DETAILED ACTION
Introduction
This office action is in response to applicant’s remarks filed 5/12/2026. Claims 1-20 are currently pending and have been examined. Applicant’s IDS have been considered. There is no claim to foreign priority.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 5/12/2026 has been entered.
Response to Arguments
Applicant's arguments filed 5/12/2026 have been fully considered but they are not fully persuasive, however in order to expedite prosecution, the Examiner has incorporated Ge et al. (Ge, US 2020/0380991), see the corresponding Ge discussion below. More specifically, applicant argues,
“Applicant respectfully disagrees with these assertions and submits that the claims have been amended to recite that a natural language utterance that containing a query is received, and a rewritten natural language utterance is generated by rewriting the natural language utterance to include one or more specific descriptors in the query.”
However, the Examiner does not concur with the applicant’s arguments. The Examiner notes the “rewriting” statement is broad enough to be interpreted in many ways. Claim 1, does not explicitly state how the query is rewritten, and in what form.
The Examiner notes, Eisner explicitly teaches,
receiving a query from a user (paragraph [0016, 0020]-his user utterance, and query)-this is not argued.
Eisner further teaches, “analyzing the query utilizing a first machine learning model to identify one or more ambiguous components of the query (paragraphs [0039]-his machine learning model and intelligent decision function assessing ambiguity in the user utterance/query).” The Examiner notes, it is the explicit query that is analyzed in the cited paragraph and further in paragraphs [0023-0026]-within the queries, the ambiguities are determined. The Examiner further notes, the applicant also discusses Eisner utilizing an intelligent search-history function with respect to concepts from the context-specific dialogue history. In the cited sections, within the utterance, ambiguous components are detected, utilizing NLP, intelligent decision machine learning model.
Therefore, the Examiner notes, it is the “rewriting the query” limitation that is the crux of the argument. The applicant’s position is that, “the program-rewriting function does not rewrite the user utterance or query.” However, the Examiner does not concur. The Examiner notes this above portion is found in the previous office action, however, the Examiner further notes that inspection of the prior art further notes, in the previously cited paragraphs [0022-0026], and Fig. 1B, it is explicitly clear, that the user utterance, is input, and there is a disambiguating concept, that identifies ambiguous entities in the query, this particular ambiguous entity is explicitly resolved “By looking up and/or rewriting concepts from context-specific dialogue history”, wherein it is clear that the clarifying entity is what is identified looked up, and used in replacement of the ambiguous entity. The Examiner notes that the code configured to find the clarifying entity may define the ambiguous entity, as the replacement or program fragment. However, in paragraph [0026], the actual concept and sub-concept are the words or phrase, such as “sushi restaurant” which is replaced specifically with “burrito restaurant”, and the rewritten query includes the new sub-concept. In order to expedite prosecution, the Examiner has incorporated Ge et al. (Ge, US 2020/0380991), which explicitly states rewriting ambiguous user utterances into a rewritten utterance that “replaces the ambiguous entity with the disambiguating entity.”
Applicant’s remaining arguments, with respect to all remaining pending claims, are based on and/or inherit the above arguments and are deemed non-persuasive as well.
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.
Claim(s) 1, 8 and 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eisner et al. (Eisner, US 2023/0367602) in view of Cho et al. (Cho, US 2023/0315766) and further in view of Ge et al. (Ge, US 2020/0380991).
