graph structure” wherein, the association here is analogous to Ghadekar’s association rule being the knowledge graph further to include a graph showing the relationship here in Lu) comprising descriptions of items provided by the online platform (Par. [0036] discloses “the item may include at least one of the text, the link, the product, the serve, and the concept information” wherein, concept information, text indicating descriptions; moreover, Par. [0036] discloses “the item recommendation…new application program” such as a website according to Par. [0099]).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar of having a method performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images; analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph associated with an online platform to create query image feature data for each query image; aggregating the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images; identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images; and identifying, by a demand analysis module, a user demand for an item from the plurality of users that the online platform has not been met based on comparing the query theme to catalog data of the online platform, with the teachings of Lu of having the knowledge graph comprising descriptions of items provided by the online platform.
Wherein having Ghadekar’s method of having the knowledge graph comprising descriptions of items provided by the online platform.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve target recommendation by using graph neural network so that model operation time is greatly reduced. Since both Ghadekar and Lu both perform target product recommendation based on machine learning. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Lu’s system improves target recommendation by using graph neural network so that model operation time is greatly reduced, (see Lu’s Abstract).
However, Ghadekar in view of Lu does not explicitly teach determining that a similarity between the query theme and the catalog data of the online platform is below a predetermined threshold.
Kumar teaches determining that a similarity (Par. [0021] discloses “the cluster model uses a technique based on grouping objects or dividing a population based on similarities”) between the query theme and the catalog data (Par. [0031] discloses “the weighted sums of the similarity score calculated based on the number of purchases completed by the user with merchants may be used” wherein the purchases is analogous to Ghadekar’s transactions being analogous to the recited query theme) of the online platform (Par. [0061] discloses “the other applications may include security applications…that enable designation of a primary interactive device, and applications that allow for web site searches…including merchant websites”) is below a predetermined threshold (Par. [0031] discloses “a charity having the highest score for a user may then be recommended to the user” indicating a threshold to be met).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar in view of Lu of having a method performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images; analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph associated with an online platform to create query image feature data for each query image; aggregating the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images; identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images; and identifying, by a demand analysis module, a user demand for an item from the plurality of users that the online platform has not been met based on comparing the query theme to catalog data of the online platform, with the teachings of Kumar of having determining that a similarity between the query theme and the catalog data of the online platform is below a predetermined threshold.
Wherein having Ghadekar’s method of having determining that a similarity between the query theme and the catalog data of the online platform is below a predetermined threshold.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve user’s recommendation system that is tailored to the user’s interests, likes, purchase, and/or associations, which increase efficiency for a user interactive system. Since both Ghadekar and Kumar both perform target product recommendation based on cross-domain collaborative filtering. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Kumar’s system improves user’s recommendation system that is tailored to the user’s interests, likes, purchase, and/or associations, which increase efficiency for a user interactive system, (see Kumar’s Par. [0003]).
Regarding claim 3, Ghadekar in view of Lu and Kumar teaches the method of claim 1, Ghadekar teaches further comprising receiving multiple query images associated with a user query on the online platform (Since the processing of Ghadekar is based on other users’ clicked images or input images to generate the associations rule for recommendation based on user-user similarity as disclosed in section VI, “Collaborative Filtering User-User Similarity” section, it can be understood to have the use of receiving multiple query images associated with other user query on the online platform), wherein the multiple query images comprise image metadata comprising user action data associated with aspects of one or more of the multiple query images (Since the multiple query images are images of other users’ clicked images or purchasing images or input images, they indicates metadata associated with multiple query images, according to section VI, “Collaborative Filtering User-User Similarity” section).
Regarding claim 4, Ghadekar in view of Lu and Kumar teaches the method of claim 1, Ghadekar teaches wherein the multiple image tags comprise object categories, object names, and object attributes (The product contents and labels being image tags, as discussed previously in claim 1, such as shown in FIG. 1 of object names/labels, and confidence value [such as shown in FIG. 6 which can be understood to be object attributes, since a confidence value is an attribute of the object in the image]; moreover, the output also include an association rule such as disclosed in section IX, 1st 2 pars, which can be understood to be object category as claimed).
Regarding claim 5, Ghadekar in view of Lu and Kumar teaches the method of claim 1, Ghadekar teaches wherein the online platform is a marketplace platform (Section I, 1st par., discloses the platform being e-commerce website), and wherein the knowledge graph comprises product information for products associated with the marketplace platform (The knowledge graph include product information as association information between products from the marketplace platform; Section V, 1st 3 pars., discloses when the image labels are identified, the association rule mining is used to state association between the products in the image with another product likely to be bought or be interested in by the user, the association rule here can be understood as a knowledge graph as claimed).
