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 .
This communication is in response to Applicant’s Arguments filed 04/12/2026. Claims 1, 7-8, 10, and 13 are amended and hereby entered. Claims 2-3 are cancelled. No claims are allowed.
Regarding 35 USC 101:
The applicant submits the amended claims recite more than an abstract idea as they recite a specific machine-executed Artificial intelligence driven content quality analysis (ACQA) workflow that improves the operation of a contact-center workflow. Specifically, the applicant states there are more additional elements than the generic computing components cited in the previous office action. However, the argument is not persuasive. While the now positively recited LLM is also considered an additional element, the rest of the claim limitations continue to fall under the abstract ideas of Mental Processes and Mathematical concepts. The examiner must look at the claims individually and as a whole, and the steps outlined in the claims recite a process previously performed mentally by humans (i.e. looking at posts, determining priority based on criteria or factors, using a scoring system to prioritize response, and routing information to the appropriate responder). Further, the additional elements are simply used to perform this quality analysis. The Applicant admits the analysis workflow is “machine-executed” further supporting the assertion that the abstract idea is being performed by generic computing components and an LLM recited at a high level of generality, see MPEP 2106.04(f). Therefore, both individually and as whole, the claims continue to recite an abstract idea without practical application or significantly more. The examiner respectfully disagrees with the applicant and the rejection is maintained.
Regarding 35 USC 103:
The applicant submits the amended limitations added to claim 1 (from canceled claim 2) is not taught by the office action. Specifically, the applicant argues Makanawala does not teach the ACQA sequence of preprocessing and analyzing text to yield a factor of quality. However, the claims are taken under broadest reasonable interpretation. The examiner cites Makanawala to show a specific version of processing and analyzing to yield quality factors, that meets the broad limitation recited by the applicant. Therefore, the examiner respectfully disagrees and the rejection is maintained.
Further, the applicant submits the secondary reference, NPL, at most teaches an off-the shelf model with a prompt for sentiment analysis, further stating the NPL is “limited” to a specific implementation of analysis. However, the claims are taken under broadest reasonable interpretation. The “limited” implementation provided by the NPL describes a specific implementation that meets the applicant’s broad recitation of claim limitations. Additionally, the steps described in the claims can be performed in conjunction with an off-the-shelf model, which is further supported by the specification paragraph 73 stating, “The LLM may be implemented by Generative Pretrained Transformer (GPT) models. For example, GPT models developed by OpenAI@, which may generate coherent and contextually relevant text based on a given prompt”. Therefore, the examiner respectfully disagrees with the applicant’s arguments and the rejection is maintained.
Further the applicant submits the examiner’s rationale to combine is too general. In response to applicant’s argument that there is no teaching, suggestion, or motivation to combine the references, the examiner recognizes that obviousness may be established by combining or modifying the teachings of the prior art to produce the claimed invention where there is some teaching, suggestion, or motivation to do so found either in the references themselves or in the knowledge generally available to one of ordinary skill in the art. See In re Fine, 837 F.2d 1071, 5 USPQ2d 1596 (Fed. Cir. 1988), In re Jones, 958 F.2d 347, 21 USPQ2d 1941 (Fed. Cir. 1992), and KSR International Co. v. Teleflex, Inc., 550 U.S. 398, 82 USPQ2d 1385 (2007). In this case, the examiner relies on knowledge generally available to one of ordinary skill in the art. In other words, it would have been obvious to one of ordinary skill to use an LLM to efficiently perform what a person would otherwise perform when analyzing text/media. This is shown in the rejection of claim 1, where the examiner states, “One of ordinary skill would have further recognized using an LLM response to analyze text and calculate scores would yield the predictable results of efficient text analysis of social media messages.” Therefore, the examiner respectfully disagrees and the rejection is maintained.
Further, the applicant submits the examiner does not explain why one of ordinary skill would have combined Makanawala, NPL, and Tanniru in a manner that yields the particular claim architecture. However, it is shown in the rejection that examiner relies on KSR rationale A “merely a combination of old elements”. In other words, the invention not only incorporates text processing but now image processing, which is a combination of old elements. Therefore, the examiner respectfully disagrees and the rejection is maintained.
