DETAILED ACTION
Status of Claims
The action is in reply to the Application 19/307,635 filed on 08/22/2025.
Claims 1-4, 6-11, 13-18, and 20 are currently pending and have been examined.
Claims 5, 12, and 19 are allowed.
The action is made NON-FINAL.
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 .
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-4, 6-7, 8-11, 13-18, and 20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claims 1-4, 6-7, 8-11, 13-18, and 20 are directed to one of the four statutory categories (process, machine, article of manufacture, or composition of matter) since the claimed invention falls into “a process” (a method for determining similarity between goods and services), “a machine” (a system for determining similarity between goods and services), and “an article of manufacture” (a computer-readable medium for determining similarity between goods and services) categories.
Regarding Claims 1-4, 6-7, 8-11, 13-18, and 20, the claim invention is directed to a judicial exception to patentability, an abstract idea.
Claim 1 recites the following limitations:
A method for determining a similarity between goods and services, comprising:
receiving a query including a first goods and services description and a second goods and services description;
providing the first goods and services description and the second goods and services description to a first … model;
receiving, from the first … model, a first set of goods and services classifications for the first goods and services description and a second set of goods and services classifications for the second goods and services description;
determining a plurality of goods and services similarity scores, each goods and services similarity score of the determined plurality of goods and services similarity scores indicating a similarity between a first goods and services classification from the first set of goods and services classifications and a second goods and services classification from the second set of goods and services classifications;
determining an aggregate goods and services similarity score based on the determined plurality of goods and services similarity scores; and
providing the aggregate goods and services similarity score as a query result.
Step 2A, Prong 1: The limitations for Claim 1 described above fall within “Certain Methods of Organizing Human Activity” for managing personal interactions between people such as following rules and instructions. Accordingly, this claim recites an abstract idea.
Step 2A, Prong 2: This judicial exception is not integrated into a practical application. Claim 1 recites one additional element – “machine learning”. This additional element represents mere generally linking of the use of the judicial exception (the abstract idea) to a particular technological environment or field of use (See MPEP 2106.05(h)). Accordingly, alone and in combination, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. This claim is directed to an abstract idea.
Step 2B: Claim 1 does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional element amounts to no more than representing mere generally linking of the use of the abstract idea to a particular technological environment or field of use (See MPEP 2106.05(h)). Representing mere generally linking the use of the abstract idea to a particular technological environment or field of use cannot provide an inventive concept. As a result, this claim is not patent eligible.
Claims 2 and 7 are directed to substantially the same abstract idea as Claim 1 and are rejected for substantially the same reasons. The additional recited limitations of the dependent claims fail to establish that the claims do not recite an abstract idea because the additional recited limitations of the claims further narrow the abstract idea. These dependent claims further narrow the abstract idea of Claim 1 such as by defining “wherein said determining the plurality of goods and services similarity scores comprises determining a goods and services similarity score based on a proportion of historical trademark cases where a similarity assessment has determined that the first goods and services classification related to a first trademark is similar to the second goods and services classification related to a second trademark” in Claim 2 and by defining “wherein the first set of goods and services classifications or the second set of goods and services classifications include Nice Classification (NCL) classifications and/or sub-classifications” in Claim 7.
Step 2A, Prong 2: These dependent claims do not integrate the abstract idea into practical application because they do not recite additional elements.
Step 2B: These dependent claims do not amount to significantly more than the abstract idea because they do not recite additional elements. Therefore, these claims are not patent eligible.
Claims 3-4 and 6 are directed to substantially the same abstract idea as Claim 1 and are rejected for substantially the same reasons. The additional recited limitations of the dependent claims fail to establish that the claims do not recite an abstract idea because the additional recited limitations of the claims further narrow the abstract idea. These dependent claims further narrow the abstract idea of Claim 1 such as by defining “wherein said determining the plurality of goods and services similarity scores comprises: determining a goods and services similarity score using … using data from historical trademark cases where a similarity assessment has been performed” in Claim 3, by defining “determining a first set of text fragments that are semantically similar to the first goods and services description and a second set of text fragments that are semantically similar to the second goods and services description; and providing the first set of text fragments and the second set of text fragments to the first … model; wherein the first … model: determines the first set of goods and services classifications based on the first goods and services description and the first set of text fragments, and determines the second set of goods and services classifications based on the second goods and services description and the second set of text fragments” in Claim 4, and by defining “determining a main goods and services classification associated with the first goods and services description, wherein the first set of goods and services classifications provided by the first … model include sub-classifications of the main goods and services classification” in Claim 6.
Step 2A, Prong 2: Claims 3-4 and 6 do not integrate the abstract idea into practical application. Claim 3 recites an additional element – “a machine learning prediction model that is trained”, and Claim 4 and 6 recite an additional element – “machine learning”. These additional elements amount to no more than representing mere generally linking the use of the judicial exception (the abstract idea) to a particular technological environment or field of use (See MPEP 2106.05(h)). Accordingly, alone and in combination, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
Step 2B: Claims 3-4 and 6 do not amount to significantly more than the abstract idea. Claims 3-4 and 6 do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, these additional elements amount to no more than representing mere generally linking the use of the abstract idea to a particular technological environment or field of use (See MPEP 2106.05(h)). Representing mere generally linking the use of the abstract idea to a particular technological environment or field of use cannot provide an inventive concept. As a result, these claims are not patent eligible.