Regarding claim 7, Ghadekar in view of Lu and Kumar teaches the method of claim 1, Ghadekar teaches further comprising: aggregating the query image feature data for the plurality of query images corresponding to the plurality of user queries to generate the aggregated query image feature data (The association as discussed previously, as disclosed in section V, 1st 2 paragraphs, to provide transaction list of what to be recommended to the user such as the list of equation (1) in section V, the list here can be understood to be a query theme as claimed, based on aggregated query image feature data, since each item in the list if obtained from an association, the list includes aggregated list of associated items to be recommended to the user; the processor programed to execute this step is analogous to the theme identification module as claimed; the items from the list is based on images from other users’ queries); and storing the aggregated query image feature data (Processing the aggregated data indicates a storing of the data, as disclosed in section V, 1st 2 paragraphs, to provide transaction list of what to be recommended to the user such as the list of equation (1) in section V).
Regarding claim 8, Ghadekar in view of Lu and Kumar teaches the method of claim 1, Ghadekar teaches further comprising transmitting data associated with the user demand to the online platform (The recommendation as discussed previously in claim 1 is being shown to the user using the online shopping website, is analogous to the demand data being transmitted to the website to show to the user; the demand data is based on comparing the input image to the images of what other users purchased in the past of catalog data of an online platform, such as finding the similarity between user-user according to section VI, since determining of similarity indicates a comparison), wherein the demand data comprises a synthetic object image generated based on the demand data (The recommendation data includes images output by the system [generated from the processing has been mentioned] as shown in FIG. 7, according to section VI).
Regarding claim 10, Ghadekar teaches a system, comprising (Abstract discloses the invention uses neural network which can be understood to include the use of processing devices as a system such as computer): an image tagging module configured to generate multiple image tags (Section III, “Pre-processing and Pruning” section, discloses “to send the images for training, two lists have been created, one for labels and the other one for images”) for each query image of a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time (Section VII, 1st par., discloses a scenario where the user upload an image of a product to the smart engine and the smart engine recommend related products; the uploaded image here is analogous to the query image as claimed and the smart engine is analogous to the image tagging module as claimed; moreover, in section VI’s “Collaborative Filtering User-User Similarity” section, it discloses the system identifies another user who is similar and has same interests as that of the current user include comparison involves liking of different users, indicating that the system has collected information regarding other user’s performing the same querying steps including a plurality of images have been updated for search during a period of time); an image content analyzer module configured to analyze the multiple image tags (Section III, “Pre-processing and Pruning” section, discloses “one for labels and the other one for images” wherein, the data being analyzed for machine learning) based on a knowledge graph associated with an online platform (Section VII, 1st Par. discloses “the proposed model provides a smart engine for online websites”) to create query image feature data from each query image (section VII, 2nd par., discloses identifying distinct products in the image by a CNN, wherein the identified information include contents of the product [analogous to multiple image tags as claimed, such as the labels given to each product in the image as shown in FIG. 1, when an image includes multiple products, include multiple labels such as an image of a pencil and an eraser together such as shown in section V, 1st par.], as part of the smart engine; section V, 1st 3 pars, discloses when the image labels are identified, the association rule mining is used to state association between the products in the image with another product likely to be bought or be interested in by the user; and the machine learning for the associations rule mining here is analogous to the recited image content analyzer module as claimed; moreover, in section VI’s “Collaborative Filtering User-User Similarity” section, it discloses the system identifies another user who is similar and has same interests as that of the current user include comparison involves liking of different users, this also indicates that the system has collected information regarding other user’s performing the same querying steps, it can be understood that the same image tags have been generated for these images); a theme identification module configured to aggregate the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images (Section IX, 1st Par., discloses “associations are generated based on the most frequent products bought by similar users” indicating different users perform the query [each query image of the plurality of query images as claimed], Section V, 2nd par., discloses “the above association states that if a person buys a pencil and an eraser, the he is most likely to buy a notebook. However, in the proposed solution, image will be provided as an input, and the recommendations generated will also be imaged”, and the associations/recommendations are generated through showing a list of products [section VII, 1st Par., discloses “the system will identify the product and would then recommend the related products to the user…the user is shown the list of products”], wherein recommending the list of products indicating generating an aggregated list of products through the associations/recommendations) and identify a query theme based on aggregated query image feature data for the plurality of query images (The association as discussed previously, as disclosed in section V, 1st 2 paragraphs, to provide transaction list of what to be recommended to the user such as the list of equation (1) in section V, the list here can be understood to be a query theme as claimed, based on aggregated query image feature data, since each item in the list if obtained from an association, the list includes aggregated list of associated items to be recommended to the user; the processor programed to execute this step is analogous to the theme identification module as claimed; the same feature data has been analyzed for other users’ steps); and a demand detection module configured to identify a user demand for an item from the plurality of users that the online platform has not been met (Section VI., “Collaborative Filtering User-User Similarity” section, discloses “products are recommended to the user based on similar user liking. The system identifies another user who is similar and has same interests as that of the current user and recommends the products which the current user lacks. Hence the comparison involves liking of different users with the same taste” wherein, “the other users’ similar likings” indicating “from the plurality of users” moreover, “recommending the products which the current user lacks” indicating “that the online platform has not been met” or the products that the current user has not thought of or met using the current online platform) based on comparing the query theme to catalog data of the online platform (Section VI discloses the output of the processing is to provide the user with recommendation of items based on the input image of what other users have had purchased in the past to follow interest rule of what items the current user may be interested in, it can be understood as a demand data has been generated of what users would be interested in of which products are being in demand according to the input image, moreover, the recommendation here is based on the output of equation 1 of section V; the processor with program to perform this steps can be understood to be analogous to the recited demand detection module as claimed; any user demand would be indicating a level of demand such as being analogous to the recited “indicating a level of user demand”).