Further, the applicant submits Tanniru does not teach continuous LLM training using labeled data updates. However, the examiner cites Tanniru paragraph 30, demonstrating a specific example of labeled training (i.e. using updated historical data to further update the model). Therefore, the examiner respectfully disagrees and the rejection is maintained.
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, 4-13 are rejected under 35 U.S.C. 101 because the claimed invention is directed to judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) with no practical application and without significantly more.
The claimed invention is directed to an abstract idea in that the instant application is directed to a mental process, see MPEP 2106.04(a)(2)(III). The independent claims (1 and 13) recite a method and systems to prioritize social media responses based on received data. These claim elements are being interpreted as concepts performed in the human mind (including observation, evaluation, judgement, and opinion). Using social media data to decide what message to address first can equivalently be achieved by human observation and evaluation of posts. The claims recite an abstract idea consistent with the “mental process” grouping set forth in the MPEP 2106.04(a)(2)(III).
Additionally, the claimed invention is directed to an abstract idea in that the instant application is directed to mathematical concepts, see MPEP 2106.04(a)(2)(I). The independent claims (1 and 13) recite a method and systems to calculate scores with mathematical relationships. The claims recite an abstract idea consistent with the “mathematical concepts” grouping set forth in the MPEP 2106.04(a)(2)(I).
The instant application fails to integrate the judicial exception into a practical application because the instant application merely recites an “apply it” (or an equivalent) with the judicial exception, or merely includes instructions to implement an abstract idea. The instant application is directed towards a method and systems to implement the identified abstract ideas of receiving information, processing information, and displaying the result of the analysis (i.e. processing social media information to recommend priority of customer response) and mathematical concepts (i.e. calculating a score using mathematical relationships), on a generically claimed computer structure. The claims do not include additional elements that integrate the abstract idea into practical application or amount to significantly more than the judicial exception. The independent claims recite the additional elements “a computerized system”, “a Large Language Model”, and “one or more processors”. These claim elements are recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a general computer environment. The machines merely act as a modality to implement the abstract idea and are not indicative of integration into a practical application (i.e., the additional elements are simply used as a tool to perform the abstract idea), see MPEP 2106.05(f).
Further, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed, the additional elements in the claims amount to no more than mere instructions to apply the exception using generic computer components. The same analysis applies here in step 2B analysis, and does not provide an inventive concept.
In regards to the dependent claims
Claims 2-3 (Cancelled)
Claims 4-12 introduce no new additional abstract ideas or new additional elements and do not impact analysis under 35 USC 101.
Claim 12 further limits the abstract idea of mathematical concepts and does not introduce any additional elements.
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 (i.e., changing from AIA to pre-AIA ) 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, 4-6, 9, and 11-13 are rejected under 35 U.S.C. 103 as being unpatentable over Makanawala (US 20130262320 A1) in view of (NPL: “Sentiment Analysis: Comprehensive Beginners Guide”) hereinafter referred to as NPL “Sentiment Analysis”.
Regarding Claim 1 and 13, (substantially similar in scope and language) Makanawala teaches:
A computerized-method for dynamically prioritizing social-media interactions in real-time, based on social-media posts in feeds, in a contact center, said computerized-method comprising: monitoring by one or more processors the social-media posts in the feeds of one or more social-media platforms which are integrated to the contact center, wherein said social-media posts have been published during a preconfigured period, for each social-media post of a customer in each feed in the feeds: [(Figure 2 and 3), (Para 0022-0023) “A "social activity management system" described in various embodiments may receive social media postings from social media users and facilitate the management of a company's resources to resolve the problems that are described in the social media postings.”]
