DETAILED ACTION
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
The information disclosure statement (IDS) submitted on 2024/03/13. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1–4 and 6–11 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Claim 5 is rejected under 35 U.S.C. 101 because it is directed to non-statutory subject matter.
Regarding independent claims 1 and 4
Step 1 — whether the claim falls within any statutory category. See MPEP 2106.03.
Claim 1 is drawn to a device (machine) claim, and claim 4 is drawn to a method (process) claim. Therefore, each of these claims falls under one of the four categories of statutory subject matter (process/method, machine/product/apparatus, manufacture, or composition of matter).
Step 2A Prong One — whether the claim recites a judicial exception. See MPEP 2106.04, subsection II.
Regarding independent claim 1, the claim is directed to a question and answer device that generates answer data in response to question data. The claim recites the limitations of "receives a predetermined question type into which a feature of the question data is classified and passage data that matches the question data," "determines whether machine reading is necessary on a basis of the question type," "sets the passage data as the answer data when determining that the machine reading is not necessary," and "sets answer text data extracted from the passage data by the machine reading on a basis of the question data and the passage data when determining that the machine reading is necessary."
These limitations, under their broadest reasonable interpretation, cover performance of the recited steps in the human mind, or by a human using pen and paper. A person can mentally receive a question together with its classified type and a matching passage, evaluate and judge whether reading the passage is necessary in view of the question type, and either designate the passage itself as the answer or read the passage and write down the extracted answer text. Such concepts of observation, evaluation, judgment, and opinion are directed towards the abstract idea of a mental process, or a concept that can be performed in the human mind (see MPEP § 2106.04(a)(2), subsection III).
Independent claim 4 is a method claim reciting limitations commensurate in scope with claim 1 and is directed towards the abstract idea for similar reasons.
Step 2A Prong Two — whether the claim as a whole integrates the recited judicial exception into a practical application of the exception. This evaluation is performed by (1) identifying whether there are any additional elements recited in the claim beyond the judicial exception, and (2) evaluating those additional elements individually and in combination to determine whether the claim as a whole integrates the exception into a practical application. See MPEP 2106.04(d).
Regarding independent claim 1, this claim recites the additional elements of "a question and answer device," "question determination circuitry," and "machine reading." These additional elements are recited at a high level of generality and amount to no more than generic computer components that merely act as a tool to apply the abstract idea (see MPEP § 2106.05(f)) and generally link the use of the judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). The recited "machine reading" amounts to mere instructions to implement, on a computer, the abstract idea of reading a passage and determining an answer, and the claim recites the result of determining the answer rather than a specific technical means that improves the functioning of a computer or any other technology (see MPEP § 2106.05(a), (f)). Accordingly, the additional elements, individually and in combination, do not integrate the abstract idea into a practical application.
Regarding independent claim 4, this claim is drawn to a method reciting similar limitations of claim 1 and is rejected under the same rationale, the additional element being a generic computer-implemented method that merely acts as a tool on which the abstract idea operates.
Step 2B — whether the claim provides an inventive concept, i.e., whether the additional elements amount to significantly more than the judicial exception. See MPEP 2106.05.
Regarding independent claims 1 and 4, the additional elements identified above, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea. The "question and answer device," "question determination circuitry," "machine reading," and the generic computer-implemented method perform only the well-understood, routine, and conventional computer functions of receiving data, processing and analyzing data, and outputting data (see MPEP § 2106.05(d)). The recitation of these generic components and functions does not add an inventive concept. Therefore, claims 1 and 4 are directed to an abstract idea without significantly more and are rejected under 35 U.S.C. § 101.
Regarding claim 5
Step 1 — whether the claim falls within any statutory category. See MPEP 2106.03.
Claim 5 recites "A program for causing a computer to function as the question and answer device according to claim 1." A claim drawn to a "program" per se, which is not recited as being embodied on a non-transitory computer-readable medium, constitutes software per se and does not fall within any of the four statutory categories of 35 U.S.C. 101 (process, machine, manufacture, or composition of matter). See MPEP § 2106.03(I). Although the specification describes that the program may be stored on a storage medium or provided through a network (specification, [0086]–[0088]), the claim itself does not recite a non-transitory computer-readable medium, and the term "program," under its broadest reasonable interpretation, encompasses non-functional descriptive material and/or a transitory propagating signal. Therefore, claim 5 is directed to non-statutory subject matter and is rejected under 35 U.S.C. § 101. Furthermore, to the extent the recited "program" is interpreted as being embodied on a non-transitory medium, claim 5 would remain directed to the same abstract idea as claim 1, set forth above, without significantly more.