As per claim 1, Eisner teaches a computer-implemented method comprising:
receiving a natural language utterance from a user (paragraph [0016, 0020]-his user utterance, and query, Fig. 1E, as a natural language utterance), wherein the natural language utterance comprises a query (ibid- example… “when is my next meeting with Tom Jones?”, as his user utterance, hereinafter the query is interpreted as the natural language utterance);
analyzing the query utilizing a first machine learning model to identify one or more ambiguous components of the query (paragraphs [0039]-his machine learning model and intelligent decision function assessing ambiguity in the user utterance/query);
determining, for each of the one or more ambiguous components of the query, a specific descriptor, utilizing the first machine learning model and a conversation history associated with the query (ibid-see above machine learning, ambiguous discussion, see also paragraphs [0023- 0026]-his ambiguities with respect to the utterance/query, and context-specific dialog history, wherein the clarifying entity is deemed the specific descriptor, from the dialogue history, which is associated with the query), wherein the conversation history includes one or more turns in a conversation between a user and a chatbot system that occur before the query is received (ibid-as defined by his dialogue, wherein the query comprising the anaphoric entity and content, uses context in a previous turn, comprising the extracted specific descriptor, to his automated assistant as the chatbot, paragraphs [0013-0016], Fig. 1E);
generating a rewritten natural language utterance by rewriting the natural language utterance to include the one or more specific descriptors [in the query as a substitute for the one or more ambiguous components of the query] (ibid, see also paragraphs [0021, 0022, 0023, 0026, 0028, 0029]-his “rewriting” from the context-specific dialog history, replacing the ambiguity, with the context-specific history clarifying entity, which is applied to the user utterance and query);
computing, utilizing an encoder model, an embedding vector for the rewritten natural language utterance (paragraphs [0017-0019]-his query to encoder machine, vector space, and answering questions, using AI, wherein Figs. 1A-1C, illustrate the “rewritten” concept, item 140, 112’, and item 130, in communication with his encoder, item 104, the vector encoded information/representation used in question answering);
retrieving a subset of textual passages from a knowledge base [utilizing the embedding vector] for the rewritten natural language utterance (ibid-see above question answering discussion, paragraph [0017, 0032, 0042, 0043]-his text response, from all data-flow of events, as stored);
determining, [utilizing a second machine learning model], an answer to the rewritten natural language utterance (ibid-his answer, as the “assistant response”, Figs. 1A, 1E, as applied to the rewritten query, hereinafter), wherein the determining comprises taking as input the rewritten natural language utterance and each of the textual passages from the subset of textual passages and extracting or generating the answer based on the rewritten natural language utterance and information within the subset of textual passages (ibid, paragraphs [0042, 0043, 0019]-his answer, from stored information, as a subset of textual passages, and generated answer); and
providing the answer to the user as a response to the query (ibid-see above response discussion).
Eisner lacks explicitly teaching, that which Cho teaches, retrieving a subset of textual passages from a knowledge base utilizing the embedding vector for the rewritten query (paragraphs [0048, 0043-0048]-his document, and corresponding document query/embeddings, and clusters);
determining, utilizing a second machine learning model, an answer to the rewritten query (ibid, see also paragraph [0015, 0048-0052]-his FAISS-AI and neural network, which provides an answer to the query).
Thus, it would have been obvious to one of ordinary skill in the linguistics art, before the effective filing date of the invention, as all the claimed elements were known in the prior art and one skilled in the art could have combined the elements as claimed by known methods (computer implemented techniques and algorithms combining processes and steps in natural language processing), in view of the teachings of Eisner and Cho to combine the prior art element of utilizing a rewriting process for a query, in order to provide and generate an answer as taught by Eisner with using a document embedding for text passages/documents, and indexing model for answering queries as taught by Cho as each element performs the same function as it does separately, as the combination would yield predictable results, KSR International Co. v. Teleflex Inc., 550 US. -- 82 USPQ2nd 1385 (2007), wherein the predictable result would be an accurate and fast retrieval of answers/responses to queries (as rewritten queries, for purposes of clarifying ambiguities, with respect to the combination with Eisner) based on the ranking, clustered and indexed responses/answers (ibid-Cho, see also paragraphs [0009, 0017]-his relatively fast and accurate retrieval of a search query response, based on semantic difference in the embeddings space).
Ge further teaches and makes explicitly clear that which Eisner refers to as discussed above, generating a rewritten natural language utterance by rewriting the natural language utterance to include the one or more specific descriptors in the query as a substitute for the one or more ambiguous components of the query (paragraph [0094]-his “rewritten utterance that replaces the ambiguous entity with the disambiguating entity”).
Thus, it would have been obvious to one of ordinary skill in the linguistics art, before the effective filing date of the invention, as all the claimed elements were known in the prior art and one skilled in the art could have combined the elements as claimed by known methods (computer implemented techniques and algorithms combining processes and steps in natural language processing), in view of the teachings of Eisner and Cho and Ge to combine the prior art element of utilizing a rewriting process for a query, in order to provide and generate an answer as taught by Eisner with using a document embedding for text passages/documents, and indexing model for answering queries as taught by Cho with generating a rewritten natural language utterance by rewriting the natural language utterance to include the one or more specific descriptors in the query as a substitute for the one or more ambiguous components of the query as taught by GE, as each element performs the same function as it does separately, as the combination would yield predictable results, KSR International Co. v. Teleflex Inc., 550 US. -- 82 USPQ2nd 1385 (2007), wherein the predictable result would be an accurate and fast retrieval of answers/responses to natural language utterances (which are rewritten natural language utterances, for purposes of clarifying ambiguities and sending to downstream processing for question answering-see Ge, with respect to the combination with Eisner) based on the ranking, clustered and indexed responses/answers (ibid-Cho, see also paragraphs [0009, 0017]-his relatively fast and accurate retrieval of a search query response, based on semantic difference in the embeddings space, ibid-Ge).