Claim 8 recites the following limitations:
A system for determining a similarity between goods and services, comprising:
… to:
receive a query including a first goods and services description and a second goods and services description;
provide the first goods and services description and the second goods and services description to a first … model;
receive, from the first … model, a first set of goods and services classifications for the first goods and services description and a second set of goods and services classifications for the second goods and services description;
determine a plurality of goods and services similarity scores, each goods and services similarity score of the determined plurality of goods and services similarity scores indicating a similarity between a first goods and services classification from the first set of goods and services classifications and a second goods and services classification from the second set of goods and services classifications;
determine an aggregate goods and services similarity score based on the determined plurality of goods and services similarity scores; and
provide the aggregate goods and services similarity score as a query result.
Step 2A, Prong 1: The limitations for Claim 8 described above fall within “Certain Methods of Organizing Human Activity” for managing personal interactions between people such as following rules and instructions. Accordingly, this claim recites an abstract idea.
Step 2A, Prong 2: This judicial exception is not integrated into a practical application. Claim 8 recites additional elements – “a processor; and a computer-readable medium comprising computer-executable instructions, that when executed by the processor, causes the processor” and “machine learning”. The additional element “machine learning” represents mere generally linking of the use of the judicial exception (the abstract idea) to a particular technological environment or field of use (See MPEP 2106.05(h)). The claim as a whole merely describes how to generally “apply” the abstract idea by using generic computer components. The claimed computer components are recited at high level of generality and merely invoked as a tool to perform a process for determining similarity between goods and services (See MPEP 2106.05(f)). Simply implementing the abstract idea on a generic computer component is not a practical application. Accordingly, alone and in combination, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. This claim is directed to an abstract idea.
Step 2B: Claim 8 does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of using a computer system to perform a process for determining similarity between goods and services amount to no more than how to generally “apply” the exception using a generic computer component (See MPEP 2106.05(f)) and representing mere generally linking the use of the abstract idea to a particular technological environment or field of use (See MPEP 2106.05(h)). Mere instructions to apply an exception using a generic computer component and representing mere generally linking the use of the abstract idea to a particular technological environment or field of use cannot provide an inventive concept. As a result, this claim is not patent eligible.
Claims 9 and 14 are directed to substantially the same abstract idea as Claim 8 and are rejected for substantially the same reasons. The additional recited limitations of the dependent claims fail to establish that the claims do not recite an abstract idea because the additional recited limitations of the claims further narrow the abstract idea. These dependent claims further narrow the abstract idea of Claim 8 such as by defining “wherein said determining the plurality of goods and services similarity scores comprises determining a goods and services similarity score based on a proportion of historical trademark cases where a similarity assessment has determined that the first goods and services classification related to a first trademark is similar to the second goods and services classification related to a second trademark” in Claim 9 and by defining “wherein the first set of goods and services classifications or the second set of goods and services classifications include Nice Classification (NCL) classifications and/or sub-classifications” in Claim 14.
Step 2A, Prong 2: These dependent claims do not integrate the abstract idea into practical application because they do not recite additional elements.
Step 2B: These dependent claims do not amount to significantly more than the abstract idea because they do not recite additional elements. Therefore, these claims are not patent eligible.
Claims 10-11 and 13 are directed to substantially the same abstract idea as Claim 8 and are rejected for substantially the same reasons. The additional recited limitations of the dependent claims fail to establish that the claims do not recite an abstract idea because the additional recited limitations of the claims further narrow the abstract idea. These dependent claims further narrow the abstract idea of Claim 8 such as by defining “wherein said determining the plurality of goods and services similarity scores comprises: determining a goods and services similarity score using … using data from historical trademark cases where a similarity assessment has been performed” in Claim 10, by defining “wherein the instructions, when executed by …, further cause … to: determine a first set of text fragments that are semantically similar to the first goods and services description and a second set of text fragments that are semantically similar to the second goods and services description; and provide the first set of text fragments and the second set of text fragments to the first … model; wherein the first … model: determines the first set of goods and services classifications based on the first goods and services description and the first set of text fragments, and determines the second set of goods and services classifications based on the second goods and services description and the second set of text fragments” in Claim 11, and by defining “wherein the instructions, when executed by the processor, further cause the processor to: determine a main goods and services classification associated with the first goods and services description, wherein the first set of goods and services classifications provided by the first … model include sub-classifications of the main goods and services classification” in Claim 13.
Step 2A, Prong 2: Claims 10-11 and 13 do not integrate the abstract idea into practical application. Claim 10 recites an additional element – “a machine learning prediction model that is trained”, and Claims 11 and 13 recite additional elements – “the processor” and “machine learning”. The additional element “the processor” amounts to no more than mere instructions to apply the exception using generic computer components (See MPEP 2106.05(f)). Other additional elements amount to no more than representing mere generally linking the use of the judicial exception (the abstract idea) to a particular technological environment or field of use (See MPEP 2106.05(h)). Accordingly, alone and in combination, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
Step 2B: Claims 10-11 and 13 do not amount to significantly more than the abstract idea. Claims 10-11 and 13 do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of using a computer system to perform a process for determining similarity between goods and services amount to no more than how to generally “apply” the exception using a generic computer component (See MPEP 2106.05(f)) and representing mere generally linking the use of the abstract idea to a particular technological environment or field of use (See MPEP 2106.05(h)). Mere instructions to apply an exception using a generic computer component and representing mere generally linking the use of the abstract idea to a particular technological environment or field of use cannot provide an inventive concept. As a result, these claims are not patent eligible.
