DETAILED ACTION
This office action is in response to Applicant’s submission filed on 12/12/2024. Claims 1-20 are pending in the application of which Claims 1, 19, and 20 are independent and have been examined.
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 § 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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 5, 8, 10, 13, and 18-20 are rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee et al. (US20250363315A1)(herein "Chatterjee"), and in further view of Basnight et al. (US20250112845A1)(herein "Basnight").
Regarding claims 1, 19 and 20 Chatterjee teaches [One or more computing devices, comprising one or more processors, configured to: - claim 1], [A method performed by one or more computing devices, the method comprising: - claim 19], and [A system comprising: one or more processors; and a memory coupled with the one or more processors, the memory storing executable instructions that when executed by the one or more processors cause the one or more processors to effectuate operations comprising: - claim 20] (Chatterjee, Par. 0044:” … The process 400 may be implemented by the computing device 202 of the system 200.”, and Par. 0125:” … methods and systems may be implemented on a conventional or a general-purpose computer system, such as a personal computer (PC) or server computer. … The computer system 1702 may include a central processing unit (“CPU” or “processor”) 1704. The processor 1704 may include at least one data processor for executing program components for executing user-generated or system-generated requests.”, and Par. 0130:” The memory devices 1730 may store a collection of program or database components, including, without limitation, an operating system 1732, …”, and Par. 0133:” … the memory 1730 may store process-executable instructions, which when executed by the processor 1704, may cause the processor 14 to implement mixed language text understanding for GenAI models.”)
[determine training data comprising text of one or more languages for a machine translation model; - claim 1], [determining training data comprising text of one or more languages for a machine translation model; - claim 19], and [determining training data for a machine translation model; - claim 20] (Chatterjee, Par. 0081:” The pre-trained multilingual translation model 1110 with curriculum learning offers a systematic and pedagogically sound approach to understanding and translating codemixed languages. The generic pre-trained codemix understanding model 1112 so obtained may then be used for understanding codemix texts that include languages L1 and L2. By harnessing structured sequencing of training data based on the complexity metric, the pre-trained multilingual translation model 1110 establishes a strong foundation for subsequent domain specific fine-tuning. This methodical training regimen not only enhances performance of the pre-trained multilingual translation model 1110 in codemixed translation but also sets the stage for the creation of highly specialized and accurate domain specific codemix translation models.”)
determine/determining, based on the training data, one or more first model parameters for the machine translation model;- claim 1], [Chatterjee, Par. 0042:” … The model fine-tuning engine 320 calculates a complexity metric of each of a set of training data (obtained from the cross-domain codemix parallel corpus 316) based on a curriculum learning framework. Further, the model fine-tuning engine 320 ranks each of the set of training data based on the complexity metric. The curriculum learning framework enables the pre-trained multilingual translation model 324 to gradually learn from simpler to more complex data in the set of training data, thereby enhancing ability of the pre-trained multilingual translation model 324 to learn intricacies and nuances of various degrees and types of codemixing.”, and Par. 0120:” … Feedback from these evaluations is used to iteratively improve the domain specific codemix understanding model 1252, with adjustments made to the training data or model hyperparameters as necessary.”)
determine/determining, based on the one or more first model parameters, a first performance of the machine translation model; (Chatterjee, Par. 0120:” The process 1400 provides for evaluation and iteration. After each fine-tuning stage at steps 1410, the domain specific codemix understanding model 1252 undergoes rigorous evaluation using domain specific benchmarks and performance metrics. This includes evaluating ability of the domain specific codemix understanding model 1252 to handle codemixed text that accurately reflects real-world use cases within the domain. Feedback from these evaluations is used to iteratively improve the domain specific codemix understanding model 1252, with adjustments made to the training data or model hyperparameters as necessary.”)
determine/determining, based on the first performance, one or more second model parameters for the machine translation model; (Chatterjee, Par. 0120:” The process 1400 provides for evaluation and iteration. After each fine-tuning stage at steps 1410, the domain specific codemix understanding model 1252 undergoes rigorous evaluation using domain specific benchmarks and performance metrics. This includes evaluating ability of the domain specific codemix understanding model 1252 to handle codemixed text that accurately reflects real-world use cases within the domain. Feedback from these evaluations is used to iteratively improve the domain specific codemix understanding model 1252, with adjustments made to the training data or model hyperparameters as necessary.”)
