DETAILED ACTION
This non-final office action is responsive to application 18/387,591 as submitted 07 Nov. 2023.
Claim status is currently pending and under examination for claims 1-20 of which independent claims are 1, 13 and 20.
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 .
Information Disclosure Statement
As required by MPEP 609(c), the applicant’s submissions of the Information Disclosure Statement dated 11/07/23 is acknowledged by the examiner and the cited references have been considered in the examination of the claims now pending. As required by MPEP 609 C(2), a copy of the PTOL-1449 initialed and dated by the examiner is attached to the instant office action.
Claim Objections
Claims 13-19 are objected to because of the following minor informalities: independent claim 13 preamble recites “One or more non-transitory computer-readable storage media” however claims 14-19 depending from 13 all recite a preamble of “The non-transitory computer-readable storage medium” emphasis medium v. media. While it is understood that the medium is of media, it is recommended to maintain consistent terminology as a matter of form reciting either of medium or media, but not both. Appropriate correction and/or clarification is respectfully requested.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claim 17 is rejected under 35 U.S.C. 112(b), as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor, regards as the invention. Particularly, claim 17 recites "the determination of the first set of embeddings and the determination of the second set of embeddings" in last limitation. There is insufficient antecedent basis for this limitation in the claim which should read "the generation of the first set of embeddings and the generation of the second set of embeddings" similar as in claim 5 which will readily address the issue.
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, 13, 17 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over:
Gupta et al., “MatSciBERT: A Materials Domain Language Model for Text Mining and Information Extraction” hereinafter Gupta (arXiv: 2109.15290v1), in view of
Mukherjee et al., US PG Pub No 2023/0420085A1 hereinafter Mukherjee, in view of
Ben-Joseph, Jonathan, US PG Pub No 2025/0037807A1 hereinafter BJ, in view of
Kumar et al., “Recycle-BERT: Extracting Knowledge about Plastic Waste Recycling by Natural Language Processing” hereinafter Kumar.
With respect to claim 1, Gupta teaches:
A method, executed by a processor {Gupta [P.3] Fig 1 “methodology for training MatSciBERT” implemented [P.5 ¶2] “GPU” is processor}, comprising:
receiving a dataset that includes information associated with scientific literature {Gupta see [P.3] Fig 1 “Download research papers” download is receive, for “Materials Science Corpus”};
applying one or more neural network models on the dataset {Gupta [P.3] Fig 1 illustrates MatSciBERT where BERT model is based on neural networks such as BiLSTM [P.6-7 Sect. 2.4], applied by training MatSciBERT on the MatSci Corpus Fig 1, [P.4-5 Sect. 2.4]};
determining a set of materials and information associated with each material of the set of materials, based on the application of the one or more neural network models {Gupta [P.3] Fig 1 materials corpus with information determined from selected materials science papers and further determining comprises “Named entity recognition” (NER) and/or “classification” upon MatSciBERT model applied by training. See [P.5 Sect. 2.3.1] and [P.6-7 Sect. 2.4.1-3]. An example is shown Fig 4};
generating a first set of embeddings indicative of a first set of features of each material of the set of materials {Gupta [P.6-7 Sect. 2.4.1-3] “BERT embeddings” generated by BERT over word tokens such that features of the materials “encode the entire text” see e.g. Fig 4 or Table 8};
receiving a user input indicative of information associated with a queried material {Gupta see [P.3] Fig 1 “Query search…materials” described e.g. [P.4 ¶2] “using their sanctioned API… queries such as ‘cement’, ‘interfacial transition zone’, ‘magnesium alloy’, and ‘magnesium alloy composite materials’ to name a few”};
However, Gupta does not appear to disclose the following limitations which are met by Mukherjee:
generating a second set of embeddings associated with textual content that describes effects of the set of materials on resources of a living environment {Mukherjee Fig 3:214 described [0036] “generating the reaction embeddings 214” again at [0074]. The reaction is an effect and such effect described for a living environment is e.g. [0018] “toxic byproducts” resulting from chemical reaction, chemical reaction represented by “text” e.g. [0084,89] and queried from scientific publications or documents [0035,77], Fig 1};
training a generative artificial intelligence (AI) model based on the first set of embeddings and the second set of embeddings {Mukherjee [0091] “trained model” shown Fig 5:518 and again Fig 3 where models are encoders 308,316 with respective embeddings 214,216 in a shared embedding space the generative model may include “SciBERT” [0077,87] and training may repeat for 100 epochs [0090]};
Mukherjee is directed to generative AI models with embeddings and application to materials thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to generate second embeddings and train a generative model per Mukherjee in combination for a motivation [0038] “training maximizes the similarity in the shared embedding space” with [0026] “texts in a joint embedding space, thereby enabling such search are retrieval operations” e.g. [0018] “effective searching and retrieval system makes it possible to explore opportunities for optimizations of chemical reactions that may lower cost, improve selectivity, and decrease non-desirable byproducts” further noting [0003] “It would be of great value to chemists and others to be able to fetch textual procedures of chemical reactions based on information about the chemical reaction without the limitations of exact keyword matching.”