said ACQA module comprises: preprocessing text and media elements in the social-media post; [(Figure 1 and 4), (Para 0024) “Broadly, a customer 101 on the left posts a social media message (such as a "social feed" 102 about a problem) to the social media systems 22 that, in turn, communicate the social media message 102 to the social activity management system 26 where it is analyzed and prioritized 103 before being inserted into a queue”, (Para 0038) “FIG. 4 is a block diagram illustrating the database 30, according to an embodiment. The database 30 may store customer information 3001, company resource information 3002, product information 3003, a knowledgebase repository 3004, configuration information 3005 and message information 3006”]
analyzing the preprocessed text and the preprocessed media elements to yield one or more factors of quality. [(Para 0024) “Broadly, a customer 101 on the left posts a social media message (such as a "social feed" 102 about a problem) to the social media systems 22 that, in turn, communicate the social media message 102 to the social activity management system 26 where it is analyzed and prioritized 103 before being inserted into a queue”, (Para 0045) “For example, the system may include a list or database of words or phrases such as "disappointed", "angry", "waiting, " "junk," and "excellent" to determine the sentiment of the author of the message. Words or phrases that signify a bad or poor sentiment may result in a message sentiment score”]
wherein said one or more factors of quality includes at least one of: i) relevance ii) accuracy: iii) clarity: iv) sentiment; and v) total-quality; [(Para 0048) “This internal sentiment aggregate score may also be factored into the determination of the message sentiment score and/or overall message score for the original social media message”]
calculating a quality-score by operating an Artificial intelligence (AI) driven Content Quality Analysis (ACQA) module, [(Para 0062) “In one embodiment the overall message score for the message may be the sum of each of the weighted parameter scores (i.e., the sum of each of the parameter scores multiplied by each of the associated weights).]
calculating a total-content quality score, by summing the one or more quality scores based on each quality-score preconfigured weight; and [(Para 0062) “In one embodiment the overall message score for the message may be the sum of each of the weighted parameter scores (i.e., the sum of each of the parameter scores multiplied by each of the associated weights).]
calculating a social-impact score based on the calculated total-content quality score and one or more parameters, [(Para 0029) “These parameters may be computed into a single priority score which determines and assigns a priority (e.g., Urgent, Normal and Low) to each message”, (Para 0062) “In one embodiment the overall message score for the message may be the sum of each of the weighted parameter scores (i.e., the sum of each of the parameter scores multiplied by each of the associated weights).]
wherein said social-impact score is updated in real-time; [(Para 0029) “The queue prioritization system 2811 may be utilized to provide auto prioritization of the message queue containing incoming social media messages to help agents focus on high profile customers and key issues.”]
automatically prioritizing the social-media posts based on the calculated social-impact score to yield a priorities queue of social-media interactions, wherein each social-media post represents a social-media interaction, [(Para 0042) “Broadly, the queue prioritization system 2811 may receive incoming social media messages, and then automatically evaluate the incoming messages for their importance using one or more procedures and based on one or more parameters”]
and automatically routing social-media interactions by a routine-engine to an available agent based on the yielded priorities queue of social-media interactions and agent skills [(Figures 1-3) “Accordingly, reassignment of the social media message to another agent or escalation to another agent]
However, Makanawala does not explicitly teach but NPL “Sentiment Analysis” does teach:
constructing a prompt of a Large Language Model (LLM), wherein the prompt includes the text of the post and instructions to assess the one or more factors of quality, [(Page 20) “For example, you could ask ‘What’s the theme and sentiment of the sentence below explained in simple terms’ or ‘What’s the sentiment of each sentence below? Say Positive, Neutral or Negative’.”]
