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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 submission filed on 12/02/2025 has been entered.
Accordingly, claims 1-9, 17, 21-22, and 24-31 are pending in this application. Claims 1, 7-9, and 17 are currently amended. New claim 31 is added. Claim 23 is cancelled.
Response to Arguments
Applicant’s arguments with respect to amended pending claims filed on 12/02/2025 have been fully considered. In view of the claim amendment filed, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made.
Further, regarding the new limitations recited in claims 1, 7-9, and 17, it is submitted that they are properly addressed by the new ground of rejection.
Furthermore, it is also submitted that all limitations in pending claims, including those not specifically argued, are properly addressed. The reason is set forth in the rejections. See claim analysis below for detail.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
Claims 1-9, 17, 21-22, and 24-31 are rejected under 35 U.S.C. 103 as being unpatentable over Chen et al. (US 20200311599 A1) in view of Puthenputhussery et al (US 20210240722 A1), Chen et al (US Patent No. 11514321 B1, hereinafter Chen), and Saha et al. (US 20230049418 A1).
Regarding Claim 1, Chen (US 20200311599 A1) discloses a system comprising: a processor; and a non-transitory memory storing instructions that, when executed, cause the processor to ([0037]: FIG. 5 is a simplified block diagram of the environment 10 according to another example. The environment 10 includes the computing device 12, the memory 16, and the processor device 14) :
receive a request to generate a training dataset for an attribute (Fig. 2; The computing device 12 receives the request 30 from the requestor 28-1 for the machine learning training dataset 22-1 that comprises the plurality of objects 24-1-24-3. The plurality of objects 24-1-24-3 includes data for training the MLM 29-1 (FIG. 2, block 1000)),
wherein the attribute (Fig. 3; [Abstract]: The plurality of objects includes data for training a machine learning mode; [0034]: In this example, the selection criteria 56 may include a parameter 56-1 that indicates that the object metadata… contains the uniqueness metric 38)
and an attribute threshold value are included in the request ([0034]: The selection criteria 56 may include a parameter 56-4 that indicates that all objects 24 that have a uniqueness metric 38 greater than X should be provided by the reverse proxy 18, where X is a percentage in a range of 0 and 100 (inclusive));
determine a similarity score for each item identifier stored in an item database based on one or more queries ([0033]:The computing device 12 determines a uniqueness characteristic for the plurality of objects 24-1-24-3, the uniqueness characteristic indicative of how unique each object 24-1-24-3 is relative to each other object 24-1-24-3 (FIG. 2, block 1002). The uniqueness characteristic may include, by way of non-limiting example, the uniqueness metric 38, [0002]: The examples disclosed herein automatically curate machine learning training datasets based on one or more uniqueness characteristics, such as a uniqueness metric that quantifies how unique each object in the training dataset is with respect to the other objects in the training dataset; [The similarity score corresponds to a uniqueness metric]),
select a set of item identifiers from the item database based on the similarity score ([0002]; In some examples, in response from a requestor for a machine learning training dataset, the examples disclosed herein may, based on the uniqueness metric, return only a subset of the most unique objects to optimize machine learning training),
wherein the similarity score indicates how related a corresponding item identifier is to the attribute ([0017]: The examples disclosed herein automatically curate machine learning training datasets based on one or more uniqueness characteristics, such as a uniqueness metric that quantifies how unique objects in the training dataset are with respect to the other objects in the training dataset; [0033]: the uniqueness characteristic indicative of how unique each object 24-1-24-3 is relative to each other object 24-1-24-3 (FIG. 2, block 1002));
generate the training dataset by: for each item identifier of the set of item identifiers: compare the similarity score to the attribute threshold value (Figs. 2-3; [0034]: The selection criteria 56 may include a parameter 56-4 that indicates that all objects 24 that have a uniqueness metric 38 greater than X should be provided by the reverse proxy 18, where X is a percentage in a range of 0 and 100 (inclusive)); and
in response to the similarity score being above the attribute threshold value, include the corresponding item identifier in the training dataset for the attribute (Figs. 2-3: [0033]: The computing device 12 sends, to the requestor 28-1, the group 52 of objects 24-1-24-3, the group 52 of objects 24-1-24-3 being selected based at least partially on the uniqueness characteristic);
store the training dataset for the attribute in a training dataset database (Fig. 1A; [0020]: The machine learning training dataset storage 20 contains one or more machine learning training datasets 22-1-22-N (generally, training datasets 22));
generate a machine learning model using the training dataset for the attribute (Fig. 5; ([0021]: Thus, objects 24 contain training data used to generate, or update, a machine learning model. Different types of training datasets 22 may be used to train different machine learning models; [0023]: At a time T1, the requestor 28-1 sends a request 30 toward the machine learning training dataset storage 20 that requests the training dataset 22-1 for training the MLM 29-1… [0037]: The plurality of objects 24-1-24-3 comprises data for training the machine learning model 29-1); and
However Chen does not explicitly teach “each respective query in the one or more queries is (i) associated, at least in part, with one or more corresponding interactions with a respective item displayed at a corresponding computing device and (ii) received, at least in part, as one or more words defining the attribute and inputted at a graphical user interface of the corresponding computing device, and store the machine learning model in a model database”.