However, the combination Gupta and Mukherjee does not disclose the following limitations which are met by BJ:
generating a third embedding indicative of a second set of features of the queried material and a fourth embedding associated with textual content that describes effects of the queried material on the resources of the living environment {BJ discloses [0030] “generate a third text embedding… generate a fourth text embedding” are 3rd and 4th embeddings generated, [0076-77] “embeddings are vectors that capture some of the essential qualities or features of the elements the represent… embeddings of SMILES strings” represent molecular properties of chemicals, e.g. per [0054,59] “Chemical property analysis is used in… material sciences …molecules in the field of material science” and [0066] “search query may include a description of any property of the molecule, such as toxicity, permeability, solubility, or bioavailability” noting e.g. “HIV effect, an Alzheimer’s effect, a side-effect” are effects on living environments, illustrated Fig 3 embedding space. See also Figs 1, 4 and 7};
applying the generative Al model on the third embedding and the fourth embedding {BJ [030] “third text embedding…third error value” and “fourth text embedding…fourth error value” error regards model’s loss function, applied e.g. [090-92, 081], and generative AI model includes “DistilBERT” [077], Fig 4:406. A training is applied iteratively over epochs [081,106]};
BJ is directed to generative AI models with embeddings and application to materials thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to generate third and fourth embeddings for applying to generative model per BJ in combination for a motivation similar to Mukherjee at BJ [0075] similarity in combined embedding space and where [076] “embedding space 300 is a high-dimensional vector space… such as 128-dimensional vector space” the higher dimension allowing for additional information to be represented by the additional embeddings, so as to [0073] “enhance and/or enrich the query results” as well as [0089] “improve the performance of the embedding model” which [0058] “can significantly enhance the development of new drugs, materials, and devices.”
However, the combination of Gupta, Mukherjee and BJ does not disclose the following limitations which are met by Kumar:
determining sustainability information associated with the queried material based on the application of the generative Al model {Kumar Figs 1-2, 4 show Recycle-BERT, [P.12124 ¶1] “sustainable option for processing plastic waste is recycling” recycling information is sustainability information, hence Recycle-BERT in journal of sustainable chemistry, determined by the Recycle-BERT model. Figs 1-2 and 4 show queries/questions, materials are e.g. [P.12126 Sect. 3.2 ¶2] “catalyst materials used for recycling… reactant materials recycled” as well as plastics and/or polymers described throughout as generally apply to organic chemistry}; and
controlling a display device to render the sustainability information associated with the queried material {Kumar Fig 8 shows rendered information using Recycle-BERT to answer questions queries regarding the materials and provide predictions, display is requisite and includes API Figs 1-2, computer implementation via GPU control device using Python is disclosed [P.12128 ¶2]}.
Kumar is directed to generative AI models with embeddings and application to materials thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to determine sustainability/recycling information for display with Recycle-BERT per Kumar in combination to arrive at the invention as claimed for a motivation [P.12124 ¶4] “The key aim of the recycle BERT is to abstract and distill relevant information related to the recycling” and so as to “help practitioners and scientists in this area” e.g. [P.12132 ¶1] “this tool would come in handy for technologists, engineers, and scientists for summarizing and distilling relevant scientific articles in the area of waste plastic recycling literature.”