sending the constructed prompt to be executed via an Application Programing Interface (API) platform of the LLM and receiving a response; [(Page 20) “Now in order to perform sentiment analysis you can use an off-the-shelf model and write a prompt describe the task you want it to do. For example, you could ask ‘What’s the theme and sentiment of the sentence below explained in simple terms’ or ‘What’s the sentiment of each sentence below? Say Positive, Neutral or Negative’.”, (Page 20, figure: testing chatgpt on tricky sentiment examples) “My coffee was great – Positive”]
parsing the response to extract one or more quality-scores; wherein the response is a string of text that includes a quality-score for each quality factor in the one or more factors of quality; and [(Page 20, figure: testing chatgpt on tricky sentiment examples), (Page 14) “The final step is to calculate the overall sentiment score for the text. As mentioned previously, this could be based on a scale of -100 to 100. In this case a score of 100 would be the highest score possible for positive sentiment. A score of 0 would indicate neutral sentiment. The score can also be expressed as a percentage, ranging from 0% as negative and 100% as positive”, (Page 20) “Now in order to perform sentiment analysis you can use an off-the-shelf model and write a prompt describe the task you want it to do. For example, you could ask ‘What’s the theme and sentiment of the sentence below explained in simple terms’ or ‘What’s the sentiment of each sentence below? Say Positive, Neutral or Negative’”]
Further, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the social media score calculation of text analysis (Makanawala) with use of prompting an external system for a response to aid in score calculations (NPL “Sentiment Analysis”). One of ordinary skill would have recognized that prompting an LLM to aid in text analysis would be beneficial when analyzing social media text for prioritization. One of ordinary skill would have further recognized using an LLM response to analyze text and calculate scores would yield the predictable results of efficient text analysis of social media messages.
Regarding Claim 2, Makanawala in view of NPL “Sentiment Analysis” teaches the limitations set forth above, Makanawala further teaches:
wherein said ACQA module further comprising before constructing of the prompt: (i) preprocessing text and media elements in the social-media post; and [(Figure 1 and 4), (Para 0024) “Broadly, a customer 101 on the left posts a social media message (such as a "social feed" 102 about a problem) to the social media systems 22 that, in turn, communicate the social media message 102 to the social activity management system 26 where it is analyzed and prioritized 103 before being inserted into a queue”, (Para 0038) “FIG. 4 is a block diagram illustrating the database 30, according to an embodiment. The database 30 may store customer information 3001, company resource information 3002, product information 3003, a knowledgebase repository 3004, configuration information 3005 and message information 3006”]
(ii) analyzing the preprocessed text and the preprocessed media elements to yield the one or more factors of quality; [(Para 0045) “For example, the system may include a list or database of words or phrases such as "disappointed", "angry", "waiting, " "junk," and "excellent" to determine the sentiment of the author of the message. Words or phrases that signify a bad or poor sentiment may result in a message sentiment score”]
Regarding Claim 4, Makanawala in view of NPL “Sentiment Analysis” teaches the limitations set forth above, Makanawala further teaches:
wherein said one or more parameters comprising at least one of: (i) user reach; (ii) engagement metrics; (iii) social-media post relevance; (iv) social-media post accuracy; (v) social-media post clarity; (vi) response time sensitivity; (vii) customer emotion intensity; (viii) contextual keywords; (ix) customer sentiment; (x) customer loyalty; and (xi) customer feedback. [(See at least, Para 0045) “For example, the system may include a list or database of words or phrases such as "disappointed", "angry", "waiting, " "junk," and "excellent" to determine the sentiment of the author of the message. Words or phrases that signify a bad or poor sentiment may result in a message sentiment score”]
Regarding Claim 5, Makanawala in view of NPL “Sentiment Analysis” teaches the limitations set forth above, Makanawala further teaches:
wherein said customer loyalty parameter of the customer is retrieved by the computerized-method further comprising: [(Para 0053) “the queue prioritization system may also determine the internal or external importance of the customer.”]
operating a social-media- feeds computation module, said social-media-feeds computation module comprising: [(Figures 1-3)]
retrieving the customer loyalty parameter of the customer, from a customers-database, and wherein the computerized-method further comprising: [(Para 0053) “For example, many companies may store information or history regarding known customers in an internal customer record or customer relationship management (CRM) record”, (Para 0054) “For example, even if the customer has never conducted business with the company before, the queue prioritization system 2811 may crawl a database or network (such as web pages accessible via the world wide web) for Information regarding the customer”]
constructing a social-impact-prompt LLM, that includes the text of the social-media post [(Para 0043) “Thereafter, the queue prioritization system 2811 may analyze each message to determine an overall message score for the message.”]