On the other hand, in the same field of endeavor, Puthenputhussery teaches
each respective query in the one or more queries is (i) associated, at least in part, with one or more corresponding interactions with a respective item displayed at a corresponding computing device ([0005]-[0007]: a computing device is configured to receive item data identifying a plurality of items for a search query provided by a user… the user engagement data identifies engagements by the user on a webpage) and
(ii) received, at least in part, as one or more words defining the attribute and inputted at a graphical user interface of the corresponding computing device (Fig. 2; [0035]: For example, search result determination computing device 102 may present (via user interface 205, as described below with respect to FIG. 2) a search query; Fig. 3; [0080]-[0083]: In this example, user session data 320 includes item engagement data 360 and search query data 330… In some examples, search result determination computing device 102 may receive a search request 310 identifying and characterizing a search query for a user. The search query may include data identifying and characterizing one or more words, for example).
Additionally, Chen (US Patent No. 11514321 B1) teaches
store the machine learning model in a model database (Fig. 8; [Col. 5, lines 2-4]: A trained version of the second machine learning model may be stored in various embodiments; [Col. 19, lines 16-19]: provider network 801 may comprise resources used to implement a plurality of services, including for example… a database/storage service 823).
Furthermore, Saha teaches generate a trained machine learning model using the training data set for the attribute applied to a machine learning model ([Abstract]: Datasets may then be selected based on the set of input features, which may be used to generate training data for training the machine learning model); and store the trained machine learning model in a model database and use the trained machine learning model to perform a process ([0011]: FIG. 6 shows a model database storing data related to various machine learning models; [0013]: FIG. 8 shows a training subsystem for generating training data for training a machine learning model, in accordance with one or more embodiments) comprising:
select a subsequent set of item identifiers from the item database based on a subsequent similarity score generated by the trained machine learning model, wherein the subsequent similarity score comprises how related an item identifier in the subsequent set of item identifiers is to the attribute (Fig. 3; [0037]-[0041]: In some embodiments, a subset of the datasets previously obtained, or additional datasets from dataset database 132 may be selected… the features may be grouped based on similarity to one another);
receive an updated attribute threshold value ([0021]-[0046]: each connection (or the neural unit itself) may have a threshold function such that the signal must surpass the threshold before it propagates to other neural units);
for each item identifier of the subsequent set of item identifiers: compare the similarity score to the updated attribute threshold value, and in response to the similarity score being above the attribute threshold value, assign the corresponding item identifier to an updated training dataset for the attribute: and store the updated training dataset for the attribute in the training dataset database (Figs. 3-5; [0042]-[0046]: In some embodiments, clustering data 408 may include the correlation scores associated with each of the feature clusters…. features included within feature cluster A are determined to have no correlation, or less than a threshold amount of correlation (e.g., correlation score being less than a threshold correlation score), with respect to features included within feature cluster B).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Chen to incorporate the teachings of Puthenputhussery and Chen and Saha to determine a similarity score for each item identifier based on queries associated with corresponding interactions with a respective item and received as a text object defining the attribute, and to store the machine learning model in a model database.