As per claim 8, claim 8 sets forth limitations similar to claim 1 and is thus rejected under similar reasons and rationale, wherein the system is deemed to embody the method, such that Eisner with Cho make obvious a system comprising:
one or more processors (Eisner, paragraphs [0121]-see his processor, instructions, software on storage devices, and execution discussion); and
one or more computer-readable media storing instructions which, when executed by the one or more processors, cause the system to perform operations comprising (ibid):
receiving a natural language utterance from a user, wherein the natural language utterance comprises a query (ibid-see claim 1, corresponding and similar limitation); analyzing the query utilizing a first machine learning model to identify one or more ambiguous components of the query (ibid); determining, for each of the one or more ambiguous components of the query, a specific descriptor, utilizing the first machine learning model and a conversation history associated with the query, wherein the conversation history includes one or more turns in a conversation between a user and a chatbot system that occur before the query is received (ibid); generating a rewritten natural language utterance by rewriting the natural language utterance to include the one or more specific descriptors in the query as a substitute for the one or more ambiguous components of the query (ibid); computing, utilizing an encoder model, an embedding vector for the rewritten natural language utterance (ibid); retrieving a subset of textual passages from a knowledge base utilizing the embedding vector for the rewritten natural language utterance (ibid); determining, utilizing a second machine learning model, an answer to the rewritten natural language utterance, wherein the determining comprises taking as input the rewritten natural language utterance and each of the textual passages from the subset of textual passages and extracting or generating the answer based on the rewritten natural language utterance and information within the subset of textual passages; and providing the answer to the user as a response to the query (ibid).
As per claim 15, claim 15 sets forth limitations similar to claim 1 and is thus rejected under similar reasons and rationale, wherein one or more non-transitory computer-readable media storing instructions is deemed to embody the method, such that Eisner with Cho make obvious one or more non-transitory computer-readable media storing instructions which, when executed by one or more processors, cause a system to perform operations comprising (Eisner, paragraphs [0121]-see his processor, instructions, software on storage devices, and execution discussion):
receiving a natural language utterance from a user, wherein the natural language utterance comprises a query (ibid-see claim 1, corresponding and similar limitation); analyzing the query utilizing a first machine learning model to identify one or more ambiguous components of the query (ibid); determining, for each of the one or more ambiguous components of the query, a specific descriptor, utilizing the first machine learning model and a conversation history associated with the query, wherein the conversation history includes one or more turns in a conversation between a user and a chatbot system that occur before the query is received (ibid); generating a rewritten natural language utterance by rewriting the natural language utterance to include the one or more specific descriptors in the query as a substitute for the one or more ambiguous components of the query(ibid); computing, utilizing an encoder model, an embedding vector for the rewritten natural language utterance (ibid); retrieving a subset of textual passages from a knowledge base utilizing the embedding vector for the rewritten natural language utterance (ibid); determining, utilizing a second machine learning model, an answer to the rewritten natural language utterance, wherein the determining comprises taking as input the rewritten natural language utterance and each of the textual passages from the subset of textual passages and extracting or generating the answer based on the rewritten natural language utterance and information within the subset of textual passages (ibid); and providing the answer to the user as a response to the query (ibid).
Claim(s) 2, 9 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Eisner et al. (Eisner, US 2023/0367602) in view of Cho et al. (Cho, US 2023/0315766) in view of Ge, as applied to claim 1, and further in view of Morrill et al. (Morrill, US 2022/0374405).