Regarding dependent claims 2, 3, 6, 7, 8, 9, 10, and 11
Step 1 — whether the claim falls within any statutory category. See MPEP 2106.03.
Claims 2, 3, 6, 7, 8, and 9 are drawn to device claims, and claims 10 and 11 are drawn to method claims. Therefore, each of these claims falls under one of the four categories of statutory subject matter.
Step 2A Prong One — whether the claim recites a judicial exception. See MPEP 2106.04, subsection II.
Regarding claim 2, this claim recites the limitations of "question classification circuitry that classifies the feature of the question data into the question type," "passage search circuitry that searches for the passage data," and "extraction/reading circuitry that extracts the answer text data from the passage data by the machine reading, on a basis of the question data and the passage data." Classifying a question by its type, searching for text that matches the question, and reading a passage to extract the answer are acts of observation, evaluation, and judgment that can be performed in the human mind or with pen and paper, and are directed towards the abstract idea of a mental process.
Regarding claim 3, this claim recites the limitation of "determines whether the number of characters in the passage data is smaller than a threshold before determining whether the machine reading is necessary, and, when the number of characters is smaller than the threshold, sets the passage data as the answer data." Counting the number of characters in a passage, comparing that number to a threshold, and deciding on that basis to designate the passage as the answer are directed towards the abstract idea of a mental process (observation, evaluation, and judgment) and a mathematical concept.
Regarding claim 6, this claim recites a limitation commensurate in scope with the limitation of claim 3 and is directed towards the abstract idea of a mental process for the same reasons.
Regarding claim 7, this claim recites the limitation of "the question determination circuitry determines machine reading is necessary based on a predetermined question type acquired from the question classification circuitry." Determining whether reading is necessary on the basis of the question type is an act of evaluation and judgment that can be performed in the human mind, and is directed towards the abstract idea of a mental process.
Regarding claim 8, this claim recites the limitation of "a generative machine reading is used when the answer data exceeds the range of passage data." Determining that the answer exceeds the passage and producing the answer accordingly is an act of evaluation, judgment, and composition that can be performed in the human mind or with pen and paper, and merely narrows the abstract idea of a mental process.
Regarding claim 9, this claim recites the limitation of "the question data used as the input comprises a plurality of pieces of question data," which merely specifies that the abstract idea is applied to more than one piece of input question data and does not alter the mental-process character of the claim.
Regarding claims 10 and 11, these claims recite limitations commensurate in scope with claims 8 and 9, respectively, and are directed towards the abstract idea of a mental process for the same reasons.
Step 2A Prong Two — whether the claim as a whole integrates the recited judicial exception into a practical application. See MPEP 2106.04(d).
Regarding claims 2, 3, 6, 7, 8, 10, and 11, these claims recite the additional elements of generic "circuitry," "machine reading," "generative machine reading," and a generic computer-implemented method. These additional elements amount to no more than generic computer components that merely act as a tool to apply the abstract idea (see MPEP § 2106.05(f)) and generally link the use of the judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)), and thus fail to integrate the exception into a practical application. Regarding claims 9 and 11, the recited "plurality of pieces of question data" further constitutes insignificant extra-solution activity in the nature of data gathering (see MPEP § 2106.05(g)), and likewise fails to integrate the exception into a practical application.
Step 2B — whether the claim provides an inventive concept. See MPEP 2106.05.
Claims 2, 3, and 6–11 merely narrow the previously cited abstract idea and recite the same generic computer components and functions discussed above. For the reasons described with respect to independent claims 1 and 4, the additional elements of these claims, considered individually and as an ordered combination, perform only well-understood, routine, and conventional computer functions (see MPEP § 2106.05(d)) and do not amount to significantly more than the abstract idea. Therefore, claims 2, 3, and 6–11 are directed to an abstract idea without significantly more and are rejected under 35 U.S.C. § 101.