Claim 15 recites the following limitations:
… to:
receive a query including a first goods and services description and a second goods and services description;
provide the first goods and services description and the second goods and services description to a first … model;
receive, from the first … model, a first set of goods and services classifications for the first goods and services description and a second set of goods and services classifications for the second goods and services description;
determine a plurality of goods and services similarity scores, each goods and services similarity score of the determined plurality of goods and services similarity scores indicating a similarity between a first goods and services classification from the first set of goods and services classifications and a second goods and services classification from the second set of goods and services classifications;
determine an aggregate goods and services similarity score based on the determined plurality of goods and services similarity scores; and
provide the aggregate goods and services similarity score as a query result.
Step 2A, Prong 1: The limitations for Claim 15 described above fall within “Certain Methods of Organizing Human Activity” for managing personal interactions between people such as following rules and instructions. Accordingly, this claim recites an abstract idea.
Step 2A, Prong 2: This judicial exception is not integrated into a practical application. Claim 15 recites additional elements – “a computer-readable medium comprising computer-executable instructions, that when executed by a processor, causes the processor” and “machine learning”. The additional element “machine learning” represents mere generally linking of the use of the judicial exception (the abstract idea) to a particular technological environment or field of use (See MPEP 2106.05(h)). The claim as a whole merely describes how to generally “apply” the abstract idea by using generic computer components. The claimed computer components are recited at high level of generality and merely invoked as a tool to perform a process for determining similarity between goods and services (See MPEP 2106.05(f)). Simply implementing the abstract idea on a generic computer component is not a practical application. Accordingly, alone and in combination, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. This claim is directed to an abstract idea.
Step 2B: Claim 15 does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of using a computer system to perform a process for determining similarity between goods and services amount to no more than how to generally “apply” the exception using a generic computer component (See MPEP 2106.05(f)) and representing mere generally linking the use of the abstract idea to a particular technological environment or field of use (See MPEP 2106.05(h)). Mere instructions to apply an exception using a generic computer component and representing mere generally linking the use of the abstract idea to a particular technological environment or field of use cannot provide an inventive concept. As a result, this claim is not patent eligible.
Claims 16 and 20 are directed to substantially the same abstract idea as Claim 15 and are rejected for substantially the same reasons. The additional recited limitations of the dependent claims fail to establish that the claims do not recite an abstract idea because the additional recited limitations of the claims further narrow the abstract idea. These dependent claims further narrow the abstract idea of Claim 15 such as by defining “wherein said determining the plurality of goods and services similarity scores comprises: determining a goods and services similarity score based on a proportion of historical trademark cases where a similarity assessment has determined that the first goods and services classification related to a first trademark is similar to the second goods and services classification related to a second trademark” in Claim 16 and by defining “wherein the first set of goods and services classifications or the second set of goods and services classifications include Nice Classification (NCL) classifications and/or sub-classifications” in Claim 20.
Step 2A, Prong 2: These dependent claims do not integrate the abstract idea into practical application because they do not recite additional elements.
Step 2B: These dependent claims do not amount to significantly more than the abstract idea because they do not recite additional elements. Therefore, these claims are not patent eligible.
Claims 17-18 are directed to substantially the same abstract idea as Claim 15 and are rejected for substantially the same reasons. The additional recited limitations of the dependent claims fail to establish that the claims do not recite an abstract idea because the additional recited limitations of the claims further narrow the abstract idea. These dependent claims further narrow the abstract idea of Claim 15 such as by defining “wherein said determining the plurality of goods and services similarity scores comprises: determining a goods and services similarity score using … using data from historical trademark cases where a similarity assessment has been performed” in Claim 17 and by defining “wherein the instructions, when executed by …, further cause … to: determine a first set of text fragments that are semantically similar to the first goods and services description and a second set of text fragments that are semantically similar to the second goods and services description; and provide the first set of text fragments and the second set of text fragments to the first … model; wherein the first … model: determines the first set of goods and services classifications based on the first goods and services description and the first set of text fragments, and determines the second set of goods and services classifications based on the second goods and services description and the second set of text fragments” in Claim 18.
Step 2A, Prong 2: Claims 17-18 do not integrate the abstract idea into practical application. Claim 17 recites an additional element – “a machine learning prediction model that is trained”, and Claim 18 recites additional elements – “the processor” and “machine learning”. The additional element “the processor” amounts to no more than mere instructions to apply the exception using generic computer components (See MPEP 2106.05(f)). Other additional elements amount to no more than representing mere generally linking the use of the judicial exception (the abstract idea) to a particular technological environment or field of use (See MPEP 2106.05(h)). Accordingly, alone and in combination, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
Step 2B: Claims 17-18 do not amount to significantly more than the abstract idea. Claims 17-18 do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of using a computer system to perform a process for determining similarity between goods and services amount to no more than how to generally “apply” the exception using a generic computer component (See MPEP 2106.05(f)) and representing mere generally linking the use of the abstract idea to a particular technological environment or field of use (See MPEP 2106.05(h)). Mere instructions to apply an exception using a generic computer component and representing mere generally linking the use of the abstract idea to a particular technological environment or field of use cannot provide an inventive concept. As a result, these claims are not patent eligible.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-4, 6-11, 13-18, and 20 are rejected under 35 U.S.C. 102(a)(1) as being clearly anticipated by Keyngnaert et al. (US PG Pub. No. 2016/0260033 A1; hereinafter "Keyngnaert").