and instructions to assess at least one parameter of: (i) user reach; (ii) engagement metrics; (iii) social- media post relevance; (iv) social-media post accuracy; (v) social-media post clarity; (vi) response time sensitivity; (vii) customer emotion intensity; (viii) contextual keywords; (ix) customer sentiment; and (x) customer feedback; [(See at least Para 0054) “Depending on the quantity, sources and nature of information received, the queue prioritization system 2811 may determine that the customer is an "externally important" individual (e.g. a politician, celebrity, high profile business executive, etc.) that is associated with a score that may be used to compute a customer level score.”]
and sending the constructed social-impact-prompt to be executed via an API platform of the LLM and receiving a response that includes a score for each parameter. [(Figures 2-3), (Para 0043) “Thereafter, the queue prioritization system 2811 may analyze each message to determine an overall message score for the message”, (Para 0060) “The various parameters scores for each of the aforementioned parameters may each be associated with a weight that may be individually configured to customize the computation of respective weighted scores that are utilized to compute the overall message score for the message”]
Regarding Claim 6, Makanawala in view of NPL “Sentiment Analysis” teaches the limitations set forth above, Makanawala further teaches:
wherein said processing of the text includes at least one of: (i) tokenizing; (ii) lowercasing; (iii) removing punctuation and stop-word; (iv) text-feature extraction, wherein said text-feature extraction includes at least one of: (i) word frequency; (ii) Term Frequency-Inverse Document Frequency (TF-IDF); (iii) word embeddings; and (iv) contextual embeddings. [(Para 0035) “The expert finder system 2817 may utilize the text analysis system 2820 to identify the product and other keywords mentioned in the social media message and provide a list of recommended experts within the context of the social media message”, (Para 0045) “The queue prioritization system 2811 may utilize text analysis system 2820 to perform a text analysis (e.g. words) of the message to determine a message sentiment score. For example, the system may include a list or database of words or phrases such as "disappointed", "angry", "waiting, " "junk," and "excellent" to determine the sentiment of the author of the message.”
Regarding Claim 9, Makanawala in view of NPL “Sentiment Analysis” teaches the limitations set forth above, Makanawala further teaches:
wherein said computerized-method is further comprising normalizing the calculated total-content quality score to a standardized scale. [(Para 0063) “For example, assignment of a priority level 505 that is assigned to the message (e.g., low, medium, high) for display in a message queue may be based on thresholds values for the overall, message score”]
Regarding Claim 11, Makanawala in view of NPL “Sentiment Analysis” teaches the limitations set forth above, Makanawala further teaches:
wherein said computerized-method is further comprising forwarding the calculated social-impact score of each social-media post and the related social-media post to a recommendation engine, said recommendation engine comprising: sending the calculated social-impact score of each social-media post and the related social-media post to at least one of: (i)knowledgebase; (ii) agent-dashboard; (iii) reporting module; and (iv) supervisor dashboard. [(Para 0023) “The problems may be resolved by utilizing the resources of the social activity management system including agents, experts, local and third party knowledgebases as well as a multiplicity of internal systems that facilitate the prioritization, management, and resolution of customer problems.”]