determine/determining, based on the one or more second model parameters, information; and (Chatterjee, Par. 0120:” The process 1400 provides for evaluation and iteration. After each fine-tuning stage at steps 1410, the domain specific codemix understanding model 1252 undergoes rigorous evaluation using domain specific benchmarks and performance metrics. This includes evaluating ability of the domain specific codemix understanding model 1252 to handle codemixed text that accurately reflects real-world use cases within the domain. Feedback from these evaluations is used to iteratively improve the domain specific codemix understanding model 1252, with adjustments made to the training data or model hyperparameters as necessary.”) Note: At each iteration the result is compared to the previous result in order to evaluate the improvement that model is providing.
Chatterjee does not teach, however Basnight teaches transmit the information. (Basnight, Par. 0049:” … Each of the cores 404 is configured to transmit output data to the output pipeline circuits 308 of the nodes 402 operably associated with the core 404.”)
Basnight is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee further in view of Basnight to transmit the information. Motivation to do so would enable cross-lingual interoperability to extract insights from global data using monolingual models.
Regarding claim 5, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, further teaches wherein the one or more first model parameters are determined based on a plurality of hyperparameters. (Chatterjee, Par. Par. 0120:” … Feedback from these evaluations is used to iteratively improve the domain specific codemix understanding model 1252, with adjustments made to the training data or model hyperparameters as necessary.”)
Regarding claim 8, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, further teaches configured to determine a translation accuracy associated with the machine translation model, wherein the first performance is determined based on the translation accuracy. (Chatterjee, Par. 0081:” … This methodical training regimen not only enhances [accuracy] performance of the pre-trained multilingual translation model 1110 in codemixed translation but also sets the stage for the creation of highly specialized and accurate domain specific codemix translation models.”) Note: to enhance translation accuracy, implies first to determine a translation accuracy.
Regarding claim 10, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, further teaches determine, based on the one or more second model parameters, a second performance, wherein the second performance exceeds the first performance. (Chatterjee, Par. 0120:” The process 1400 provides for evaluation and iteration. After each fine-tuning stage at steps 1410, the domain specific codemix understanding model 1252 undergoes rigorous evaluation using domain specific benchmarks and performance metrics. This includes evaluating ability of the domain specific codemix understanding model 1252 to handle codemixed text that accurately reflects real-world use cases within the domain. Feedback from these evaluations is used to iteratively improve the domain specific codemix understanding model 1252, with adjustments made to the training data or model hyperparameters as necessary.”) Note: evaluation and iteration/iteratively reads on second performance exceeds the first performance.
Regarding claim 13, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, further teaches wherein the first performance of the machine translation model is determined based on an evaluation associated with a large language model (LLM). (Chatterjee, Par. 0041:” The fine-tuning module 304 may include a data storage 318, a model fine-tuning engine 320, and a data storage 322. The data storage 318 may include a pre-trained multilingual translation model 324. The pre-trained multilingual translation model 324 may be a pre-trained GenAI model, such as, but not limited to, Generative Pre-trained Transformers (GPT), Gemini, Large Language Model Meta AI (LLaMA), and the like. The model fine-tuning engine 320 may receive the cross-domain codemix parallel corpus 316 in the pre-processed format from the data pre-processing engine 306.”)
Regarding claim 18, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however Basnight further teaches wherein the information is transmitted to one or more data pipelines or an application programming interface (API) associated with integration into a client-facing service that classifies job titles into corresponding occupations based on a global taxonomy. (Basnight, Par. 0049:” … Each of the cores 404 is configured to transmit output data to the output pipeline circuits 308 of the nodes 402 operably associated with the core 404.”)
Claims 2, and 6 are rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Liu et al. (US20180032870A1)(herein "Liu").