Regarding Claim 1, Keyngnaert teaches a method for determining a similarity between goods and services, comprising: receiving a query including a first goods and services description and a second goods and services description; providing the first goods and services description and the second goods and services description to a first machine learning model (See “FIG. 1 is a block diagram of a goods and/or services similarity engine 100 that can be implemented in accordance with embodiments of the present disclosure. The goods and/or services similarity engine 100 can include a distribution analyzer 110 and machine learning engine 120 configured to implement word relatedness component 122. The goods and/or services similarity engine 100 can be configured to receive input data 130 including an input string forming a reference description of goods and/or services and one or more trademark classes to be associated with the reference description of goods and/or services. For each of the one or more trademark classes included in the input data, the goods and/or services similarity engine 100 can be trained to detect and distinguish between words/terms in the reference description of goods and/or services that are strongly associated with the one or more class (kernel terms), words/terms in the reference description of goods and/or services that are generic across the trademark classes (generic terms), and/or words/terms that can modify the words that are kernel terms (modifier terms), and/or process descriptions of goods and/or services associated with trademarks stored in one or more trademark repositories and one or more input descriptions of goods and/or services. In response to detecting and distinguishing kernel terms, modifier terms, and generic terms, the goods and/or services similarity engine 100 can be trained to determine similarity values between the word/terms in the reference description of goods and/or services and the words/terms in the description of goods and/or services in existing trademarks based on the type of term (e.g., kernel, modifier, generic).” in Paragraph [0061] and “The machine learning engine 120 implemented by the goods and/or services similarity engine 100 can utilize machine learning techniques, such as deep neural network techniques and/or any other suitable machine learning techniques. For example, the similarity engine can implement word relatedness based on, for example, word embeddings that can be trained based on a corpus of existing descriptions of goods and/or services for registered trademarks in the training data 140.” in Paragraph [0067]); receiving, from the first machine learning model, a first set of goods and services classifications for the first goods and services description and a second set of goods and services classifications for the second goods and services description (See “As a non-limiting example, the input data reference description of goods and/or services can be “PAPER CUPS FOR HOLDING POPCORN” for class 21 (Housewares and Glass). FIGS. 3A-B show example normalized distributions 300 and 350 for the terms “CUP” and “POPCORN”, respectively, as a function of the trademark class.” in Paragraph [0070], “In exemplary embodiments, scores can be generated for each token in the reference description of goods and/or services as they relate to the specified class(es) in the input data 130 (e.g., class 21) and as they relate to a class (or classes) outside of the specified class in the input data 130 (e.g., classes other than class 21).” in Paragraph [0071], “FIG. 4A shows a graph 400 illustrating class specific scores 401-405 for the terms “PAPER”, “CUPS”, “FOR”, “HOLDING”, and “POPCORN”, respectively, for the specified class 21. As shown in graph 400, the term “CUPS” is the only term that is strongly associated with class 21 based on its class specific score, and therefore is identified by the distribution analyzer 110 as the only kernel in the reference description of goods and/or services. FIG. 4B shows a graph 450 illustrating excluded class specific scores 451-455 for the terms “PAPER”, “CUPS”, “FOR”, “HOLDING”, and “POPCORN”, respectively, for classes outside of the specified class 21. As shown in graph 450, the terms “PAPER”, “HOLDING”, and “POPCORN” have strong associations with classes outside of class 21, and therefore are identified by the distribution analyzer 110 as the modifiers in the reference description of goods and/or services.” in Paragraph [0072], “The distributed analyzer 110 can utilize the normalized scaled class distribution generated for tokens based on the distribution of tokens across the class using the training data to generate distribution similarity scores between two tokens. For example, distribution similarity scores between tokens in the reference description of goods and tokens in one of the description of goods in the set of trademark data 140 can be generated by the distribution analyzer 110.” in Paragraph [0073], “Similarity scores between two descriptions of goods and/or services T and U can be generated by the distribution analyzer 110 based on the class specific scores and the excluded class specific scores. FIG. 5A shows class specific similarity scores 502 being generated between tokens 511-515 (e.g., tokens t1 to tn) of a first description of goods T (e.g., PAPER CUPS FOR HOLDING POPCORN) and tokens 551-556 (e.g., tokens t1 to tm) of a second description of goods U (e.g., PAPER CUPS AND CONTAINERS FOR FOOD) based on class specific scores and FIG. 5B shows similarity scores 504 being generated between tokens 511-515 of a first description of goods and/or services T and tokens 551-556 of a second description of goods and/or services U based on excluded class specific scores.” in Paragraph [0074], and “When determining a description similarity between the reference description of goods and/or services and one of the descriptions of goods and/or services in the set of trademark data 140, the distribution analyzer 110 can utilize a similarity function to determine the similarity between the two descriptions of goods and/or services by determining the normalized similarity of the two descriptions of goods and/or services based on the class specific score and determining the similarity of the two descriptions of goods