The motivation for doing so would be to determine a value based on a relevance of each item to the search query, as recognized by Puthenputhussery ([0005] of Puthenputhussery: For example, in some embodiments, a computing device is configured to… determine, for each of the plurality of items, a first value based on a relevance of each of the plurality of items to the search query), to analyze various clusters, as recognized by Chen ([Col. 5, lines 7-11] of Chen: The trained versions of such models may then be used to analyze various clusters to determine whether any action (such as automated partitioning or auditing for possible partitioning) should be performed on the clusters) and to select dataset based on the input features, as recognized by Saha ([Abstract] of Saha: Datasets may then be selected based on the set of input features, which may be used to generate training data for training the machine learning model).
Regarding Claim 2, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Chen (US 20200311599 A1) further teaches wherein the instructions cause the processor to: identify the set of item identifiers from the item database based on the similarity score by selecting item identifiers from the item database including a corresponding order frequency above an order threshold (Fig. 3; [0034]: The selection criteria 56 may include a parameter 56 - 4 that indicates that all objects 24 that have a uniqueness metric 38 greater than X should be provided by the reverse proxy 18 , where X is a percentage in a range of 0 and 100 (inclusive)).
Regarding Claim 3, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1, wherein the instructions further cause the processor to:
Chen (US Patent No. 11514321 B1) further teaches identify the set of item identifiers from the item database based on the similarity score by: determining an engagement value based on at least one of a number of orders, a number of add-to-cart selections, and a number of view selections; and selecting a predetermined number of item identifiers corresponding to highest engagement values (Fig. 1; [Col. 10, lines 29-46]: Clusters identified using subsystems 120 and 170 may be used to respond to specific clustering requests… In a store problem domain, for example, the downstream applications 134 may comprise… customer views of similar items; [Col. 4, lines 25-27]: The entity records of individual ones of all of the clusters may each satisfy an intra-cluster entity-level similarity criterion; Fig. 7; [Col. 18, lines 11-34]: Any of a number of techniques may be employed to select clusters from which training data for the attribute-level similarity prediction model(s) in various embodiments).
Regarding Claim 4, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Chen (US Patent No. 11514321 B1) further teaches wherein the instructions further cause the processor to: identify the set of item identifiers from the item database based on the similarity score by: determining an engagement value based on at least one of a number of orders, a number of add-to-cart selections, and a number of view selections; and selecting the set of item identifiers as item identifiers with a corresponding engagement value above a first engagement threshold ([Col. 4, lines 25-27]: The entity records of individual ones of all of the clusters may each satisfy an intra-cluster entity-level similarity criterion; Fig. 7; [Col. 18, lines 11-34]: Any of a number of techniques may be employed to select clusters from which training data for the attribute-level similarity prediction model(s) in various embodiments).
Regarding Claim 5, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Chen (US 20200311599 A1) teaches wherein comparing the similarity score to the attribute threshold value includes identifying a subset of queries of a query list including a number of engagements between the corresponding item identifier and a user being above a second engagement threshold Figs. 2-3; [0034]: The selection criteria 56 may include a parameter 56-4 that indicates that all objects 24 that have a uniqueness metric 38 greater than X should be provided by the reverse proxy 18, where X is a percentage in a range of 0 and 100 (inclusive)).
Regarding Claim 6, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Chen (US 20200311599 A1) further teaches wherein the attribute includes at least one of: (i) a gender, (ii) an age, and (iii) a color ([0022]: The segmentation of the objects into blocks may be based on any criterion or criteria, such as a predetermined number of bytes, and/or based on the content of the data in the object; [Non-functional descriptive material describing the data]).
Regarding Claim 7, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Chen (US 20200311599 A1) further teaches wherein the instructions further cause the processor to: classify item identifiers as including the attribute using the trained machine learning model ([0003]: The method further includes sending, to the first requestor, a first group of objects from the plurality of objects, the first group of objects being selected based at least partially on the uniqueness characteristic or sent in an order based at least partially on the uniqueness characteristic).
Regarding Claim 8, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 7.