Regarding claim 2, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however Liu teaches wherein determining the training data comprises applying a plurality of language-specific training parameters comprising a mean word length and a target-to-source ratio. (Liu, Par. 0008:” … to obtain training data via the interface circuitry, analyze the training data to obtain training measurements of the language features from the training data, use the training measurements of the language features as parameters to establish classifiers corresponding to abstract text features by machine learning, and fuse the classifiers into a decision tree to form the evaluation model.”, and Par. 0037:” …, the basic language features include at least statistical features, …, a maximum sentence length, and a ratio of a number of interrogative sentences; the word features include at least: an average word length, a category and a number of longest words, …; and the punctuation features include at least: a ratio of a number of question marks and a ratio of a number of exclamation marks.”)
Liu is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Liu to applying a plurality of language-specific training parameters comprising a mean word length and a target-to-source ratio. Motivation to do so would enhance translation accuracy and fluency by preventing under-translation or over-translation, and allows the system to estimate how many tokens are required to accurately output a translation.
Regarding claim 6, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however Liu teaches wherein the first performance is determined by testing the machine translation model with a validation dataset. (Liu, Par. 0096:” … the classification of the target variable, and the data is usually divided into training sets and test sets [validation dataset].”, and Par. 0110:”After the machine learning models are trained, the accuracy of the models further needs to be verified. In the experiment of this embodiment, the five-fold cross validation is used. Cross validation is a practical method for cutting a data sample into relatively small subsets when the data volume is not big enough. First, a subset therein is used as a training sample to train a classifier, and other subsets are used as test sets to verify indexes such as accuracy of the classifier. The five-fold cross validation is to divide the data set into five parts, and one part is selected each time as a test set, and the remaining four parts are used as training sets, and therefore five experiments are made.”)
Liu is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Liu to wherein the first performance is determined by testing the machine translation model with a validation dataset. Motivation to do so would provide an unbiased way to test a model's ability to translate unseen text.
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Shotaro et al. (JP2020140580A)(herein " Shotaro ").
Regarding claim 3, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Shotaro teaches detect variations in the training data comprising inconsistent capitalization and punctuation, wherein the first model parameters are determined based on the variations in the training data. (Shotaro, Page 8, Par. 3:” The preprocessing unit 131 includes at least one model group generated for each of a plurality of different normalization methods for the data for training, or at least one model group in which different model parameters are set for each model group. Data whose anomaly degree calculated using one model is higher than a predetermined value is excluded from the training data. “, and Page 8, Par. 7:” Here, the detection device 10 executes normalization for the training data for each variation (step S103). The variation is a normalization method, and includes min-max normalization, standardization (Z-score), robust normalization, and the like shown in FIG.”, and Page 10, Par. 6:” The preprocessing unit 131 calculates using at least one model group generated for each of a plurality of different normalization methods for the data for training, or a model group in which different model parameters are set for each model group. Data whose degree of anomaly is higher than a predetermined value is excluded from the training data. As a result, the detection device 10 can exclude data that reduces the detection accuracy.”) Note: normalization method describes stripping out punctuation, accents, and capital letters to create a consistent baseline of text.
Shotaro is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Shotaro to detect variations in the training data comprising inconsistent capitalization and punctuation, wherein the first model parameters are determined based on the variations in the training data. Motivation to do so would simplify feature representation and reduces computational overhead and learn meaningful patterns more effectively without being confused by arbitrary variations.
Claims 4, and 7 are rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Liu et al. (US20180032870A1)(herein "Liu").
Regarding claim 4, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Fatemi teaches wherein the one or more first model parameters are associated with gender normalization or cultural translation adjustments. (Fatemi, Abstract:” … New parameters are added to the language model. The new parameters may be associated with gender related terms, such as profession names. In a subsequent training phase the new parameters of the language model are trained using a gender neutral dataset.”)
Fatemi is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Fatemi to wherein the one or more first model parameters are associated with gender normalization or cultural translation adjustments. Motivation to do so would accurately deduce and translate the correct gender of non-gendered subjects, avoiding defaults to masculine pronouns for leadership or high-status roles.
Regarding claim 7, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Fatemi teaches wherein the machine translation model comprises cultural or gender normalization parameters. (Fatemi, Abstract:” … New parameters are added to the language model. The new parameters may be associated with gender related terms, such as profession names. In a subsequent training phase the new parameters of the language model are trained using a gender neutral dataset.”)