As per claims 2, 9 and 16, Eisner with Cho with Ge make obvious the computer-implemented method of claim 1, further comprising:
[converting a plurality of documents in a variety of document formats into a plurality of text documents];
dividing each of the plurality of text documents into textual passages (ibid-Cho, paragraph [0006, 0007]-his document database, comprising document queries, and document responses, as his textual passages);
encoding, utilizing the encoder model or a different encoder model, semantics of each of the textual passages (ibid-his embedding space, mapping and semantic model, for all the textual passages), wherein the encoding comprises taking as input each of the textual passages and computing an embedding vector for each of the textual passages (ibid); and
indexing and storing the textual passages in a data store to generate the knowledge base, wherein the textual passages are indexed in accordance with the embedding vectors (ibid-see also, paragraph [0015]-his indexing).
Eisner with Cho with Ge lack teaching that which Morrill teaches, converting a plurality of documents in a variety of document formats into a plurality of text documents (paragraph [0031, 0034, 0036, 0047]-his plurality of different document formats, all converted into textual documents, i.e. his JSON, PDF, or the like, and converted into text documents, and also, “partitioning the text” into segments of texts, which is also interpreted as dividing a plurality of documents into text passages).
Thus, it would have been obvious to one of ordinary skill in the linguistics art, before the effective filing date of the invention, as all the claimed elements were known in the prior art and one skilled in the art could have combined the elements as claimed by known methods (computer implemented techniques and algorithms combining processes and steps in natural language processing), in view of the teachings of Eisner and Cho and Morrill to combine the prior art element of utilizing a rewriting process for a query, in order to provide and generate an answer as taught by Eisner with using a document embedding for text passages/documents, and indexing model for answering queries as taught by Cho with converting multiple document formats into text documents as taught by Morrill as each element performs the same function as it does separately, as the combination would yield predictable results, KSR International Co. v. Teleflex Inc., 550 US. -- 82 USPQ2nd 1385 (2007), wherein the predictable result would be an accurate and fast retrieval of answers/responses to queries (as rewritten queries, for purposes of clarifying ambiguities, with respect to the combination with Eisner) based on the ranking, clustered and indexed responses/answers (ibid-Cho, see also paragraphs [0009, 0017]-his relatively fast and accurate retrieval of a search query response, based on semantic difference in the embeddings space), the documents, which come in a plurality of forms, converted in a form usable downstream applications (ibid-Morrill, abstract).
Claim(s) 3, 5, 6, 10, 12, 13, 17, 19 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Eisner et al. (Eisner, US 2023/0367602) in view of Cho et al. (Cho, US 2023/0315766) in view of Ge, as applied to claim 1, and further in view of Zhuo et al. (Zhuo, US 2021/0357441).
As per claims 3, 10 and 17, Eisner with Cho with Ge make obvious the computer-implemented method of claim 1, but lack the method further comprising, that which Zhuo teaches:
evaluating, utilizing a cross-encoder model, how well each of the textual passages from the subset of textual passages answer the query, wherein the evaluating comprises taking as input each of the textual passages from the subset of textual passages and computing a score for each of the textual passages from the subset of textual passages that is indicative of answerability (Zhuo, paragraph [0049-0051, 0109]-his BERT model, and corresponding natural language query, by voice, and answer candidates, cross-encoding of both, and corresponding scoring, as the textual passages comprising answers/answerability, Fig. 3);
ranking the textual passages from the subset of textual passages based on the score computed for each of the textual passages from the subset of textual passages (ibid-his ranking of the answer candidates, based on the scoring); and
grouping some of the textual passages from the subset of textual passages into a revised subset of textual passages based on the ranking and a predetermined answerability threshold (ibid-his scoring threshold for candidates, based on ranking, and ranking list changed, as the grouping, based on fine-tuning),
wherein the determining the answer to the rewritten natural language utterance comprises taking as input the rewritten natural language utterance and each of the textual passages from the revised subset of textual passages and extracting or generating the answer based on the rewritten natural language utterance and information within the revised subset of the textual passages (ibid, paragraphs [0060-0071]-his, rewritten queries, and corresponding response candidate having the highest score, is provided to the chatbot, from the group of ranked candidates, and then, based on the input query, the final answer is provided).