With respect to claim 5, the combination of Gupta, Mukherjee, BJ and Kumar teaches the method according to claim 1, further comprising:
applying a natural language model on each material of the set of materials and the information associated with each material of the set of materials {Gupta [P.1 ¶1] “natural language processing, such as bidirectional encoder representations from transformers (BERT)” cont’d [P.2 ¶2] “NLP and ML in the material science domain” applied as MatSciBERT shown Fig 1}, wherein
Gupta further discloses [P.7 ¶1] “BERT embeddings.” However, the second set of embeddings is disclosed by Mukherjee:
the generation of the first set of embeddings and the generation of the second set of embeddings are further based on application of the natural language model {Mukherjee Fig 3 embedding space of combined embeddings shown with encoder networks/models which may encode via [0077] “BERT-based language model of scientific publications (e.g., SciBERT)” again at [0087] or similar [0038]}. A person having ordinary skill in the art would have considered it obvious to employ BERT-based encoder per Mukherjee in combination with Gupta as obvious to try since maintains consistent model structure of a BERT-based NLP model known in the art.
With respect to claim 13, the rejection of claim 1 is incorporated. The difference in scope being a non-transitory computer-readable media storing instructions executed causing electronic device to perform operations of method claim 1. Gupta discloses [P.5 ¶2] “32GB GPUs” GB denoting memory medium for storing instructions executed by GPU processor electronic device, and “code is written using PyTorch …available at our GitHub repository.” The remainder of this claim is rejected for the same rationale as claim 1.
With respect to claim 17, the combination of Gupta, Mukherjee, BJ and Kumar teaches the non-transitory computer-readable storage medium according to claim 13, and further teaches the limitation similar to claim 5. Therefore, the rejection of claim 5 is applied to claim 17.
With respect to claim 20, the rejection of claim 1 is incorporated. The difference in scope being an electronic device comprising memory storing instructions and coupled to processor for execution to perform a process of limitations in method claim 1. Gupta discloses [P.5 ¶2] “32GB GPUs” GPU being a processor electronic device for execution instructions stored in GB denoting memory medium, and “code is written using PyTorch …available at our GitHub repository.” The remainder of this claim is rejected for the same rationale as claim 1.
Claims 2-4 and 14-16 are rejected under 35 U.S.C. 103 as being unpatentable over Gupta, Mukherjee, BJ and Kumar in view of
Sun et al., US PG Pub No 2023/0223112A1 hereinafter Sun.
With respect to claim 2, the combination of Gupta, Mukherjee, BJ and Kumar teaches the method according to claim 1, further comprising:
applying a first neural network model of the one or more neural network models on the dataset {Kumar [P.12126 Sect.4 ¶1] “BERT neural network” again at [P.12124 ¶2] “BERT employs transformer neural networks” applied to the database Figs 1-2};
identifying a set of reactants based on the application of the first neural network model {Kumar [P.12123 ¶1] “Reactant-BERT to identify the reactants” shown Fig 4(b) and described e.g. [P.12129 Sect. 5.2 ¶1], introduced [P.12124 Last¶], and example use Fig 8(3)};
However, Kumar in combination does not appear to disclose the following limitations which are disclosed by Sun:
applying a second neural network model of the one or more neural network models on the set of reactants {Sun Fig 3B “Second Neural Network” 320 inputs “reactants” 352, similar at Fig 3A:320,312 second subnetwork 320 input reactants 312, again Fig 2}; and
selecting a subset of reactants from the set of reactants based on the application of the second neural network model {Sun Fig 4:410 “Select a particular candidate set of reactants” upon 404 “second subnetwork” where [0059] “second subnetwork 220 is a self-attention based neural network …second subnetwork 220 can be a transformer neural network” the selecting described e.g. [0097,40]}, wherein
the determination of the set of materials is further based on the identification of the set of reactants and the selection of the subset of reactants {Sun discloses [0025] “determine a target compound” and [0013] “determine an optimal set of reactants” based on [0073] “identify a final set of reactants” and [0097] “select a particular candidate set of reactants” similar [0040]. The target compound e.g. chemical solvent or reagent is a material, introduced [0010] “chemically synthesize, i.e. to physically generate, the target compound”}.