Chen (US 11514321 B1) further teaches wherein the instructions further cause the processor to, in response to receiving a new item identifier:
determine at least one attribute of the new item identifier by applying a plurality of machine learning models, comprising the trained machine learning model, stored in the model database to the new item identifier ([0054]: FIG. 6 illustrates aspects of an example machine learning model which may be use to predict attribute-level similarity; [0054]-[0057]: The attribute matching model 690 (which may also be referred to as an attribute-level similarity prediction model) depicted in FIG. 6 comprises several neural network-based stages, and may be employed for generating similarity scores for a given critical attribute); and
identify and tag the new item identifier based on the at least one attribute ([0055]: the most frequently represented byte pairs (e.g., with each byte corresponding to a character) in the tokens may be identified using an iterative technique).
Regarding Claim 9, Chen (US 20200311599 A1) discloses a method comprising:
receiving a request to generate a training dataset for an attribute (Fig. 2; The computing device 12 receives the request 30 from the requestor 28-1 for the machine learning training dataset 22-1 that comprises the plurality of objects 24-1-24-3. The plurality of objects 24-1-24-3 includes data for training the MLM 29-1 (FIG. 2, block 1000)),
wherein the attribute (Fig. 3; [Abstract]: The plurality of objects includes data for training a machine learning mode; [0034]: In this example, the selection criteria 56 may include a parameter 56-1 that indicates that the object metadata… contains the uniqueness metric 38)
and an attribute threshold value are included in the request ([0034]: The selection criteria 56 may include a parameter 56-4 that indicates that all objects 24 that have a uniqueness metric 38 greater than X should be provided by the reverse proxy 18, where X is a percentage in a range of 0 and 100 (inclusive));
determining a similarity score for each item identifier stored in an item database based on one or more queries ([0033]:The computing device 12 determines a uniqueness characteristic for the plurality of objects 24-1-24-3, the uniqueness characteristic indicative of how unique each object 24-1-24-3 is relative to each other object 24-1-24-3 (FIG. 2, block 1002). The uniqueness characteristic may include, by way of non-limiting example, the uniqueness metric 38, [0002]: The examples disclosed herein automatically curate machine learning training datasets based on one or more uniqueness characteristics, such as a uniqueness metric that quantifies how unique each object in the training dataset is with respect to the other objects in the training dataset; [The similarity score corresponds to a uniqueness metric]),
selecting a set of item identifiers from the item database based on the similarity score ([0002]; In some examples, in response from a requestor for a machine learning training dataset, the examples disclosed herein may, based on the uniqueness metric, return only a subset of the most unique objects to optimize machine learning training),
wherein the similarity score indicates how related a corresponding item identifier is to the attribute ([0017]: The examples disclosed herein automatically curate machine learning training datasets based on one or more uniqueness characteristics, such as a uniqueness metric that quantifies how unique objects in the training dataset are with respect to the other objects in the training dataset; [0033]: the uniqueness characteristic indicative of how unique each object 24-1-24-3 is relative to each other object 24-1-24-3 (FIG. 2, block 1002));
generating the training dataset by: for each item identifier of the set of item identifiers: compare the similarity score to the attribute threshold value (Figs. 2-3; [0034]: The selection criteria 56 may include a parameter 56-4 that indicates that all objects 24 that have a uniqueness metric 38 greater than X should be provided by the reverse proxy 18, where X is a percentage in a range of 0 and 100 (inclusive)); and
in response to the similarity score being above the attribute threshold value, including the corresponding item identifier in the training dataset for the attribute (Figs. 2-3: [0033]: The computing device 12 sends, to the requestor 28-1, the group 52 of objects 24-1-24-3, the group 52 of objects 24-1-24-3 being selected based at least partially on the uniqueness characteristic);
storing the training dataset for the attribute in a training dataset database (Fig. 1A; [0020]: The machine learning training dataset storage 20 contains one or more machine learning training datasets 22-1-22-N (generally, training datasets 22));
generating a machine learning model using the training dataset for the attribute (Fig. 5; ([0021]: Thus, objects 24 contain training data used to generate, or update, a machine learning model. Different types of training datasets 22 may be used to train different machine learning models; [0023]: At a time T1, the requestor 28-1 sends a request 30 toward the machine learning training dataset storage 20 that requests the training dataset 22-1 for training the MLM 29-1… [0037]: The plurality of objects 24-1-24-3 comprises data for training the machine learning model 29-1); and
However Chen does not explicitly teach “each respective query in the one or more queries is (i) associated, at least in part, with one or more corresponding interactions with a respective item displayed at a corresponding computing device and (ii) received, at least in part, as a text object defining the attribute and inputted at the corresponding computing device, and storing the machine learning model in a model database”.