Regarding Claim 12, Makanawala in view of NPL “Sentiment Analysis” teaches the limitations set forth above, Makanawala further teaches:
where said calculated social-impact score is according to formula I: (I) social-impact score = I (user reach * WI)+ (engagement metrics * W2) + (social media interaction relevance * W3) +(social media post accuracy * W4) +(social media post clarity*W5)+ (response time sensitivity * W6)+ (customer emotion intensity * W7) + (contextual keywords *W8) + (customer sentiment * W9) + (customer loyalty * W10) + (customer feedback*W1l), [(Para 0061) “The user interface 500 may be displayed by the queue prioritization system 2811 to enable a user (such as a customer service agent) to adjust weighting for each of the aforementioned parameter scores”, (Para 0062) “In one embodiment the overall message score for the message may be the sum of each of the weighted parameter scores (i.e., the sum of each of the parameter scores multiplied by each of the associated weights)”; the variables data types listed are nonfunctional descriptive material. However, art is still provided:]
whereby: the user reach is a parameter that measures an influence of the customer, [(Para 0054) “Depending on the quantity, sources and nature of information received, the queue prioritization system 2811 may determine that the customer is an "externally important" individual (e.g. a politician, celebrity, high profile business executive, etc.) that is associated with a score”, (Para 0056) “Customer Influence: The queue prioritization system 2811 may determine a score for customer influence of the customer that approximates the influence of the customer on various social media systems 22 (e.g., social networking platforms).”] the engagement metrics is a parameter that assesses level of engagement generated by the social- media post, [(Para 0049) “the message/post based on a number of views, feedback indicators (e.g., "likes"), messages, follow-up comments, shares, reposts, etc. that are posted by other users in association with the original message.”] the social-media post relevance is a parameter that evaluates relevance of the social-media post to objectives of the contact center, [(Para 0059)“Social Amplification: The queue prioritization system 2811 may determine a social amplification score of the message/post based on a number of same or similar messages/posts as reported by other users on social media systems.”] the social-media post accuracy is a parameter that refers to correctness of information presented in the social-media post, [(Para 0086) “the social media identity information of the user that posted the message, along with a match measurement indicator indicating the accuracy of the match”] the social-media post clarity is a parameter that assesses readability and comprehensibility of language used in the social-media post, [(Para 0032) “The recommended knowledgebase system may utilize the text analysis system 2820 to perform a text analysis and keyword identification of the social media message to identify a product and issue(s) in the social media message.”] the time sensitivity is a parameter that indicates urgency of the content of the social-media post, [(Para 0045) “For example, the identification of a bad or poor sentiment may result in a higher message sentiment score and higher overall message score that causes the message to be assigned to an agent more quickly and processed with greater urgency.”] the customer emotion intensity is a parameter that evaluates strength of emotions expressed in the social-media post, [(Para 0045) “For example, the system may include a list or database of words or phrases such as "disappointed", "angry", "waiting, " "junk," and "excellent" to determine the sentiment of the author of the message.”] the contextual keywords is a parameter that provides an analysis of presence of keywords relevant to domain of the contact center, [(Para 0045) “For example, the system may include a list or database of words or phrases such as "disappointed", "angry", "waiting, " "junk," and "excellent" to determine the sentiment of the author of the message.”] the customer sentiment is a parameter that assesses overall sentiment of the social-media post, [(Para 0048) “This internal sentiment aggregate score may also be factored into the determination of the message sentiment score and/or overall message score for the original social media message.”] the customer loyalty is a parameter that measures loyalty of the customer, [(Para 0053) “For example, many companies may store information or history regarding known customers in an internal customer record or customer relationship management (CRM) record (e.g., how long has the user been a customer, what is the quantity of products/services they have procured from the company, a user's previous purchase history with respect to the company, what is the sentiment history of messages/posts from the customer, etc.)”] the customer feedback is a parameter that captures response of other customers to the social- media post, [(Para 0046) “This message sentiment score for responses/comments to the original social media message may also be factored into the determination of the message sentiment score”]
and W1 - W 11 are preconfigured weights assigned to the parameters. [(Para 0061) “The various parameters scores for each of the aforementioned parameters may each be associated with a weight that may be individually configured to customize the computation of respective weighted scores that are utilized to compute the overall message score for the message.”]
Claims 7-8 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Makanawala (US 20130262320 A1) in view of NPL “Sentiment Analysis” in further view of Tanniru (US 20210295103 A1)
Regarding Claim 7, Makanawala in view of NPL “Sentiment Analysis” teach the limitations of claim 1,
While the combination teaches a computerized method that process text and media, it does not explicitly teach processing visual features:
wherein said processing media elements includes at least one of: (i) resizing; (ii) normalization; and (iii) visual-feature extraction, wherein said visual-feature extraction is operated by at least one of: (i) Convolutional Neural Networks (CNNs); and (ii) pretrained models.