Sun is directed to generative transformer neural networks for chemical reactants thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to specify second neural network and selecting reactants for determining chemical reactants per Sun in combination to arrive at the invention as claimed for a stated motivation [0012-13] “realize one or more of the following advantages… automatically perform retrosynthesis in an efficient and accurate manner to determine an optimal set of reactants to use to synthesize a target compound. The search space of possible sets of reactants for synthesizing a particular target compound can be very large, growing exponentially with the number of individual reactants… use the techniques herein to drastically reduce the time and computational cost required to determine an optimal set of reactants.”
With respect to claim 3, the combination of Gupta, Mukherjee, BJ, Kumar and Sun teaches the method according to claim 2, further comprising:
extracting information associated with each reactant of the set of reactants based on the application of the first neural network model {Kumar see [P.12129 Sect. 5.2] “Information Extraction” with Reactant-BERT, e.g. “extract more detailed information. For this purpose, the selected relevant abstracts are fed into individual Q&A BERT models (i.e., Catalyst-BERT, Method-BERT, Reactant-BERT …extract more information, such as recycling reaction conditions” and/or [P.12130] “extracted answers of catalyst, methodology, reactant”}, wherein
the information associated with each material of the set of materials includes information associated with each reactant of one or more reactants of the set of reactants included in the corresponding material {Kumar Fig 8(3) shows individual reactant information as answers extracted, e.g. [P.12129 Sect. 5.2] “extract more information, such as recycling reaction conditions” and/or [P.12130] “extracted answers of catalyst, methodology, reactant”}.
With respect to claim 4, the combination of Gupta, Mukherjee, BJ, Kumar and Sun teaches the method according to claim 3, wherein the determined information associated with each reactant of the set of reactants includes at least one of:
an organic structure of a corresponding reactant, a decay rate associated with the corresponding reactant, a biodegradability associated with the corresponding reactant, one or more catalysts facilitating a chemical reaction that involves the corresponding reactant, one or more products generated due to the chemical reaction, a temperature requirement for triggering the chemical reaction, or one or more precursors involved in the chemical reaction {Sun [0044] “determine, for each candidate set 122 of reactants, a temperature at which the candidate set 122 of reactants can synthesize the target compound… determine to discard candidate sets 122 that require temperatures that are outside of a predetermined acceptable range” see also [0045] “retrosynthesis… generate the target compound” and/or [0051] structure of chemical representations as smiles/selfies. Additionally, note Kumar’s catalyst-bert}.
With respect to claim 14, the combination of Gupta, Mukherjee, BJ and Kumar teaches the non-transitory computer-readable storage medium according to claim 13, and further combination with Sun teaches the limitation of claim 2. Therefore, the rejection of claim 2 with equal motivation is applied to claim 14.
With respect to claim 15, the combination of Gupta, Mukherjee, BJ, Kumar and Sun teaches the non-transitory computer-readable storage medium according to claim 14, and further teaches the limitation of claim 3. Therefore, the rejection of claim 3 is applied to claim 15.
With respect to claim 16, the combination of Gupta, Mukherjee, BJ, Kumar and Sun teaches the non-transitory computer-readable storage medium according to claim 15, and further teaches the limitation of claim 4. Therefore, the rejection of claim 4 is applied to claim 16.
Claims 6 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Gupta, Mukherjee, BJ and Kumar in view of
Iwayama et al., “Functional Output Regression for Machine Learning in Materials Science” hereinafter Iwayama.
With respect to claim 6, the combination of Gupta, Mukherjee, BJ and Kumar teaches the method according to claim 1. Iwayama teaches wherein
the generative AI model corresponds to a conditional generative adversarial network (GAN) model that includes a generator model and a discriminator model {Iwayama Fig 2 cGAN described [P.4840-41 Pg.Brk] “Conditional GAN… generator (G) and a binary classifier called a discriminator (D) …adversarial training process is to balance the learning progress of G and D” implemented Eq.3}.