On the other hand, in the same field of endeavor, Puthenputhussery teaches
each respective query in the one or more queries is (i) associated, at least in part, with one or more corresponding interactions with a respective item displayed at a corresponding computing device ([0005]-[0007]: a computing device is configured to receive item data identifying a plurality of items for a search query provided by a user… the user engagement data identifies engagements by the user on a webpage) and
(ii) received, at least in part, as one or more words defining the attribute and inputted at a graphical user interface of the corresponding computing device (Fig. 2; [0035]: For example, search result determination computing device 102 may present (via user interface 205, as described below with respect to FIG. 2) a search query; Fig. 3; [0080]-[0083]: In this example, user session data 320 includes item engagement data 360 and search query data 330… In some examples, search result determination computing device 102 may receive a search request 310 identifying and characterizing a search query for a user. The search query may include data identifying and characterizing one or more words, for example).
Additionally, Chen (US Patent No. 11514321 B1) teaches
storing the machine learning model in a model database (Fig. 8; [Col. 5, lines 2-4]: A trained version of the second machine learning model may be stored in various embodiments; [Col. 19, lines 16-19]: provider network 801 may comprise resources used to implement a plurality of services, including for example… a database/storage service 823).
Furthermore, Saha teaches generating a trained machine learning model using the training dataset for the attribute applied to a machine learning model ([Abstract]: Datasets may then be selected based on the set of input features, which may be used to generate training data for training the machine learning model) and storing the trained machine learning model in a model database; and using the trained machine learning model to perform a process ([0011]: FIG. 6 shows a model database storing data related to various machine learning models; [0013]: FIG. 8 shows a training subsystem for generating training data for training a machine learning model, in accordance with one or more embodiments) comprising:
selecting a subsequent set of item identifiers from the item database based on a subsequent similarity score generated by the trained machine learning model, wherein the subsequent similarity score comprises how related an item identifier in the subsequent set of item identifiers is to the attribute (Fig. 3; [0037]-[0041]: In some embodiments, a subset of the datasets previously obtained, or additional datasets from dataset database 132 may be selected… the features may be grouped based on similarity to one another);
receiving an updated attribute threshold value ([0021]-[0046]: each connection (or the neural unit itself) may have a threshold function such that the signal must surpass the threshold before it propagates to other neural units);
for each item identifier of the subsequent set of item identifiers: comparing the similarity score to the updated attribute threshold value; and in response to the similarity score being above the attribute threshold value, assigning the corresponding item identifier to an updated training dataset for the attribute; and storing the updated training dataset for the attribute in the training dataset database (Figs. 3-5; [0042]-[0046]: In some embodiments, clustering data 408 may include the correlation scores associated with each of the feature clusters…. features included within feature cluster A are determined to have no correlation, or less than a threshold amount of correlation (e.g., correlation score being less than a threshold correlation score), with respect to features included within feature cluster B).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teaching of Chen to incorporate the teachings of Puthenputhussery and Chen and Saha to determine a similarity score for each item identifier based on queries associated with corresponding interactions with a respective item and received as a text object defining the attribute, and to store the machine learning model in a model database.
The motivation for doing so would be to determine a value based on a relevance of each item to the search query, as recognized by Puthenputhussery ([0005] of Puthenputhussery: For example, in some embodiments, a computing device is configured to… determine, for each of the plurality of items, a first value based on a relevance of each of the plurality of items to the search query), to analyze various clusters, as recognized by Chen ([Col. 5, lines 7-11] of Chen: The trained versions of such models may then be used to analyze various clusters to determine whether any action (such as automated partitioning or auditing for possible partitioning) should be performed on the clusters) and to select dataset based on the input features, as recognized by Saha ([Abstract] of Saha: Datasets may then be selected based on the set of input features, which may be used to generate training data for training the machine learning model).