However, Ghadekar does not explicitly teach the knowledge graph comprising descriptions of items provided by the online platform.
Lu teaches the knowledge graph (Pars [0014-0015] disclose “the target recommendation model further indicates an association relationship between at least two sample items…a relationship between items and a relationship between objects are established, so that information hidden in a graph structure can be further used to effectively use the other information in the entire graph structure” wherein, the association here is analogous to Ghadekar’s association rule being the knowledge graph further to include a graph showing the relationship here in Lu) comprising descriptions of items provided by the online platform (Par. [0036] discloses “the item may include at least one of the text, the link, the product, the serve, and the concept information” wherein, concept information, text indicating descriptions; moreover, Par. [0036] discloses “the item recommendation…new application program” such as a website according to Par. [0099]).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar of having a system performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images; analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph associated with an online platform to create query image feature data for each query image; aggregating the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images; identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images; and identifying, by a demand analysis module, a user demand for an item from the plurality of users that the online platform has not been met based on comparing the query theme to catalog data of the online platform, with the teachings of Lu of having the knowledge graph comprising descriptions of items provided by the online platform.
Wherein having Ghadekar’s system of having the knowledge graph comprising descriptions of items provided by the online platform.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve target recommendation by using graph neural network so that model operation time is greatly reduced. Since both Ghadekar and Lu both perform target product recommendation based on machine learning. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Lu’s system improves target recommendation by using graph neural network so that model operation time is greatly reduced, (see Lu’s Abstract).
However, Ghadekar in view of Lu does not explicitly teach determining that a similarity between the query theme and the catalog data of the online platform is below a predetermined threshold.
Kumar teaches determining that a similarity (Par. [0021] discloses “the cluster model uses a technique based on grouping objects or dividing a population based on similarities”) between the query theme and the catalog data (Par. [0031] discloses “the weighted sums of the similarity score calculated based on the number of purchases completed by the user with merchants may be used” wherein the purchases is analogous to Ghadekar’s transactions being analogous to the recited query theme) of the online platform (Par. [0061] discloses “the other applications may include security applications…that enable designation of a primary interactive device, and applications that allow for web site searches…including merchant websites”) is below a predetermined threshold (Par. [0031] discloses “a charity having the highest score for a user may then be recommended to the user” indicating a threshold to be met).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar in view of Lu of having a system performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images; analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph associated with an online platform to create query image feature data for each query image; aggregating the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images; identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images; and identifying, by a demand analysis module, a user demand for an item from the plurality of users that the online platform has not been met based on comparing the query theme to catalog data of the online platform, with the teachings of Kumar of having determining that a similarity between the query theme and the catalog data of the online platform is below a predetermined threshold.
Wherein having Ghadekar’s system of having determining that a similarity between the query theme and the catalog data of the online platform is below a predetermined threshold.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve user’s recommendation system that is tailored to the user’s interests, likes, purchase, and/or associations, which increase efficiency for a user interactive system. Since both Ghadekar and Kumar both perform target product recommendation based on cross-domain collaborative filtering. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Kumar’s system improves user’s recommendation system that is tailored to the user’s interests, likes, purchase, and/or associations, which increase efficiency for a user interactive system, (see Kumar’s Par. [0003]).
Regarding claim 11, Ghadekar in view of Lu and Kumar teaches the system of claim 10, Ghadekar teaches wherein the demand detection module is further configured to: determine a query result by comparing each query image associated with a user query to the catalog data of the online platform (The demand data is based on comparing the input image to the images of what other users purchased in the past of catalog data of an online platform, such as finding the similarity between user-user according to section VI, since determining of similarity indicates a comparison); and provide the query result to a user computing device associated with the user query (The recommendation result is shown to the user at the website; Section VI., “Collaborative Filtering User-User Similarity” section, discloses “products are recommended to the user based on similar user liking. The system identifies another user who is similar and has same interests as that of the current user and recommends the products which the current user lacks. Hence the comparison involves liking of different users with the same taste”).
Regarding claim 12, Ghadekar in view of Lu and Kumar teaches the system of claim 10, Ghadekar teaches wherein the image tagging module is configured to receive multiple query images associated with a user query on the online platform (Since the processing of Ghadekar is based on other users’ clicked images or input images to generate the associations rule for recommendation based on user-user similarity as disclosed in section VI; it can be understood to have the use of receiving multiple query images associated with other user query on the online platform), wherein the multiple query images comprise image metadata comprising user action data associated with aspects of one or more of the multiple query images (Since the multiple query images are images of other users’ clicked images or purchasing images or input images, hence, they indicates metadata associated with multiple query images as disclosed in section V).