Iwayama is directed to materials science using generative machine learning models thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to employ cGAN per Iwayama in combination to arrive at the invention as claimed as applying known techniques to known methods ready improvement to yield predictable results where [P.4841 Rt.Col] “cGANs have been intensively studied… we can take advantage of the wealth of tips and various extended works that have been accumulated in machine-learning research” and/or for a further motivation that cGAN are beneficial as they increase data diversity [P.4846 ¶5] as well as [P.4838 ¶3] “can cover various potential applications in material science.”
With respect to claim 18, the combination of Gupta, Mukherjee, BJ and Kumar teaches the non-transitory computer-readable storage medium according to claim 13, and further combination with Iwayama teaches the limitation of claim 6. Therefore, the rejection of claim 6 with equal motivation is applied to claim 18.
Claims 7-11 are rejected under 35 U.S.C. 103 as being unpatentable over Gupta, Mukherjee, BJ, Kumar and Iwayama in view of
Aarif et al., “Smart bin: Waste segregation system using deep learning-Internet of Things for sustainable smart cities” hereinafter Aarif.
With respect to claim 7, the combination of Gupta, Mukherjee, BJ, Kumar and Iwayama teaches the method according to claim 6. Aarif teaches wherein
the generator model is trained to generate an output for each material of the set of materials such that the discriminator model classifies each material of the set of materials as sustainable {Aarif Fig 3 GAN generator & discriminator for Fig 4 waste materials classified as bio- and non-biodegradable waste shown Fig 5 “Bio- and non-biodegradable waste classification” biodegradable is sustainable}.
Aarif is directed to sustainability concerning material waste using machine learning models thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to classify waste as biodegradable/sustainable with GAN as per Aarif in combination to arrive at the invention as claimed for the motivation of a smart bin for material waste that segregates type of waste for disposal, addressing a need where [P.1-2 ¶1] “Waste management has become a serious problem as different sorts of waste like bio- and non-biodegradable waste should be processed differently as the large portion of the waste item has considerably reusable or recyclable substances… segregating waste has become a primary need” and [Abst] “Our smart bin intends to provide optimized waste management of bio and non-bio-waste and help build an ecologically safe society.”
With respect to claim 8, the combination of Gupta, Mukherjee, BJ, Kumar, Iwayama and Aarif teaches the method according to claim 7, wherein
the output is generated based on at least one of: a random input, the first set of embeddings, the second set of embeddings, a generator loss received from the discriminator model based on previously generated outputs, or the information associated with each material of the set of materials {Aarif Fig 3 output of generator indicated by arrow, and at least associated with information of materials by way of images taken from camera Fig 4. See also Iwayama Fig 2 “input Z denotes a Gaussian random noise” and [P.4841 ¶1] “input (X,Z) is first transformed into an embedding vector” Eq.3 loss function}.
With respect to claim 9, the combination of Gupta, Mukherjee, BJ, Kumar, Iwayama and Aarif teaches the method according to claim 7, wherein
the discriminator model is configured to classify the material as sustainable or hazardous {Aarif Fig 3 discriminator D(x) for Fig 5 waste classification as bio- and non-biodegradable where biodegradable is sustainable and non-biodegradable is hazardous, see [P.3 Sect. 2.1]}.
With respect to claim 10, the combination of Gupta, Mukherjee, BJ, Kumar, Iwayama and Aarif teaches the method according to claim 9, wherein
the classification of the discriminator model is based on at least one of: the output generated for each material of the set of materials, the first set of embeddings, the second set of embeddings, a discriminator loss generated based on previous classification results, and the information associated with each material of the set of materials {Aarif Fig 3 discriminator arrow from generator’s output for classification of waste materials Fig 5, limitation recites ‘at least one of’ in the alternative. See also Iwayama Fig 2, Eq.3 [P.4841 ¶1-2] “classifier called a discriminator… discriminator D is learned such that the classification error is minimized”}.