and/or services based on the excluded class specific score and combining these (potentially weighted) in an overall score.” in Paragraph [0075]); determining a plurality of goods and services similarity scores, each goods and services similarity score of the determined plurality of goods and services similarity scores indicating a similarity between a first goods and services classification from the first set of goods and services classifications and a second goods and services classification from the second set of goods and services classifications (See “Referring still to FIG. 1, the machine learning engine 120 can utilize the vectors generated for tokens based on word relatedness using the training data to generate word relatedness similarity scores between two tokens. For example, similarity scores between tokens in the reference description of goods and/or services and tokens in one of the description of goods and/or services in the set of trademark data 140 can be generated by the machine learning engine 120.” in Paragraph [0076], “The similarity engine 100 can use one or more of the similarity scores generate by the distribution analyzer 110 and/or machine learning engine 120 to generate an overall similarity score. The similarity engine 100 can use the word relatedness similarity score, the distribution similarity score, the class specific similarity scores, the excluded class specific similarity scores, and/or a combination thereof. For example, the similarity engine can generate an aggregate of the similarity scores. Each similarity score that forms the aggregate can be assigned a weighting factor to emphasis some of the similarity scores contributions to overall similarity score and to de-emphasize some of the similarity scores contributions to the overall similarity score.” in Paragraph [0077] and “The output of the goods and/or services similarity engine 100 can include the goods and/or services similarity scores and an association between the goods and/or services similarity scores and each of the corresponding descriptions of goods and/or services of the trademark. For example, the goods and/or services similarity engine 100 can output the class specific similarity score, the excluded class specific similarity score, and the overall similarity score for each comparison between the reference description of goods and/or services and the descriptions of goods and/or services in the set of trademark data 140.” in Paragraph [0078]); determining an aggregate goods and services similarity score based on the determined plurality of goods and services similarity scores (See “The similarity engine 100 can use one or more of the similarity scores generate by the distribution analyzer 110 and/or machine learning engine 120 to generate an overall similarity score. The similarity engine 100 can use the word relatedness similarity score, the distribution similarity score, the class specific similarity scores, the excluded class specific similarity scores, and/or a combination thereof. For example, the similarity engine can generate an aggregate of the similarity scores. Each similarity score that forms the aggregate can be assigned a weighting factor to emphasis some of the similarity scores contributions to overall similarity score and to de-emphasize some of the similarity scores contributions to the overall similarity score.” in Paragraph [0077]); and providing the aggregate goods and services similarity score as a query result (See “For example, the similarity engine can generate an aggregate of the similarity scores.” in Paragraph [0077] and “The goods and/or services similarity values output by the similarity engine can be used, for example, in post-search processes after search results have been retrieved from the trademark repository and before the results are presented to the user via a graphical user interface. For example, the similarity values can be used by a search engine to facilitate detection of trademarks in the search results that are more relevant to a search string received from a user (e.g., a trademark name, a reference description of goods and/or services, and associated classes). In embodiments of the present disclosure, the trademarks returned by a search that have the highest similarities between the reference description of goods and/or services and the description of goods and/or services for the trademarks can be ranked or prioritized by the search engine based on the similarity values to present those trademarks to a user as being particularly relevant or important.” in Paragraph [0079]).
Regarding Claim 2, Keyngnaert teaches all the limitations of Claim 1 as described above. Keyngnaert also teaches wherein said determining the plurality of goods and services similarity scores comprises determining a goods and services similarity score based on a proportion of historical trademark cases where a similarity assessment has determined that the first goods and services classification related to a first trademark is similar to the second goods and services classification related to a second trademark (See “Each of the example resources can have specific strengths and specific weaknesses, such that none of the resource can be used alone to provide a complete or accurate semantic model for trademarks. The approach to get to a complete and accurate model is by identifying those combination of patterns of the aforementioned resources of the semantic modeler 1048 that lead to valid results based on historic data and/or training data. The semantic modeler 1048 can interact with each of the resources to identify whether each of the resources consider which words the resources consider to be semantically related to a term or element. A resource may vote for a particular word by indicating that it is semantically related to a term. The semantic modeler 1048 can consider the votes from the resources to select the most correct, accurate, or appropriate semantic relationships based on the voting patterns of the resources. For example, it can be determined that certain voting patterns can provide correct semantic relationships. The voting patterns can be integrated into logic of the semantic modeler 1048 that decides which conditions allow for which patterns to be used.” in Paragraph [0211]).