Regarding claim 14, Ghadekar in view of Lu and Kumar teaches the system of claim 10, Ghadekar teaches wherein the theme identification module comprises a clustering model configured to cluster the aggregated query image feature data to one or more query clusters, wherein the one or more query clusters correspond to one or more query themes (The association as discussed previously, as disclosed in section V, 1st 2 paragraphs, to provide transaction list of what to be recommended to the user such as the list of equation (1) in section V, the list here can be understood to be a query theme as claimed, based on aggregated query image feature data, since each item in the list if obtained from an association, the list includes aggregated list of associated items to be recommended to the user; the processor programed to execute this step is analogous to the theme identification module; the items from the list is based on images from other users’ queries, according to the association rule [according to the query them as claimed]; wherein the recommendation processing is based on a support calculation such as shown in equation 2 of section V, hence, indicates a clustering technique).
Regarding claim 15, Ghadekar in view of Lu and Kumar teaches the system of claim 10, Ghadekar teaches wherein the demand detection module is further configured to transmit data associated with the user demand to the online platform (The recommendation as discussed previously in claim 10 is being shown to the user using the online shopping website, hence, it can be understood as the demand data being transmitted to the website to show to the user; Section VII, 1st par., discloses a scenario where the user upload an image of a product to the smart engine and the smart engine recommend related product), wherein the demand data comprises a synthetic object image generated based on the query theme (The recommendation data includes images output by the system [generated from the processing has been mentioned] based on the transaction list [the query theme], as shown in FIG. 7 and Section VII, 1st Par.).
Regarding claim 16, Ghadekar teaches a non-transitory computer-readable medium, storing executable instructions, which when executed by a processing device, cause the processing device to perform operations comprising (Abstract discloses the invention uses neural network which can be understood to include the use of processing devices such as computer which includes a process to execute instructions of the invention stored in a non-transitory storage medium such as a ROM or RAM), receiving a plurality of query images corresponding to a plurality of user queries from a plurality of users via an online platform (Section VII, 1st par., discloses a scenario where the user upload an image of a product to the smart engine and the smart engine recommend related products; the uploaded image here is analogous to the query image as claimed and the smart engine is analogous to the image tagging module as claimed, of an online platform e-commerce website according to sect. I, 1st par.; moreover, in section VI’s “Collaborative Filtering User-User Similarity” section, it discloses the system identifies another user who is similar and has same interests as that of the current user include comparison involves liking of different users, this also indicates that the system has collected information regarding other user’s performing the same querying steps including a plurality of images have been updated for search during a period of time); generating multiple image tags for each query image of the plurality of query images (Section VII, 2nd par., discloses identifying distinct products in the image by a CNN, wherein the identified information include contents of the product [analogous to multiple image tags as claimed, such as the labels given to each product in the image as shown in FIG. 1, when an image includes multiple products, it would include multiple labels such as an image of a pencil and an eraser together such as shown in section V, 1st par.], as part of the smart engine); analyzing the multiple image tags based on a knowledge graph to create query image feature data for each query image (Section V, 1st 3 pars, discloses when the image labels are identified, the association rule mining is used to state association between the products in the image with another product likely to be bought or be interested in by the user, the association rule here can be understood as a knowledge graph as claimed, and the association here can be understood to be the recited query image feature data as claimed; and the machine learning for the associations rule mining here is analogous to the recited image content analyzer module as claimed); aggregating the query image feature data for each query image of the plurality of query images (Section IX, 1st Par., discloses “associations are generated based on the most frequent products bought by similar users” indicating different users perform the query [each query image of the plurality of query images as claimed], Section V, 2nd par., discloses “the above association states that if a person buys a pencil and an eraser, the he is most likely to buy a notebook. However, in the proposed solution, image will be provided as an input, and the recommendations generated will also be imaged”, and the associations/recommendations are generated through showing a list of products [section VII, 1st Par., discloses “the system will identify the product and would then recommend the related products to the user…the user is shown the list of products”], wherein recommending the list of products indicating generating an aggregated list of products through the associations/recommendations) to obtain aggregated query image feature data for the plurality of query images (Section V, 2nd par., discloses “the above association states that if a person buys a pencil and an eraser, the he is most likely to buy a notebook. However, in the proposed solution, image will be provided as an input, and the recommendations generated will also be imaged”, and the associations/recommendations are generated through showing a list of products [section VII, 1st Par., discloses “the system will identify the product and would then recommend the related products to the user…the user is shown the list of products”], wherein recommending the list of products indicating generating an aggregated list of products through the associations/recommendations); identifying a query theme based on aggregated query image feature data for the plurality of query images (The association as discussed previously, as disclosed in section V, 1st 2 paragraphs, to provide transaction list of what to be recommended to the user such as the list of equation (1) in section V, the list here can be understood to be a query theme as claimed, based on aggregated query image feature data, since each item in the list if obtained from an association, the list includes aggregated list of associated items to be recommended to the user; the processor programed to execute this step is analogous to the theme identification module as claimed); and identifying a user demand for an (Section VI discloses the output of the processing is to provide the user with recommendation of items based on the input image of what other users have had purchased in the past to follow interest rule of what items the current user may be interested in, as a demand data has been generated of what users would be interested in of which products are being in demand according to the input image, moreover, the recommendation here is based on the output of equation 1 of section V; the processor with program to perform this steps can be understood to be analogous to the recited demand analysis module as claimed) item from the plurality of users that the online platform has not been met (Section VI., “Collaborative Filtering User-User Similarity” section, discloses “products are recommended to the user based on similar user liking. The system identifies another user who is similar and has same interests as that of the current user and recommends the products which the current user lacks. Hence the comparison involves liking of different users with the same taste” wherein, “the other users’ similar likings” indicating “from the plurality of users” moreover, “recommending the products which the current user lacks” indicating “that the online platform has not been met” or the products that the current user has not thought of or met using the current online platform) based on comparing the query theme to catalog data of the online platform (The demand data is based on comparing the input image to the images of what other users purchased in the past of catalog data of an online platform, such as finding the similarity between user-user according to section VI, since determining of similarity indicates a comparison).