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 text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
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 1-7, 9, and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Kim et al. (Kim), US 11,531,818 B2, in view of Pires (Pires), Non-Patent Literature, "A Common Evaluation Setting for Just.Ask, Open Ephyra and Aranea QA systems," published May 8, 2012, pp. 2, 4, and 10, and further in view of Gupta et al. (Gupta), US 11,409,748 B1.
Regarding claim 1, Kim teaches a question and answer device that generates answer data in response to question data, the question and answer device comprising:
"A question and answer device that generates answer data in response to question data, the question and answer device comprising: question determination circuitry"
Kim teaches a machine reading comprehension (MRC) question and answer providing device that receives a user question and provides an answer to the user (Kim, Abstract; col. 1, lines 8–10; col. 3, lines 31–37; Fig. 2), the device being implemented by processing circuitry including a processor (Kim, col. 4, lines 11–26; col. 11, lines 43–67) and comprising an analyzer 220, a screener 250, and an answer unit 260 that together determine and provide the answer (Kim, col. 3, lines 38–59; Fig. 2).
"passage data that matches the question data"
This sub-limitation is taught by Kim. Kim teaches a passage searcher that selects at least one document from at least one domain corresponding to the analyzed user question and searches that document for a passage that is a candidate answer determined as being suitable for the user question (Kim, col. 3, lines 40–44; col. 6, lines 11–31; Fig. 5), thereby obtaining passage data that matches the question data.
"sets answer text data extracted from the passage data by the machine reading on a basis of the question data and the passage data when determining that the machine reading is necessary"
This limitation is taught by Kim. Kim teaches an MRC question and answer unit that applies an MRC algorithm which comprehends the user question and automatically reads and derives an answer from the passages (Kim, col. 7, lines 43–56), the unit receiving the user question and a corresponding passage as inputs and obtaining a correct answer candidate value therefrom (Kim, col. 3, lines 48–54; col. 7, lines 37–43), such that the extracted answer text data is set as the answer on the basis of the question data and the passage data.
Kim teaches subject matter related to the question-type limitations of claim 1. Specifically, Kim teaches analyzing the user question by a morphological analysis, recognizing an entity name, and mapping at least one domain corresponding to the query based on a rule-based domain classifier (Kim, col. 5, lines 30–58), and Kim further teaches determining whether to perform the MRC algorithm based on whether a stored best answer value corresponding to a similar question already exists (Kim, col. 6, line 65 – col. 7, line 5; col. 13, lines 55–67). However, Kim does not teach:
"receives a predetermined question type into which a feature of the question data is classified"
"determines whether machine reading is necessary on a basis of the question type"
In the same field of endeavor, Pires teaches "receives a predetermined question type into which a feature of the question data is classified." Pires teaches a Question Classification module that classifies the question by its expected answer type, a feature derived from the question (Pires, p. 2, § 3.1, Question Interpretation).
Pires further teaches "determines whether machine reading is necessary on a basis of the question type." Pires teaches that, based on the question category, the system applies different answer-production strategies: factoid-type questions are answered by extracting an answer from retrieved sources, whereas non-factoid-type questions are answered from a returned source passage (Pires, p. 4, § 3.2, Passage Retrieval; p. 10, § 5.3, Answer Extraction), thereby determining, on the basis of the question type, whether the answer is to be obtained by reading/extraction or by returning the passage.
Kim and Pires are analogous to the claimed invention as both are from the same field of endeavor of automated question answering that produces an answer to a user question from retrieved text. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the determination of whether to perform machine reading in the device of Kim so that the determination is made on the basis of the question type, as taught by Pires. The motivation to combine Kim and Pires is that Kim seeks to provide the best answer most suitable for the user question (Kim, col. 1, lines 40–42) and to improve searching speed and efficiency by limiting unnecessary processing (Kim, col. 5, lines 52–58; col. 7, lines 8–13), and Pires provides a known technique of selecting the answer-production strategy on the basis of the question type (Pires, p. 4, § 3.2), such that applying Pires's type-based determination to Kim would apply the appropriate answer-production strategy to each question type and avoid expending machine-reading computation where the question type does not require it, yielding the predictable result of an accurate answer with reduced computation (see KSR; MPEP § 2143(A), (C), (G)).