Regarding Claim 3, Keyngnaert teaches all the limitations of Claim 1 as described above. Keyngnaert also teaches wherein said determining the plurality of goods and services similarity scores comprises: determining a goods and services similarity score using a machine learning prediction model that is trained using data from historical trademark cases where a similarity assessment has been performed (See “FIG. 1 is a block diagram of a goods and/or services similarity engine 100 that can be implemented in accordance with embodiments of the present disclosure. The goods and/or services similarity engine 100 can include a distribution analyzer 110 and machine learning engine 120 configured to implement word relatedness component 122. The goods and/or services similarity engine 100 can be configured to receive input data 130 including an input string forming a reference description of goods and/or services and one or more trademark classes to be associated with the reference description of goods and/or services. For each of the one or more trademark classes included in the input data, the goods and/or services similarity engine 100 can be trained to detect and distinguish between words/terms in the reference description of goods and/or services that are strongly associated with the one or more class (kernel terms), words/terms in the reference description of goods and/or services that are generic across the trademark classes (generic terms), and/or words/terms that can modify the words that are kernel terms (modifier terms), and/or process descriptions of goods and/or services associated with trademarks stored in one or more trademark repositories and one or more input descriptions of goods and/or services. In response to detecting and distinguishing kernel terms, modifier terms, and generic terms, the goods and/or services similarity engine 100 can be trained to determine similarity values between the word/terms in the reference description of goods and/or services and the words/terms in the description of goods and/or services in existing trademarks based on the type of term (e.g., kernel, modifier, generic).” in Paragraph [0061], and “The machine learning engine 120 implemented by the goods and/or services similarity engine 100 can utilize machine learning techniques, such as deep neural network techniques and/or any other suitable machine learning techniques. For example, the similarity engine can implement word relatedness based on, for example, word embeddings that can be trained based on a corpus of existing descriptions of goods and/or services for registered trademarks in the training data 140. Initially, the tokens generated for the words/terms in the descriptions of goods and/or services in the training data are assigned arbitrary or random vector values. Based on the training data (the description of goods and/or services), the machine learning engine 120 can map the tokens to, for example, vectors, where each vectors includes a real number. The mapping can be based on vectors associated with tokens that surround the token for which a vector is being generated. For example, a vector for a token in a description of goods and/or services can be determined, at least in part, based on the tokens to the left or right of the token. By defining a vector of a token based on the tokens that surround the token, the machine learning engine 120 can define context-based vectors such that a similarity between two tokens can be determined based on the context within which the two tokens are used as opposed to on the form or content of the token itself. Tokens having identical or similar vectors can be identified by the machine learning algorithm as being similar. In some embodiments, the word relatedness component 122 of the machine learning engine 120 can utilize stemming, inflections, normalizations, and contextual usage to generate the vectors.” in Paragraph [0067]).
Regarding Claim 4, Keyngnaert teaches all the limitations of Claim 1 as described above. Keyngnaert also teaches determining a first set of text fragments that are semantically similar to the first goods and services description and a second set of text fragments that are semantically similar to the second goods and services description; and providing the first set of text fragments and the second set of text fragments to the first machine learning model (See “Similarity scores between two descriptions of goods and/or services T and U can be generated by the distribution analyzer 110 based on the class specific scores and the excluded class specific scores. FIG. 5A shows class specific similarity scores 502 being generated between tokens 511-515 (e.g., tokens t1 to tn) of a first description of goods T (e.g., PAPER CUPS FOR HOLDING POPCORN) and tokens 551-556 (e.g., tokens t1 to tm) of a second description of goods U (e.g., PAPER CUPS AND CONTAINERS FOR FOOD) based on class specific scores and FIG. 5B shows similarity scores 504 being generated between tokens 511-515 of a first description of goods and/or services T and tokens 551-556 of a second description of goods and/or services U based on excluded class specific scores.” in Paragraph [0074], “When determining a description similarity between the reference description of goods and/or services and one of the descriptions of goods and/or services in the set of trademark data 140, the distribution analyzer 110 can utilize a similarity function to determine the similarity between the two descriptions of goods and/or services by determining the normalized similarity of the two descriptions of goods and/or services based on the class specific score and determining the similarity of the two descriptions of goods and/or services based on the excluded class specific score and combining these (potentially weighted) in an overall score.” in Paragraph [0075], “The training data can be used to train the distribution analyzer 110 and the machine learning engine 120. As shown in FIG. 2, each trademark (represented by identifiers 202, e.g., GUID-1-GUID-N) in the training data 140 can be associated with one or more trademark classes 204 and a description of goods and/or services 206. During the training process, trademarks that include multiple classes are split into separate entries. For example, the trademark associated with GUID-1 can include two classes, and therefore, can be split into two separate entries, where a first entry 208 includes the identifier, GUID-1, a first one of the classes 210, and a description of goods and/or services 212 for the first one of the classes 210, and a second entry 214 includes the identifier, GUID-1, a second one of the classes 216, and a description of goods and/or services 218 for the second one of the classes 216. The words/terms included in the descriptions of goods and/or services for each entry can be tokenized using tokenization processes described herein. For example, the words/terms in the first entry 208 can be tokenized such that the words/terms in the description of goods and/or services are replaced by tokens 220, e.g., token1-tokenk.” in Paragraph [0065], and Fig. 2); wherein the first machine learning model: determines the first set of goods and services classifications based on the first goods and services description and the first set of text fragments, and determines the second set of goods and services classifications based on the second goods and services description and the second set of text fragments (See “As a non-limiting example, the input data reference description of goods and/or services can be “PAPER CUPS FOR HOLDING POPCORN” for class 21 (Housewares and Glass). FIGS. 3A-B show example normalized distributions 300 and 350 for the terms “CUP” and “POPCORN”, respectively, as a function of the trademark class.” in Paragraph [0070], “In exemplary