Regarding Claim 17, Chen (US 20200311599 A1) discloses a non-transitory computer readable medium having instructions stored thereon ([0037]: FIG. 5 is a simplified block diagram of the environment 10 according to another example. The environment 10 includes the computing device 12, the memory 16, and the processor device 14), wherein the instructions, when executed by at least one processor, cause a device to perform operations comprising:
receiving a request to generate a training dataset for an attribute (Fig. 2; The computing device 12 receives the request 30 from the requestor 28-1 for the machine learning training dataset 22-1 that comprises the plurality of objects 24-1-24-3. The plurality of objects 24-1-24-3 includes data for training the MLM 29-1 (FIG. 2, block 1000)),
wherein the attribute (Fig. 3; [Abstract]: The plurality of objects includes data for training a machine learning mode; [0034]: In this example, the selection criteria 56 may include a parameter 56-1 that indicates that the object metadata… contains the uniqueness metric 38)
and an attribute threshold value are included in the request ([0034]: The selection criteria 56 may include a parameter 56-4 that indicates that all objects 24 that have a uniqueness metric 38 greater than X should be provided by the reverse proxy 18, where X is a percentage in a range of 0 and 100 (inclusive));
determining a similarity score for each item identifier stored in an item database based on one or more queries ([0033]:The computing device 12 determines a uniqueness characteristic for the plurality of objects 24-1-24-3, the uniqueness characteristic indicative of how unique each object 24-1-24-3 is relative to each other object 24-1-24-3 (FIG. 2, block 1002). The uniqueness characteristic may include, by way of non-limiting example, the uniqueness metric 38, [0002]: The examples disclosed herein automatically curate machine learning training datasets based on one or more uniqueness characteristics, such as a uniqueness metric that quantifies how unique each object in the training dataset is with respect to the other objects in the training dataset; [The similarity score corresponds to a uniqueness metric]),
selecting a set of item identifiers from the item database based on the similarity score ([0002]; In some examples, in response from a requestor for a machine learning training dataset, the examples disclosed herein may, based on the uniqueness metric, return only a subset of the most unique objects to optimize machine learning training),
wherein the similarity score indicates how related a corresponding item identifier is to the attribute ([0017]: The examples disclosed herein automatically curate machine learning training datasets based on one or more uniqueness characteristics, such as a uniqueness metric that quantifies how unique objects in the training dataset are with respect to the other objects in the training dataset; [0033]: the uniqueness characteristic indicative of how unique each object 24-1-24-3 is relative to each other object 24-1-24-3 (FIG. 2, block 1002));
generating the training dataset by: for each item identifier of the set of item identifiers: compare the similarity score to the attribute threshold value (Figs. 2-3; [0034]: The selection criteria 56 may include a parameter 56-4 that indicates that all objects 24 that have a uniqueness metric 38 greater than X should be provided by the reverse proxy 18, where X is a percentage in a range of 0 and 100 (inclusive)); and
in response to the similarity score being above the attribute threshold value, including the corresponding item identifier in the training dataset for the attribute (Figs. 2-3: [0033]: The computing device 12 sends, to the requestor 28-1, the group 52 of objects 24-1-24-3, the group 52 of objects 24-1-24-3 being selected based at least partially on the uniqueness characteristic);
storing the training dataset for the attribute in a training dataset database (Fig. 1A; [0020]: The machine learning training dataset storage 20 contains one or more machine learning training datasets 22-1-22-N (generally, training datasets 22));
generating a trained machine learning model using the training dataset for the attribute applied to a machine learning model (Fig. 5; ([0021]: Thus, objects 24 contain training data used to generate, or update, a machine learning model. Different types of training datasets 22 may be used to train different machine learning models; [0023]: At a time T1, the requestor 28-1 sends a request 30 toward the machine learning training dataset storage 20 that requests the training dataset 22-1 for training the MLM 29-1… [0037]: The plurality of objects 24-1-24-3 comprises data for training the machine learning model 29-1); and
However Chen does not explicitly teach “each respective query in the one or more queries is (i) associated, at least in part, with one or more corresponding interactions with a respective item displayed at a corresponding computing device and (ii) received, at least in part, as a text object defining the attribute and inputted at the corresponding computing device, and storing the machine learning model in a model database”.