Thus, it would have been obvious to one of ordinary skill in the linguistics art, before the effective filing date of the invention, as all the claimed elements were known in the prior art and one skilled in the art could have combined the elements as claimed by known methods (computer implemented techniques and algorithms combining processes and steps in natural language processing), in view of the teachings of Eisner and Cho and Zhuo to combine the prior art element of utilizing a rewriting process for a query, in order to provide and generate an answer as taught by Eisner with using a document embedding for text passages/documents, and indexing model for answering queries as taught by Cho with using an evaluation of the passages for determining answerability and ranking, in order to generate a final answer as taught by Zhuo, as each element performs the same function as it does separately, as the combination would yield predictable results, KSR International Co. v. Teleflex Inc., 550 US. -- 82 USPQ2nd 1385 (2007), wherein the predictable result would be an accurate and fast retrieval of answers/responses to queries (as rewritten queries, for purposes of clarifying ambiguities, with respect to the combination with Eisner) based on the ranking, as evaluated for answerability, and clustered and indexed responses/answers (ibid-Cho, see also paragraphs [0009, 0017]-his relatively fast and accurate retrieval of a search query response, based on semantic difference in the embeddings space, ibid-Zhuo).
As per claims 5, 12 and 19, Eisner with Cho with Ge make obvious the computer-implemented method of claim 1, Zhuo teaching that which the above combination lacks, wherein retrieving the subset of textual passages from the knowledge base comprises:
comparing the embedding vector for the rewritten natural language utterance to embedding vectors computed for textual passages within the knowledge base (ibid-see Zhuo, embedding vector for rewritten query and vector for document, paragraphs [0060-0071, 0076]-his, rewritten queries, and corresponding response candidate textual passages, from a database of documents); and
retrieving each of the textual passages for the revised subset of textual passages in response to determining that a semantic distance between the embedding vector for each of the textual passages and the embedding vector for the rewritten natural language utterance is less than a predetermined threshold amount (ibid-his semantic comparisons, “most semantically related” multi-dimensional vectors for distance comparison, and corresponding retrieved candidates, see claim 3, ranking and reranking, revised candidates, as the retrieved textual passages, the combination of references similarly motivated for combination, as seen in claim 3, Zhuo determining a response, based on candidate ranked passages, from documents, see also abstract).
As per claims 6, 13 and 20, Eisner with Cho with Ge make obvious the computer-implemented method of claim 1, but lack teaching that which Zhuo teaches, wherein retrieving the subset of textual passages from the knowledge base comprises:
performing a [k-nearest-neighbor] search to make classifications or predictions about groupings of textual passages within the knowledge base (ibid-see Zhuo, embedding vector for rewritten query and vector for document, paragraphs [0060-0071, 0076]-his search and corresponding response candidate textual passages, see his “nearest neighbor” discussion); and
retrieving each of the textual passages for the revised subset of textual passages in response to determining that the embedding vector for each of the textual passages and the embedding vector for the rewritten natural language utterance are classified or predicted to pertain to a same grouping of textual passages within the knowledge base (ibid-Zhuo, see retrieved semantically similar candidates as the passages, and claim 3, “revised subset” discussion, based on classification and prediction, as ranked).
Eisner teaches, using a k-nearest-neighbor search to make classifications or predictions (paragraph [0123]-his “nearest-neighbor algorithm”).
Thus, it would have been obvious to one of ordinary skill in the linguistics art, before the effective filing date of the invention, as all the claimed elements were known in the prior art and one skilled in the art could have combined the elements as claimed by known methods (computer implemented techniques and algorithms combining processes and steps in natural language processing), in view of the teachings of Eisner and Cho and Zhuo to combine the prior art element of utilizing a rewriting process for a query, in order to provide and generate an answer, and nearest-neighbor search for clustering methods, as taught by Eisner with using a document embedding for text passages/documents, and indexing model for answering queries as taught by Cho with using a selection of relevant passages, as taught by Zhuo, each element performs the same function as it does separately, as the combination would yield predictable results, KSR International Co. v. Teleflex Inc., 550 US. -- 82 USPQ2nd 1385 (2007), wherein the predictable result would be an accurate and fast retrieval of answers/responses to queries (as rewritten queries, for purposes of clarifying ambiguities, with respect to the combination with Eisner) based on the ranking, as evaluated for answerability, and clustered (ibid, see Eisner algorithm for clustering) and indexed responses/answers (ibid-Cho, see also paragraphs [0009, 0017]-his relatively fast and accurate retrieval of a search query response, based on semantic difference in the embeddings space, ibid-Zhuo).
Claim(s) 4, 11 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eisner et al. (Eisner, US 2023/0367602) in view of Cho et al. (Cho, US 2023/0315766) in view of Ge, in view of Zhuo, as applied to claim 3, and further in view of Croutwater et al. (Croutwater, US 2021/0019375).