embodiments, scores can be generated for each token in the reference description of goods and/or services as they relate to the specified class(es) in the input data 130 (e.g., class 21) and as they relate to a class (or classes) outside of the specified class in the input data 130 (e.g., classes other than class 21).” in Paragraph [0071], “FIG. 4A shows a graph 400 illustrating class specific scores 401-405 for the terms “PAPER”, “CUPS”, “FOR”, “HOLDING”, and “POPCORN”, respectively, for the specified class 21. As shown in graph 400, the term “CUPS” is the only term that is strongly associated with class 21 based on its class specific score, and therefore is identified by the distribution analyzer 110 as the only kernel in the reference description of goods and/or services. FIG. 4B shows a graph 450 illustrating excluded class specific scores 451-455 for the terms “PAPER”, “CUPS”, “FOR”, “HOLDING”, and “POPCORN”, respectively, for classes outside of the specified class 21. As shown in graph 450, the terms “PAPER”, “HOLDING”, and “POPCORN” have strong associations with classes outside of class 21, and therefore are identified by the distribution analyzer 110 as the modifiers in the reference description of goods and/or services.” in Paragraph [0072], “The distributed analyzer 110 can utilize the normalized scaled class distribution generated for tokens based on the distribution of tokens across the class using the training data to generate distribution similarity scores between two tokens. For example, distribution similarity scores between tokens in the reference description of goods and tokens in one of the description of goods in the set of trademark data 140 can be generated by the distribution analyzer 110.” in Paragraph [0073], “Similarity scores between two descriptions of goods and/or services T and U can be generated by the distribution analyzer 110 based on the class specific scores and the excluded class specific scores. FIG. 5A shows class specific similarity scores 502 being generated between tokens 511-515 (e.g., tokens t1 to tn) of a first description of goods T (e.g., PAPER CUPS FOR HOLDING POPCORN) and tokens 551-556 (e.g., tokens t1 to tm) of a second description of goods U (e.g., PAPER CUPS AND CONTAINERS FOR FOOD) based on class specific scores and FIG. 5B shows similarity scores 504 being generated between tokens 511-515 of a first description of goods and/or services T and tokens 551-556 of a second description of goods and/or services U based on excluded class specific scores.” in Paragraph [0074], “When determining a description similarity between the reference description of goods and/or services and one of the descriptions of goods and/or services in the set of trademark data 140, the distribution analyzer 110 can utilize a similarity function to determine the similarity between the two descriptions of goods and/or services by determining the normalized similarity of the two descriptions of goods and/or services based on the class specific score and determining the similarity of the two descriptions of goods and/or services based on the excluded class specific score and combining these (potentially weighted) in an overall score.” in Paragraph [0075], and “FIG. 1 is a block diagram of a goods and/or services similarity engine 100 that can be implemented in accordance with embodiments of the present disclosure. The goods and/or services similarity engine 100 can include a distribution analyzer 110 and machine learning engine 120 configured to implement word relatedness component 122. The goods and/or services similarity engine 100 can be configured to receive input data 130 including an input string forming a reference description of goods and/or services and one or more trademark classes to be associated with the reference description of goods and/or services. For each of the one or more trademark classes included in the input data, the goods and/or services similarity engine 100 can be trained to detect and distinguish between words/terms in the reference description of goods and/or services that are strongly associated with the one or more class (kernel terms), words/terms in the reference description of goods and/or services that are generic across the trademark classes (generic terms), and/or words/terms that can modify the words that are kernel terms (modifier terms), and/or process descriptions of goods and/or services associated with trademarks stored in one or more trademark repositories and one or more input descriptions of goods and/or services.” in Paragraph [0061]).
Regarding Claim 6, Keyngnaert teaches all the limitations of Claim 1 as described above. Keyngnaert also teaches determining a main goods and services classification associated with the first goods and services description, wherein the first set of goods and services classifications provided by the first machine learning model include sub-classifications of the main goods and services classification (See “As a non-limiting example, the input data reference description of goods and/or services can be “PAPER CUPS FOR HOLDING POPCORN” for class 21 (Housewares and Glass). FIGS. 3A-B show example normalized distributions 300 and 350 for the terms “CUP” and “POPCORN”, respectively, as a function of the trademark class. As shown in FIG. 3A, the term “CUP” has a peak 302 (e.g., a high frequency of occurrence) in class 21, and a generally lower value across the remaining classes. Based on this normalized class distribution, the distribution analyzer 110 can determine that the term “CUP” is a kernel term for the reference description of goods and/or services because it is strongly associated with class 21, which was specified in the input data 130. Similarly, as shown in FIG. 3B, the term “POPCORN” has peaks 352 and 354 (e.g., a high frequency of occurrence) in classes 30 and 31, and a generally lower value across the remaining classes. Based on this normalized class distribution, the distribution analyzer 110 can determine that the term “POPCORN” is not a kernel term for the reference description of goods and/or services because it is not strongly associated with class 21, but is a modifier term because it is strongly associated with one or more other classes. Similar distributions and analyses can be performed for each of the words/terms in the reference description of goods and/or services and for the class or classes specified in the input data 130 to determine whether the words/terms are kernels, modifiers, or generics.” in Paragraph [0070], “In exemplary embodiments, scores can be generated for each token in the reference description of goods and/or services as they relate to the specified class(es) in the input data 130 (e.g., class 21) and as they relate to a class (or classes) outside of the specified class in the input data 130 (e.g., classes other than class 21).” in Paragraph [0071], and “FIG. 4A shows a graph 400 illustrating class specific scores 401-405 for the terms “PAPER”, “CUPS”, “FOR”, “HOLDING”, and “POPCORN”, respectively, for the specified class 21. As shown in graph 400, the term “CUPS” is the only term that is strongly associated with class 21 based on its class specific score, and therefore is identified by the distribution analyzer 110 as the only kernel in the reference description of goods and/or services. FIG. 4B shows a graph 450 illustrating excluded class specific scores 451-455 for the terms “PAPER”, “CUPS”, “FOR”, “HOLDING”, and “POPCORN”, respectively, for classes outside of the specified class 21. As shown in graph 450, the terms “PAPER”, “HOLDING”, and “POPCORN” have strong associations with classes outside of class 21, and therefore are identified by the distribution analyzer 110 as the modifiers in the reference description of goods and/or services.” in Paragraph [0072]).