On the other hand, in the same field of endeavor, Puthenputhussery teaches
each respective query in the one or more queries is (i) associated, at least in part, with one or more corresponding interactions with a respective item displayed at a corresponding computing device ([0005]-[0007]: a computing device is configured to receive item data identifying a plurality of items for a search query provided by a user… the user engagement data identifies engagements by the user on a webpage) and
(ii) received, at least in part, as one or more words defining the attribute and inputted at a graphical user interface of the corresponding computing device (Fig. 2; [0035]: For example, search result determination computing device 102 may present (via user interface 205, as described below with respect to FIG. 2) a search query; Fig. 3; [0080]-[0083]: In this example, user session data 320 includes item engagement data 360 and search query data 330… In some examples, search result determination computing device 102 may receive a search request 310 identifying and characterizing a search query for a user. The search query may include data identifying and characterizing one or more words, for example).
Additionally, Chen (US Patent No. 11514321 B1) teaches
storing the trained machine learning model in a model database (Fig. 8; [Col. 5, lines 2-4]: A trained version of the second machine learning model may be stored in various embodiments; [Col. 19, lines 16-19]: provider network 801 may comprise resources used to implement a plurality of services, including for example… a database/storage service 823).
Furthermore, Saha teaches generating a trained machine learning model using the training dataset for the attribute applied to a machine learning model ([Abstract]: Datasets may then be selected based on the set of input features, which may be used to generate training data for training the machine learning model) and storing the trained machine learning model in a model database; and using the trained machine learning model to perform a process ([0011]: FIG. 6 shows a model database storing data related to various machine learning models; [0013]: FIG. 8 shows a training subsystem for generating training data for training a machine learning model, in accordance with one or more embodiments) comprising:
selecting a subsequent set of item identifiers from the item database based on a subsequent similarity score generated by the trained machine learning model, wherein the subsequent similarity score comprises how related an item identifier in the subsequent set of item identifiers is to the attribute (Fig. 3; [0037]-[0041]: In some embodiments, a subset of the datasets previously obtained, or additional datasets from dataset database 132 may be selected… the features may be grouped based on similarity to one another);
receiving an updated attribute threshold value ([0021]-[0046]: each connection (or the neural unit itself) may have a threshold function such that the signal must surpass the threshold before it propagates to other neural units);
for each item identifier of the subsequent set of item identifiers: comparing the similarity score to the updated attribute threshold value; and in response to the similarity score being above the attribute threshold value, assigning the corresponding item identifier to an updated training dataset for the attribute; and storing the updated training dataset for the attribute in the training dataset database (Figs. 3-5; [0042]-[0046]: In some embodiments, clustering data 408 may include the correlation scores associated with each of the feature clusters…. features included within feature cluster A are determined to have no correlation, or less than a threshold amount of correlation (e.g., correlation score being less than a threshold correlation score), with respect to features included within feature cluster B).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teaching of Chen to incorporate the teachings of Puthenputhussery and Chen and Saha to determine a similarity score for each item identifier based on queries associated with corresponding interactions with a respective item and received as a text object defining the attribute, and to store the machine learning model in a model database.
The motivation for doing so would be to determine a value based on a relevance of each item to the search query, as recognized by Puthenputhussery ([0005] of Puthenputhussery: For example, in some embodiments, a computing device is configured to… determine, for each of the plurality of items, a first value based on a relevance of each of the plurality of items to the search query), to analyze various clusters, as recognized by Chen ([Col. 5, lines 7-11] of Chen: The trained versions of such models may then be used to analyze various clusters to determine whether any action (such as automated partitioning or auditing for possible partitioning) should be performed on the clusters) and to select dataset based on the input features, as recognized by Saha ([Abstract] of Saha: Datasets may then be selected based on the set of input features, which may be used to generate training data for training the machine learning model).
Regarding Claim 21, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Puthenputhussery further teaches wherein the determining the similarity score is based, at least in part, on a first weight associated with a first type of interaction in the one or more corresponding interactions (Fig. 8; [0038]-[0044]: The chart further includes a rank weighting column 812, which identifies a weighting for original rank 810) and a second weight, different from the first weight, associated with a second type of interaction in the one or more corresponding interactions, different from the first type of interaction (Fig. 8; ([0044]: The chart further includes… a relevance probability column 814, which represents a relevance probability value (e.g., generated by a trained crowd model) for each of the items 802).