However, Ghadekar does not explicitly teach the knowledge graph comprising descriptions of items provided by the online platform.
Lu teaches the knowledge graph (Pars [0014-0015] disclose “the target recommendation model further indicates an association relationship between at least two sample items…a relationship between items and a relationship between objects are established, so that information hidden in a graph structure can be further used to effectively use the other information in the entire graph structure” wherein, the association here is analogous to Ghadekar’s association rule being the knowledge graph further to include a graph showing the relationship here in Lu) comprising descriptions of items provided by the online platform (Par. [0036] discloses “the item may include at least one of the text, the link, the product, the serve, and the concept information” wherein, concept information, text indicating descriptions; moreover, Par. [0036] discloses “the item recommendation…new application program” such as a website according to Par. [0099]).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar of having a system performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images; analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph associated with an online platform to create query image feature data for each query image; aggregating the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images; identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images; and identifying, by a demand analysis module, a user demand for an item from the plurality of users that the online platform has not been met based on comparing the query theme to catalog data of the online platform, with the teachings of Lu of having the knowledge graph comprising descriptions of items provided by the online platform.
Wherein having Ghadekar’s system of having the knowledge graph comprising descriptions of items provided by the online platform.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve target recommendation by using graph neural network so that model operation time is greatly reduced. Since both Ghadekar and Lu both perform target product recommendation based on machine learning. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Lu’s system improves target recommendation by using graph neural network so that model operation time is greatly reduced, (see Lu’s Abstract).
However, Ghadekar in view of Lu does not explicitly teach determining that a similarity between the query theme and the catalog data of the online platform is below a predetermined threshold.
Kumar teaches determining that a similarity (Par. [0021] discloses “the cluster model uses a technique based on grouping objects or dividing a population based on similarities”) between the query theme and the catalog data (Par. [0031] discloses “the weighted sums of the similarity score calculated based on the number of purchases completed by the user with merchants may be used” wherein the purchases is analogous to Ghadekar’s transactions being analogous to the recited query theme) of the online platform (Par. [0061] discloses “the other applications may include security applications…that enable designation of a primary interactive device, and applications that allow for web site searches…including merchant websites”) is below a predetermined threshold (Par. [0031] discloses “a charity having the highest score for a user may then be recommended to the user” indicating a threshold to be met).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar in view of Lu of having a system performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images; analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph associated with an online platform to create query image feature data for each query image; aggregating the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images; identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images; and identifying, by a demand analysis module, a user demand for an item from the plurality of users that the online platform has not been met based on comparing the query theme to catalog data of the online platform, with the teachings of Kumar of having determining that a similarity between the query theme and the catalog data of the online platform is below a predetermined threshold.
Wherein having Ghadekar’s system of having determining that a similarity between the query theme and the catalog data of the online platform is below a predetermined threshold.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve user’s recommendation system that is tailored to the user’s interests, likes, purchase, and/or associations, which increase efficiency for a user interactive system. Since both Ghadekar and Kumar both perform target product recommendation based on cross-domain collaborative filtering. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Kumar’s system improves user’s recommendation system that is tailored to the user’s interests, likes, purchase, and/or associations, which increase efficiency for a user interactive system, (see Kumar’s Par. [0003]).
Regarding claim 17, Ghadekar in view of Lu and Kumar teaches the non-transitory computer-readable medium of claim 16, Ghadekar teaches wherein the plurality of query images comprises image metadata comprising user action data associated with aspects of the plurality of query images (Since the processing of Ghadekar is based on other users’ clicked images or input images to generate the associations rule for recommendation based on user-user similarity as disclosed in section VI; i the use of receiving multiple query images associated with other user query on the online platform; since the multiple query images are images of other users’ clicked images or purchasing images or input images, they indicates metadata associated with multiple query images).