Fatemi is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Fatemi to wherein the machine translation model comprises cultural or gender normalization parameters. Motivation to do so would accurately deduce and translate the correct gender of non-gendered subjects, avoiding defaults to masculine pronouns for leadership or high-status roles.
Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Jing et al. (CN118709703A)(herein "Jing").
Regarding claim 9, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Jing teaches determine a cultural fit associated with the machine translation model, wherein the first performance is determined based on the cultural fit. (Jing, Page 11, Par. 4:”… the data quality monitoring result and other key performance indexes such as translation accuracy, fluency and cultural adaptability score, corresponding scoring standards are formulated, the standards are used for comprehensive evaluation of performance, and the translation model is adjusted to better adapt to the requirements of different cultural contexts by analyzing the correlation performance among indexes such as how the indexes influence each other, and finally, the analysis results are integrated to generate comprehensive performance index information.”)
Jing is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Jing to determine a cultural fit associated with the machine translation model, wherein the first performance is determined based on the cultural fit. Motivation to do so would allow the system to accurately adapt tone, idioms, and context-specific phrasing.
Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Michael Griffin (US20250336505A1)(herein " Griffin ").
Regarding claim 11, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Griffin teaches wherein the training data comprises labor market data. (Griffin, Par. 0079:” … The training data can be generated by analyzing and compiling historical nurse employment data and extracting nurse attribute information, working condition information, and whether the nurse resigned from healthcare system 170.”)
Griffin is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Griffin to wherein the training data comprises labor market data. Motivation to do so would optimize hiring processes, identify specific skill gaps, and forecast future talent needs more effectively.
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Pfitzmann et al. (US20210350274A1)(herein " Pfitzmann ").
Regarding claim 12, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Pfitzmann teaches wherein determining the one or more second model parameters comprises optimizing the first performance using multi-core processing to enhance graphics processing unit (GPU) utilization. (Pfitzmann, Par. 0001:” … Machine learning involves inherently complex processing of large quantities of training data. Significant processing resources, usually provided by powerful processing systems using multi-core central processing units (CPUs), often with accelerators such as graphics processing units (GPUs) and field-programmable gate arrays (FPGAs), are required for practical implementation.”, and Par. 0002:” Embodiments of the present invention may include optimizing a function of the predicted performance metrics, indicating a predicted improvement in model performance, to identify a set of augmentation actions to augment the dataset for further training of the model.”)
Pfitzmann is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Pfitzmann to optimize the first performance using multi-core processing to enhance graphics processing unit (GPU) utilization. Motivation to do so would speed up visual rendering, and increases overall computing throughput without requiring costly hardware upgrades.
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Cunnington et al. (US20100198579A1)(herein "Cunnington ").
Regarding claim 14, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Cunnington teaches wherein the second model parameters are determined based on translation accuracy, cultural fit, and terminology appropriateness in labor market data. (Cunnington, Par. 0027:” … In other words, the context of topics or presentations within the telepresence session 104 can assist in delivering accurate translations between languages.”, and Par. 0028:” … The interpreter component 102 can further include a cultural engine 202 that can identify a cultural cue related to at least one of a cultural trait, a cultural custom, a cultural tradition, or a cultural behavior. The cultural engine 202 can observe and evaluate data and/or communications within the telepresence session 104 in order to identify a cultural cue, wherein the data and/or communications can include any suitable data such as, but not limited to, meeting details (e.g., start time, end time, date, etc.), participant information (e.g., physical location, profile information, age, connectivity, employee information, employment position, etc.), audio communications, observed gestures, observed movements, video communications, graphics, text, etc.”).
Cunnington is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Cunnington to wherein the second model parameters are determined based on translation accuracy, cultural fit, and terminology appropriateness in labor market data. Motivation to do so would yield highly precise, localized, and contextually relevant matching between job seekers and employers.
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Ball et al. (US20250284815A1)(herein "Ball").