However, Tanniru teaches:
wherein said processing media elements includes at least one of: (i) resizing; (ii) normalization; and (iii) visual-feature extraction, wherein said visual-feature extraction is operated by at least one of: (i) Convolutional Neural Networks (CNNs); and (ii) pretrained models. [(Para 0029) “The processing platform may train the convolutional neural network model with historical data (e.g., historical image data, historical label data, and historical data identifying the first set of fields and the second set of fields) to identify visual features of image data.”]
Further, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the method of processing text and media taught by Makanawala in view of NPL “Sentiment Analysis”, with the method of processing visual features taught by Tanniru. The claimed invention is merely a combination of old elements, and in combination each element would have performed the same function as it did separately. Processing images in addition to processing text for the same recommendation purposes would have yielded a predictable result to one of ordinary skill in the art.
Regarding Claim 8, the combination of Makanawala and NPL “Sentiment Analysis” teach the limitations of claim 1, Makanawala further teaches:
wherein said analyzing of the preprocessed text is performed by applying at least one of: (i) Natural Language Processing (NLP); (ii) sentiment analysis; and (iii) readability analysis, [(Para 0045) “The queue prioritization system 2811 may utilize text analysis system 2820 to perform a text analysis (e.g. words) of the message to determine a message sentiment score. For example, the system may include a list or database of words or phrases such as "disappointed", "angry", "waiting, " "junk," and "excellent" to determine the sentiment of the author of the message”]
While Makanawala in view of NPL “Sentiment Analysis” teaches a computerized method that analyzes text and media, it does not explicitly teach analyzing visual features:
and wherein said analyzing of the preprocessed media elements is operated by visual content analysis, said visual content analysis includes at least one technique of: (i) object detections; (ii) image classification; and (iii) content moderation
However, Tanniru teaches:
and wherein said analyzing of the preprocessed media elements is operated by visual content analysis, said visual content analysis includes at least one technique of: (i) object detections; (ii) image classification; and (iii) content moderation. [(Para 0029) “The processing platform may train the convolutional neural network model with historical data (e.g., historical image data, historical label data, and historical data identifying the first set of fields and the second set of fields) to identify visual features of image data.”]
Further, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the method of analyzing text and media taught by Makanawala in view of NPL “Sentiment Analysis”, with the method of analyzing visual features taught by Tanniru. The claimed invention is merely a combination of old elements, and in combination each element would have performed the same function as it did separately. Analyzing images in addition to text for the same recommendation purposes would have yielded a predictable result to one of ordinary skill in the art.
Regarding Claim 10, the combination of Makanawala and NPL “Sentiment Analysis” teach the limitations of claim 1
While Makanawala in view of NPL “Sentiment Analysis” teaches a model with labeled data, it does not explicitly teach continuously training the model:
wherein said LLM is continuously trained using labeled data updates to adapt to evolving content types and quality standards over time
However, Tanniru teaches:
wherein said LLM is continuously trained using labeled data updates to adapt to evolving content types and quality standards over time [(Para 0030) “In this case, the processing platform may provide the other system or device with historical data for use in training the convolutional neural network model, and may provide the other system or device with updated historical data to retrain the convolutional neural network model in order to update the convolutional neural network model”]
Further, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the method of using a model with labeled data taught by Makanawala in view of NPL “Sentiment Analysis”, with the method of training the model with updated data taught by Tanniru. The claimed invention is merely a combination of old elements, and in combination each element would have performed the same function as it did separately. Updating the model taught by Makanawala in view of NPL “Sentiment Analysis” with the training method of Tanniru would have yielded predictable results to one of ordinary skill in the art.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Examiner Benjamin Truong, whose telephone number is 703-756-5883. The examiner can normally be reached on Monday-Friday from 9 am to 5 pm (EST).
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, Nathan Uber SPE can be reached on 571-270-3923. 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.
/B.L.T. /Examiner, Art Unit 3687
/NATHAN C UBER/Supervisory Patent Examiner, Art Unit 3626
Prosecution Insights