As per claims 4, 11 and 18, Eisner with Cho with Ge with Zhuo make obvious the computer-implemented method of claim 3, but lack further comprising, that which Trainor teaches, routing the query or a subsequent utterance from the user in the conversation between the user and the chatbot system to one or more skills within the chatbot system based on the score computed for each of the textual passages from the revised subset of textual passages (Croutwater, paragraphs [0054, 0055, 0062]-his routing user requests to skill bot, based on the relevancy, as scored, documents database as comprising relevancy scored passages).
Thus, it would have been obvious to one of ordinary skill in the linguistics art, before the effective filing date of the invention, as all the claimed elements were known in the prior art and one skilled in the art could have combined the elements as claimed by known methods (computer implemented techniques and algorithms combining processes and steps in natural language processing), in view of the teachings of Eisner and Cho and Ge and Zhuo and Croutwater to combine the prior art element of utilizing a rewriting process for a query, in order to provide and generate an answer as taught by Eisner with using a document embedding for text passages/documents, and indexing model for answering queries as taught by Cho with using an evaluation of the passages for determining answerability and ranking, in order to generate a final answer as taught by Zhuo with routing the query to a chatbot system to one or more skills based on relevancy of answers found in documents, as taught by Croutwater, as each element performs the same function as it does separately, as the combination would yield predictable results, KSR International Co. v. Teleflex Inc., 550 US. -- 82 USPQ2nd 1385 (2007), wherein the predictable result would be an accurate and fast retrieval of answers/responses to queries (as rewritten queries, for purposes of clarifying ambiguities, with respect to the combination with Eisner) based on the ranking, as evaluated for answerability, and clustered and indexed responses/answers (ibid-Cho, see also paragraphs [0009, 0017]-his relatively fast and accurate retrieval of a search query response, based on semantic difference in the embeddings space, ibid-Zhuo), the query routed to a chatbot based on the chatbot skills with respect to an answerability of the text passages (ibid, Croutwater-paragraph [0054, 0055]).
Claim(s) 7 and 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eisner et al. (Eisner, US 2023/0367602) in view of Cho et al. (Cho, US 2023/0315766) in view of Ge, as applied to claim 1, and further in view of Werner et al. (Werner, US 2021/0406479).
As per claims 7 and 14, Eisner with Cho with Ge make obvious the computer-implemented method of claim 1, but lack further comprising, that which Werner teaches:
providing the answer to the user as the response to the query in response to determining that the chatbot system cannot answer the query using another method (paragraph [0039]-his answer to a query, based on a different method not providing a correct answer);
providing the answer to the user as the response to the query in addition to another answer generated by the chatbot system using another method (paragraphs [0037, 0038]-his multiple models, providing an answer to the query); or
providing the answer to the user as the response to the query instead of another answer generated by the chatbot system using another method in response to determining that a confidence score calculated for the one or more answers exceeds a predetermined threshold (paragraph [0039]-his answer confidence score, from a model, different from another model, wherein the threshold of confidence determines the model and corresponding answer to be presented).
Thus, it would have been obvious to one of ordinary skill in the linguistics art, before the effective filing date of the invention, as all the claimed elements were known in the prior art and one skilled in the art could have combined the elements as claimed by known methods (computer implemented techniques and algorithms combining processes and steps in natural language processing), in view of the teachings of Eisner and Cho and Ge and Werner to combine the prior art element of utilizing a rewriting process for a query, in order to provide and generate an answer as taught by Eisner with using a document embedding for text passages/documents, and indexing model for answering queries as taught by Cho with providing an answer selection based on multiple models, as taught by Werner, as each element performs the same function as it does separately, as the combination would yield predictable results, KSR International Co. v. Teleflex Inc., 550 US. -- 82 USPQ2nd 1385 (2007), wherein the predictable result would be an accurate and fast retrieval of answers/responses to queries (as rewritten queries, for purposes of clarifying ambiguities, with respect to the combination with Eisner, ibid-Cho), wherein, answer options comprise, an answer model’s capability to answer, confidence in answer, or plurality of answers from different models (ibid-Werner, see also abstract).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LAMONT M SPOONER whose telephone number is (571)272-7613. The examiner can normally be reached 8:00 AM -5:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Daniel Washburn can be reached at (571)272-5551. 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.
/LAMONT M SPOONER/ Primary Examiner, Art Unit 2657
8/7/2026
Prosecution Insights