Regarding claim 15, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Ball teaches wherein the training data is determined based on an adaptive text cleaning pipeline comprising at least one of emoji removal, non-alphanumeric sequence handling, or rule-based language-specific post-processing. (Ball, Par. 0050:” Text cleaning of the text description refers to sentence cleaning functions (e.g. lowercases the sentences, removes non-alphanumeric characters and removes excess whitespace) and word cleaning functions (lowercases the words, removes non-alphanumeric characters and removes excess whitespace). Such cleaning of text data reduces the number of tokens, avoids punctuation, emojis, URLs, etc. which may confuse a natural language processing model, and reduces noise in the natural language processing process.”)
Ball is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Ball to wherein the training data is determined based on an adaptive text cleaning pipeline comprising at least one of emoji removal, non-alphanumeric sequence handling, or rule-based language-specific post-processing. Motivation to do so would enhance model accuracy through noise reduction and vocabulary standardization.
Claim 16 is rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Wray et al. (US11151673B1)(herein "Wray").
Regarding claim 16, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Wray teaches wherein the information comprises translated labor market data. (Wray, Col. 20, ll. 13-15:”… collects raw data from a data source, the raw data may be provided to an ingestion engine …”, and Col. 42, ll. 20-21:” … ingestion engine 1704 may provide the collected information to translation engine 1706.”, and Col. 51, ll. 11-15:” … the ingestion engine may be arranged to provide …, one or more industries, labor market information, or the like, to the translation engine.”, and Col. 52, ll. 46-52:” … the translation engine may be arranged to ... In some embodiments, translating free form job descriptions into unified facts, such as, unified job or position features enables position profiles (e.g., job profiles) to be generated ...”). Note: labor market information provided to a translation engine reads on translated labor market data.
Wray is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Wray to wherein the information comprises translated labor market data. Motivation to do so would enable global workforce mobility and cross-border recruitment, which allows organizations to tap into international talent pools.
Claim 17 is rejected under 35 U.S.C. 103 as being unpatentable over Chatterjee, and Basnight, and in further view of Qiuhuiet al. (CN117234586A)(herein " Qiuhui").
Regarding claim 17, Chatterjee, as modified above, teaches the one or more computing devices of claim 1.
Chatterjee, as modified above, does not teach, however, Qiuhui teaches wherein determining the second model parameters comprises generalizing translations by a deep learning model. (Qiuhui, Page 2, Par. 4:” the data set is prepared for training the model, and the deep learning technology used by the invention belongs to the category of neural machine translation in natural language processing, and a typical [generalized] neural machine translation method takes a word element vector as input, and word elements are obtained through the wording, so that the collected data needs to be preprocessed and converted into a word element sequence before being input into the model.”)
Qiuhui is considered to be analogous to the claimed invention because it is in the same field of endeavor. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Chatterjee, as modified above, further in view of Qiuhui to wherein determining the second model parameters comprises generalizing translations by a deep learning model. Motivation to do so would allow the model to accurately translate previously unseen sentences, handle complex idioms, and maintain natural phrasing.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure. Editorial Team (“Hyperparameter Tuning with GridSearchCV”) teaches in Page 1:” In almost any Machine Learning project, we train different models on the dataset and select the one with the best performance. However, there is room for improvement as we cannot say for sure that this particular model is best for the problem at hand. Hence, our aim is to improve the model in any way possible. One important factor in the performances of these models are their hyperparameters, once we set appropriate values for these hyperparameters, the performance of a model can improve significantly.”
Examiner's Note: Examiner has cited particular columns and line numbers and/or paragraph numbers in the references applied to the claims above for the convenience of the applicant. Although the specified citations are representative of the teachings of the art and are applied to specific limitations within the individual claim, other passages and figures may apply as well. It is respectfully requested from the applicant in preparing responses, to fully consider the references in entirety as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the Examiner.
In the case of amending the Claimed invention, Applicant is respectfully requested to indicate the portion(s) of the specification which dictate(s) the structure relied on for proper interpretation and also to verify and ascertain the metes and bounds of the claimed invention.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DARIOUSH AGAHI whose telephone number is (408)918-7689. The examiner can normally be reached Monday - Thursday and alternate Fridays, 7:30-4:30 PT.
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, Bhavesh Mehta can be reached on 571-272-7453. 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.
DARIOUSH AGAHI, P.E.
Primary Examiner
/DARIOUSH AGAHI/Primary Examiner, Art Unit 2656
Prosecution Insights