Regarding claim 18, Ghadekar in view of Lu and Kumar teaches the non-transitory computer-readable medium of claim 16, Ghadekar teaches wherein the executable instructions, which when executed by a processing device, cause the processing device to perform further operations comprising: determining a query result by comparing one or more query images corresponding to a user query to catalog data of the online platform (The demand data is based on comparing the input image to the images of what other users purchased in the past of catalog data of an online platform, such as finding the similarity between user-user according to section VI, since determining of similarity indicates a comparison); and providing the query result to a user computing device associated with user query (The recommendation result is shown to the user at the website according to section V).
Regarding claim 20, Ghadekar in view of Lu and Kumar teaches the non-transitory computer-readable medium of claim 16, Ghadekar teaches wherein the executable instructions, which when executed by a processing device, cause the processing device to perform further operations comprising: transmitting data associated with the user demand to the online platform (The recommendation as discussed previously in claim 16 is being shown to the user using the online shopping website, it can be understood as the demand data being transmitted to the website to show to the user; section VI discloses the output of the processing is to provide the user with recommendation of items based on the input image of what other users have had purchased in the past to follow interest rule of what items the current user may be interested in, it can be understood as a demand data has been generated of what users would be interested in), wherein the data comprises a synthetic object image generated based on the query theme (The recommendation data includes images output by the system [generated from the processing has been mentioned] based on the transaction list [the query theme] as shown in FIG. 7 and section VI).
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Premanand Ghadekar et. al. (“Image-Based Product Recommendations Using Market Basket Analysis, Sept. 2019, 2019 5th International Conference on Computing, Communication, Control And Automation” hereinafter as “Ghadekar”) in view of Biao Lu et. al. (“US 2024/0265309 A1” hereinafter as “Lu”) further in view of Dinesh Kumar et. al. (“US 2023/0214696 A1” hereinafter as “Kumar”) and Tingting Qiao et. al. (“MirrorGAN: Learning Text-To-Image Generation by Redescription, 2019, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition CVPR, pp. 1505-1514” hereinafter as “Qiao”).
Regarding claim 2, Ghadekar in view of Lu and Kumar teaches the method of claim 1, Ghadekar teaches wherein at least one query image of the plurality of query images (Section IX, 1st Par., discloses “associations are generated based on the most frequent products bought by similar users” indicating different users perform the query [each query image of the plurality of query images as claimed]).
However, Ghadekar in view of Lu and Kumar does not explicitly teach at least one query image of the plurality of query images is a synthetic image generated by a generative artificial intelligence model.
Qiao teaches at least one query image of the plurality of query images is a synthetic image generated by a generative artificial intelligence model (FIG. 2 illustrates and its caption discloses that an image can be generated based on giving a description, the query image of Ghadakar can be understood to be an image given by description by the user, the image generated by a neural network to generate images [being generative]).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar in view of Lu and Kumar of having a method performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images, with the teachings of Qiao of having at least one query image of the plurality of query images is a synthetic image generated by a generative artificial intelligence model.
Wherein having Ghadekar’s method of having at least one query image of the plurality of query images is a synthetic image generated by a generative artificial intelligence model.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve target image generation by using a neural network architecture to enhance the diversity and semantic consistency of the generated images. Since both Ghadekar and Qiao both perform target image generation. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Qiao’s system improves target image generation by using a neural network architecture to enhance the diversity and semantic consistency of the generated images, (see Qiao’s Par. [0003]).
Claims 6, 13 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Premanand Ghadekar et. al. (“Image-Based Product Recommendations Using Market Basket Analysis, Sept. 2019, 2019 5th International Conference on Computing, Communication, Control And Automation” hereinafter as “Ghadekar”) in view of Biao Lu et. al. (“US 2024/0265309 A1” hereinafter as “Lu”) further in view of Dinesh Kumar et. al. (“US 2023/0214696 A1” hereinafter as “Kumar”) and Thomas Deselaers et. al. (“Visual and Semantic Similarity in ImageNet, June 2011, CVPR 2011, Colorado Springs, CO, USA” hereinafter as “Deselaers”).
Regarding claim 6, Ghadekar in view of Lu and Kumar teaches the method of claim 1, Ghadekar teaches wherein analyzing, by the image content analyzer module, the multiple image tags (Section VII, 2nd par., discloses identifying distinct products in the image by a CNN, wherein the identified information include contents of the product [analogous to multiple image tags as claimed, such as the labels given to each product in the image as shown in FIG. 1, when an image includes multiple products, it would include multiple labels such as an image of a pencil and an eraser together such as shown in section V, 1st par.], as part of the smart engine) based on the knowledge graph associated with the online platform to create the query image feature data for each query image comprises (Section V, 1st 3 pars., discloses when the image labels are identified, the association rule mining is used to state association between the products in the image with another product likely to be bought or be interested in by the user, the association rule here can be understood as a knowledge graph as claimed): mapping the multiple image tags to the knowledge graph (Mapping the labeled item with the association rule such as shown by arrow in section V of an example of the association between the data); the multiple image tags based on the knowledge graph (The image tags based on the knowledge graph; Section V, 1st 3 pars., discloses when the image).