Regarding Claim 22, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Puthenputhussery further teaches wherein the determining the similarity score comprises assigning a first value to each respective interaction in the one or more corresponding interactions that satisfies a threshold value ([0005]: determine, for each of the plurality of items, a first value based on a relevance of each of the plurality of items to the search query) and assigning a second value, less than the first value, that fails to satisfy the threshold value ([0005]: The computing device may further be configured to determine, for each of the plurality of items, a second value based on the user engagement data; [0048]: In some examples, engagement data for a particular category (e.g., clicks, add-to-cart events, orders, etc.) must meet a threshold before the wt value for that particular category is used).
Regarding Claim 24, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Puthenputhussery further teaches wherein the attribute is based, at least in part, on a text description of a first length and a text description of a second length greater than the first length and defined, at least in part, by a third-party (Fig. 1; [0082]: Database 116 may further store catalog data 310, which may identify one or more attributes of a plurality of items… item description 378 (e.g., a description of the product including product features, such as ingredients, benefits, use or consumption instructions, or any other suitable description), and item options 380 (e.g., item colors, sizes, flavors, etc.)).
Regarding Claim 25, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Chen (US Patent No. 11514321 B1) further teaches wherein the set of items is selected based, at least in part, on a predetermined threshold number of items associated with the set of items ([Col. 4, lines 40-43]: clusters smaller than a threshold size (e.g., 100 entity records) may be selected as candidates for providing training data).
Regarding Claim 26, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Puthenputhussery further teaches wherein the similarity score is bound between a minimum respective similar score and a maximum respective similarity score (Fig. 1; [0051]: In one example, search result determination computing device 102 employs a “stacked ranker” configuration where the initial set of items are divided into “tiers” based on their corresponding relevance probability values (as generated by the crowd model). Each tier may correspond to a range of possible relevance probability values).
Regarding Claim 27, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Puthenputhussery further teaches wherein the one or more corresponding interactions comprises a viewing of a respective item through a graphical user interface, a selection of the respective item through the graphical user interface, a value of the respective item, or a combination thereof (Fig. 1; [0026]: For example, each of multiple computing devices 110, 112, 114 may be operable to view, access, and interact with a website hosted by web server 104… The website may allow the operator to add one or more of the items to an online shopping cart, and allow the customer to perform a “checkout” of the shopping cart to purchase the items).
Regarding Claim 28, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Puthenputhussery further teaches wherein, prior to the determining the similarity score comprises, the instructions further cause the processor to identify each item stored in the item database satisfying a threshold value of queries (Figs. 5, 9A-9B; [0100]-[0103]: Item re-ranking determination engine 504 may re-rank the ranked item recommendations 503 based on relevance probability data 407 and engagement probability data 505… engagement data (e.g., as identified by user session data 320) for the first user satisfies one or more thresholds as described above, but engagement data for the second user does not satisfy one or more of the thresholds).
Regarding Claim 29, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Puthenputhussery further teaches wherein the attribute includes (i) a gender, (ii) an age, (iii) a color, or (iv) a size ([0082]: Catalog data 310 may identify, for each of the plurality of items… item options 380 (e.g., item colors, sizes, flavors, etc.).
Regarding Claim 30, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Puthenputhussery further teaches wherein the training dataset comprises a corresponding item definition for each respective item in the set of item identifiers (Fig. 3; [0091]-[0092]: Specifically, FIG. 4 illustrates the training of a crowd model (e.g., as identified by crowd model data 390)… For example, crowd data 391 includes, for each of a plurality of items, corresponding user transaction data 340 (e.g., order numbers 342 that include an item ID 343 for each item).
Regarding Claim 31, the combined teachings of Chen, Puthenputhussery, and Chen disclose the system of claim 1.
Chen further teaches wherein the updated training dataset is accessible to a plurality of untrained models (Fig. 2; [0021]: Thus, objects 24 contain training data used to generate, or update, a machine learning model. Different types of training datasets 22 may be used to train different machine learning models).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHIRLEY D. HICKS whose telephone number is (571)272-3304. The examiner can normally be reached Mon - Fri 7:30 - 4:00.
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, Charles Rones can be reached on (571) 272-4085. 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.
/S.D.H./Examiner, Art Unit 2168
/CHARLES RONES/Supervisory Patent Examiner, Art Unit 2168