With respect to claim 11, the combination of Gupta, Mukherjee, BJ, Kumar and Iwayama teaches the method according to claim 6, further comprising:
receiving, by the generator model, a first set of inputs including a random input, the third embedding, and the fourth embedding {Iwayama Fig 2 arrows into generator G is receiving “input Z denotes a Gaussian random noise” and embedding vector of embedding layer disclosed [P.4841 ¶1] such that a vector is a sequence/series of embeddings applicable to high-dimensional spaces e.g. [P.4840 ¶3] “four dimensions” or [P.4846 ¶3] “six-dimensional real vector”};
generating, by the generator model, based on the first set of inputs, an output {Iwayama Fig 2 arrows out of the generator are output based on the input indicated by arrows [] “”};
receiving, by the discriminator model, a second set of inputs including the output, the third embedding, and the fourth embedding {Iwayama Fig 2 arrows into discriminator D are inputs from the generator G output, [P.4841 ¶1]}; and
However, Iwayama does not disclose classifying as sustainable or hazardous which is met by Aarif:
classifying, by the discriminator model, the queried material as sustainable or hazardous based on the second set of inputs {Aarif Fig 3 discriminator D for classifying material waste Fig 5 into bio- and non-biodegradable waste, respectively sustainable and hazardous}. Motivation for combination is applied similarly as in claim 7.
Claims 12 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Gupta, Mukherjee, BJ and Kumar in view of
Jiang et al., “Machine Learning Based Prediction of Enzymatic Degradation of Plastics Using Encoded Protein Sequence and Effective Feature Representation” hereinafter Jiang.
With respect to claim 12, the combination of Gupta, Mukherjee, BJ and Kumar teaches the method according to claim 1. Jiang teaches wherein
the determined sustainability information associated with the queried material corresponds to a first indication that specifies whether the queried material is sustainable or hazardous and a second indication that explains a rationale behind the first indication {Jiang Fig 1 “binary classification by predicting whether the enzyme can degrade the plastic of interest” Degradable is sustainable and Nondegradable is hazardous, material is [P.562 Sect. 3.6 ¶1] “plastic type as the input query” and explained rationale is Fig 3 “SHapley Additive exPlanations (SHAP) …SHAP value, which indicates the relative importance of features on enzymatic plastic degradation” introduced [P.559 Sect. 2.2 ¶3]}.
Jiang is directed to environmental sciences and recycling using transformer ML encoders thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to perform binary classification over degradable/nondegradable as corresponding to sustainable/hazardous and provide Shapley explanatory features per Jiang to arrive at the invention as claimed for a motivation [P.559 ¶3] “we aimed to develop an innovative and effective computational approach to predict the ability of an enzyme with a known sequence to degrade a target plastic of interest with consideration of a variety of common plastics” because [P.557 ¶1] “Plastics are extensively used globally, but improper handling of plastic waste has caused severe environmental problems. Treating and recycling postconsumer plastics is critically important for environmental protection.” The binary classification may be represented by simple one-hot encoding and Shapley features quantify feature importance through known techniques to capture marginal contribution of feature information.
With respect to claim 19, the combination of Gupta, Mukherjee, BJ and Kumar teaches the non-transitory computer-readable medium according to claim 13, and further combination with Jiang teaches the limitation of claim 12. Therefore, the rejection of claim 12 with equal motivation is applied to claim 19.
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Hillebrand et al., “sustain.AI: a Recommender System to analyze Sustainability Reports” arXiv: 2305.08711v3 see Figs 1-2 screenshots and BERT
Shen et al., “Computational Design and Manufacturing of Sustainable Materials through First-Principles and Materiomics” discloses sustainability and machine learning at length
Bahrami et Srinivasan, “Examining LLM’s Awareness of the United Nations Sustainable Development Goals” co-author is instant applicant, qualifying prior art not listed in IDS.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Chase P Hinckley whose telephone number is (571)272-7935. The examiner can normally be reached M-F 9:00 - 5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Miranda M. Huang can be reached at 571-270-7092. 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.
/CHASE P. HINCKLEY/Examiner, Art Unit 2124
Prosecution Insights