Regarding Claim 7, Keyngnaert teaches all the limitations of Claim 1 as described above. Keyngnaert also teaches wherein the first set of goods and services classifications or the second set of goods and services classifications include Nice Classification (NCL) classifications and/or sub-classifications (See “As described herein, the order 802 can include an input string (e.g., of textual elements) forming one or more words for the trademarks that the user wishes to search. The order 802 can also include international classes to be searched (e.g., as defined in the World Intellectual Property Organization (WIPO) classification) and lists of jurisdictions to be searched, as well as a list of input strings forming one or more words that indicate for which goods and services the searched trademark(s) will be or are associated with (e.g., a reference description of goods and/or services).” in Paragraph [0101] wherein the “international classes (e.g. as defined in the World Intellectual Property Organization (WIPO) classification)” is interpreted to be “Nice Classification (NCL) classifications” as described in Paragraph [0024] of the specification.).
Claims 8-11 and 13-14 are system claims corresponding to method Claims 1-4 and 6-7. All of the limitations in Claims 8-11 and 13-14 are found reciting the same scopes of the respective limitations in Claims 1-4 and 6-7. Accordingly, Claims 8-11 and 13-14 are considered anticipated by the same rationales presented in the rejection of Claims 1-4 and 6-7, respectively set forth above.
Claims 15-18 and 20 are product claims corresponding to method Claims 1-4 and 7. All of the limitations in Claims 15-18 and 20 are found reciting the same scopes of the respective limitations in Claims 1-4 and 7. Accordingly, Claims 15-18 and 20 are considered anticipated by the same rationales presented in the rejection of Claims 1-4 and 7, respectively set forth above.
REASONS FOR ALLOWANCE
The following is an examiner’s statement of reasons for allowance: Examiner knows of no art which teaches or suggests alone, or in combination with other art, dependent claims 5, 12, and 19 in their entirety; and in particular, “ranking the goods and services classifications from the first set of goods and services classifications and the third set of goods and services classifications based on a confidence probability associated with each of the goods and services classifications from the first set of goods and services classifications and the third set of goods and services classifications, wherein said determining a plurality of goods and services similarity scores comprises: determining goods and services similarity scores for highest ranking goods and services classifications having confidence probabilities that exceed a confidence probability threshold; or determining goods and services similarity scores for the goods and services classifications that appear most frequently when no goods and services classification have a confidence probability that exceeds the confidence probability threshold” in combination with other claim limitations, as recited in Claim 5, and similarly in Claims 12 and 19.
The closest prior art is found to be Keyngnaert in view of Jessen et al. (US 2016/0350886 A1; hereinafter, “Jessen”). Keyngnaert teaches a system and method for identifying similar trademarks from one or more repositories based on training a goods and/or services similarity engine to identify similarities between pairs of descriptions of goods and/or services in a corpus of training data that includes the descriptions of goods and/or services for registered trademarks and trademark classes associated with the descriptions of goods and/or services. Keyngnaert also teaches ranking based on the highest similarities between the reference description of goods and/or services and the description of goods and/or services for the trademarks. Jessen teaches a method and system for evaluating intellectual property. Jessen also teaches ranking of the results based on relevancy or similarity to the patent claim for infringement search. However, Keyngnaert in view of Jessen fails to teach “ranking the goods and services classifications from the first set of goods and services classifications and the third set of goods and services classifications based on a confidence probability associated with each of the goods and services classifications from the first set of goods and services classifications and the third set of goods and services classifications, wherein said determining a plurality of goods and services similarity scores comprises: determining goods and services similarity scores for highest ranking goods and services classifications having confidence probabilities that exceed a confidence probability threshold; or determining goods and services similarity scores for the goods and services classifications that appear most frequently when no goods and services classification have a confidence probability that exceeds the confidence probability threshold”.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TAYAR M KYU whose telephone number is (571)272-3419. The examiner can normally be reached Mon-Fri 9:00 am - 6:00 pm.
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, Jeffrey Zimmerman can be reached at 571-272-4602. 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.
/T.M.K./Examiner, Art Unit 3628
/GEORGE CHEN/Primary Examiner, Art Unit 3628
Prosecution Insights