However, Ghadekar in view of Lu and Kumar does not explicitly teach wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags, and determining a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags.
Deselaers teaches wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags (FIG. 6 illustrates and its caption discloses that the semantic distance between the categories of the neighbors are determined to show objects of the same basic-level category, wherein the labels here indicate semantic descriptions being inferred), and determining a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags (FIG. 6 illustrates and its caption discloses the semantic context being used to obtain the inferred semantic descriptions being the labels of the multiple image tags [the image labels]; further disclosed in section 3).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar in view of Lu and Kumar of having a method performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images; analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph associated with an online platform to create query image feature data for each query image; aggregating the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images; identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images; and identifying, by a demand analysis module, a user demand for an item from the plurality of users that the online platform has not been met based on comparing the query theme to catalog data of the online platform, with the teachings of Deselaers of having wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags, and determining a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags.
Wherein having Ghadekar’s method of having wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags, and determining a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve target image generation by using a neural network architecture to enhance the diversity and semantic consistency of the generated images. Since both Ghadekar and Deselaers both perform semantic similarity determination between images. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Deselaers’s system improves determination of image similarity based on visual distances more effectively, (see Deselaers’s Abstract).
Regarding claim 13, Ghadekar in view of Lu and Kumar teaches the system of claim 10, Ghadekar teaches wherein the image content analyzer module is further configured to: map the multiple image tags to the knowledge graph (Mapping the labeled item with the association rule such as shown by arrow in section V of an example of the association between the data); the multiple image tags based on the knowledge graph (The image tags based on the knowledge graph; Section V, 1st 3 pars., discloses when the image labels are identified).
However, Ghadekar in view of Lu and Kumar does not explicitly teach wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags; wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags, and determine a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags.
Deselaers teaches wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags (Section 4, 1st 2 Pars, disclose “semantic distances between categories defined on the WorldNet hierarchy relate to visual distances in ImageNet”); wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags (FIG. 6 illustrates and its caption discloses that the semantic distance between the categories of the neighbors are determined to show objects of the same basic-level category, wherein the labels here indicate semantic descriptions being inferred), and determine a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags (FIG. 6 illustrates and its caption discloses the semantic context being used to obtain the inferred semantic descriptions being the labels of the multiple image tags [the image labels]; further disclosed in section 3).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar in view of Lu and Kumar of having a system performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images; analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph associated with an online platform to create query image feature data for each query image; aggregating the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images; identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images; and identifying, by a demand analysis module, a user demand for an item from the plurality of users that the online platform has not been met based on comparing the query theme to catalog data of the online platform, with the teachings of Deselaers of having wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags; wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags, and determine a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags.
Wherein having Ghadekar’s system of having wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags; wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags, and determine a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve target image generation by using a neural network architecture to enhance the diversity and semantic consistency of the generated images. Since both Ghadekar and Deselaers both perform semantic similarity determination between images. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Deselaers’s system improves determination of image similarity based on visual distances more effectively, (see Deselaers’s Abstract).
Regarding claim 19, Ghadekar in view of Lu and Kumar teaches the non-transitory computer-readable medium of claim 16, Ghadekar teaches wherein the executable instructions, which when executed by a processing device, cause the processing device to perform further operations comprising: (FIG. 6 illustrates and its caption discloses that the semantic distance between the categories of the neighbors are determined to show objects of the same basic-level category, wherein the labels here indicate semantic descriptions being inferred): map the multiple image tags to the knowledge graph (Mapping the labeled item with the association rule such as shown by arrow in section V of an example of the association between the data); the multiple image tags based on the knowledge graph (The image tags based on the knowledge graph; Section V, 1st 3 pars., discloses when the image labels are identified).
However, Ghadekar in view of Lu and Kumar does not explicitly teach wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags, and determine a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags.
Deselaers teaches wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags (FIG. 6 illustrates and its caption discloses that the semantic distance between the categories of the neighbors are determined to show objects of the same basic-level category, wherein the labels here indicate semantic descriptions being inferred), and determine a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags (FIG. 6 illustrates and its caption discloses the semantic context being used to obtain the inferred semantic descriptions being the labels of the multiple image tags [the image labels]; further disclosed in section 3).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar in view of Lu and Kumar of having a system performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images, with the teachings of Deselaers of having wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags; wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags, and determine a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags.
Wherein having Ghadekar’s system of having wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags; wherein the query image feature data comprises inferred semantic descriptions of the multiple image tags, and determine a semantic context of the multiple image tags to obtain the inferred semantic descriptions of the multiple image tags.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve target image generation by using a neural network architecture to enhance the diversity and semantic consistency of the generated images. Since both Ghadekar and Deselaers both perform semantic similarity determination between images. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Deselaers’s system improves determination of image similarity based on visual distances more effectively, (see Deselaers’s Abstract).
Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over Premanand Ghadekar et. al. (“Image-Based Product Recommendations Using Market Basket Analysis, Sept. 2019, 2019 5th International Conference on Computing, Communication, Control And Automation” hereinafter as “Ghadekar”) in view of Biao Lu et. al. (“US 2024/0265309 A1” hereinafter as “Lu”) further in view of Dinesh Kumar et. al. (“US 2023/0214696 A1” hereinafter as “Kumar”) and Yucheng Jin et. al. (“CircleBuy: a visual search based second screen application of buying products in videos, June 2016, Proceedings of the 8th ACM SIGCHI Symposium on Engineering Interactive Computing Systems, pp. 287-292” hereinafter as “Jin”).
Regarding claim 9, Ghadekar in view of Lu and Kumar teaches the method of claim 1.
However, Ghadekar in view of Lu and Kumar does not explicitly teach further comprising transmitting to a user computing device a notification message indicating that the item has been newly added to the online platform.
Jin teaches further comprising transmitting to a to a user computing device a notification message indicating that the item has been newly added to the online platform (Page 289, 1st col., last par., discloses the similar product also is given descriptions [notification message] since, Ghadekar discloses the similar products are given to the user by transmitting it to the computing device, here, Jin further teaches the similar can further be given description [message] and the showing to the user the description is analogous to notification as claimed).
Therefore, it would have been obvious to one or ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teaches of Ghadekar in view of Lu and Kumar of having a method performed by one or more processing devices, comprising: receiving, by an image tagging module, a plurality of query images corresponding to a plurality of user queries from a plurality of users during a period of time; generating, by the image tagging module, multiple image tags for each query image of the plurality of query images; analyzing, by an image content analyzer module, the multiple image tags based on a knowledge graph associated with an online platform to create query image feature data for each query image; aggregating the query image feature data for each query image of the plurality of query images to obtain aggregated query image feature data for the plurality of query images; identifying, by a theme identification module, a query theme based on the aggregated query image feature data for the plurality of query images; and identifying, by a demand analysis module, a user demand for an item from the plurality of users that the online platform has not been met based on comparing the query theme to catalog data of the online platform, with the teachings of Jin of having further comprising transmitting to a user computing device a notification message indicating that the item has been newly added to the online platform.
Wherein having Ghadekar’s method of having further comprising transmitting to a user computing device a notification message indicating that the item has been newly added to the online platform.
The motivation behind the modification would have been to improve product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, and further to improve product recommendation to user based on trends in e-commerce. Since both Ghadekar and Jin both perform object recommendation for online shopping. Wherein Ghadekar’s system improves product recommendation system that considers not only the product history but also the linkage between the same products bought by other customers and by using a learned neural network to learn the linkage enables an efficient way of recommending products to the user, (see Ghadekar’s Abstract) and Jin’s system improves product recommendation to user based on trends in e-commerce, (see Jin’s Abstract).
Pertinent Prior Art(s)
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
LUO; JING et. al., “US 20090234842 A1”, discloses an image search method and system using face detection, begins with receiving a query submitted by a user. Next a query word is searched in the query from an image resource using an image search engine to obtain an initial image collection. Any faces are detected in each image in the initial image collection which has been searched. A search for the query word in a text surrounding each image having the face in the initial image collection is performed. A determination is made whether the query word indicates at least one person's name in the surrounding text matching the query word. An image an image in the initial image collection is returned to a user in which the face is included and the query word in the surrounding text indicates the person's name.
CHESTER; David et. al., “US 20170262479 A1”, discloses an image search engine to search against images having a similar concept. The aspects include mapping between concepts of the user drawing space and concepts of the image space such that images associated with the same concept are identified. For each input user drawing, the drawing is first processed through a concept classifier to identify a corresponding concept, and then through a feature extractor to form a corresponding feature vector. The results from the concept classifier and the feature extractor may be combined to search against a collection of images having a similar concept to determine a listing of images ranked by visual and semantic similarity to the input.
KALE; Ajinkya et. al., “US 20220114361 A1”, discloses an image caption generator model to generate phrase tags for input images. The phrase tags can include short phrases that describe the contents of the images (e.g., objects depicted therein). Once trained, the image caption generator model can be used as an image phrase tagger to tag input images from an image library with phrase tags. The image library can be indexed based on their phrase tags. Subsequently, when the image library is queried, the query can be divided into phrases and the index can be used to identify matching images.
BAO; Ling et. al., “US 20220319046 A1”, discloses a visual positioning method, the method may include obtaining a positioning image collected by an imaging device; obtaining a three-dimensional (3D) point cloud map associated with an area where the imaging device is located; determining a target area associated with the positioning image from the 3D point cloud map based on the positioning image.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PHUONG HAU CAI whose telephone number is (571)272-9424. The examiner can normally be reached M-F 8:30 am - 5:00pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Chineyere Wills-Burns can be reached at (571) 272-9752. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/PHUONG HAU CAI/Examiner, Art Unit 2673
/CHINEYERE WILLS-BURNS/Supervisory Patent Examiner, Art Unit 2673