The combination of Kim and Pires does not teach:
"sets the passage data as the answer data when determining that the machine reading is not necessary"
In the same field of endeavor, Gupta teaches "sets the passage data as the answer data when determining that the machine reading is not necessary." Gupta teaches, in response to determining that a query is a question query, generating candidate answer passages and selecting one answer passage from the candidate answer passages, and providing that selected passage directly as the answer in an answer box presented to the user (Gupta, Abstract; col. 3, lines 36–60; Fig. 2). Gupta further teaches that some question queries are better served by returning an explanatory passage as the answer (Gupta, col. 10, lines 30–37), such that the passage data is set as the answer data when extraction by machine reading is not used.
Kim, Pires, and Gupta are analogous to the claimed invention as all three are from the same field of endeavor of automated question answering that produces an answer to a user question from retrieved text. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to further modify the combined device of Kim and Pires so that, when it is determined on the basis of the question type that machine reading is not necessary, the retrieved passage data is set as the answer data, as taught by Gupta. The motivation to combine Kim, Pires, and Gupta is that Gupta teaches that returning a retrieved passage directly as the answer provides a directly responsive answer to the user (Gupta, col. 3, lines 36–60), and combining this with the type-based determination of Kim and Pires would deliver a responsive answer for question types not requiring extraction while reducing computation, which is a combination of prior-art elements according to their established functions to yield predictable results (see KSR; MPEP § 2143(A)).
Regarding Claim 2, Claim 2 depends from claim 1 and is rejected under the same rationale as claim 1. Regarding the further limitations of claim 2, Kim teaches:
"passage search circuitry that searches for the passage data"
This limitation is taught by Kim. Kim teaches a passage searcher that selects at least one document from at least one domain corresponding to the analyzed user question and searches that document for a passage that is a candidate answer determined as being suitable for the user question (Kim, col. 3, lines 40–44; col. 6, lines 11–31; Figs. 2 and 3).
"extraction/reading circuitry that extracts the answer text data from the passage data by the machine reading, on a basis of the question data and the passage data"
This limitation is taught by Kim. Kim teaches an MRC question and answer unit that applies an MRC algorithm to the retrieved passages, the algorithm comprehending the user question and automatically reading and deriving an answer from the passages on the basis of the user question and the passage (Kim, col. 3, lines 48–54; col. 7, lines 37–56; Figs. 2, 3, and 4).
Kim teaches subject matter related to classifying the feature of the question data, in that Kim teaches an analyzer that classifies a feature of the question data into a domain based on a rule-based domain classifier (Kim, col. 5, lines 30–58). However, Kim does not teach:
"question classification circuitry that classifies the feature of the question data into the question type"
In the same field of endeavor, Pires teaches "question classification circuitry that classifies the feature of the question data into the question type." Pires teaches a Question Classification module that classifies the question, based on a feature derived from the question, into its expected answer type (Pires, p. 2, § 3.1, Question Interpretation).
Kim and Pires are analogous to the claimed invention as both are from the same field of endeavor of automated question answering that produces an answer to a user question from retrieved text. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to provide the device of Kim with question classification circuitry that classifies the feature of the question data into the question type, as taught by Pires. The motivation to combine Kim and Pires is that Kim seeks to provide the best answer most suitable for the user question (Kim, col. 1, lines 40–42) and Pires provides a known technique of classifying the question into its expected answer type (Pires, p. 2, § 3.1), such that classifying the question into the question type in Kim would enable the answer-production strategy to be selected on the basis of the question type, yielding the predictable result of an accurate answer with reduced computation (see KSR; MPEP § 2143(A), (C), (G)).
Regarding Claim 3, Claim 3 depends from claim 1 and is rejected under the same rationale as claim 1. Regarding the further limitations of claim 3, Kim teaches:
"determines whether the number of characters in the passage data is smaller than a threshold before determining whether the machine reading is necessary"
This sub-limitation is taught by Kim. Kim teaches setting an extraction range for the passage in units that count characters, for example a range from 500 bytes to 800 bytes including characters and spaces, and extracting the passage based on whether the data size of each passage meets a predetermined byte value that is automatically adjusted to an optimal byte value (Kim, col. 7, lines 14–37; col. 14, lines 36–42, claim 5).
However, Kim does not teach "when the number of characters is smaller than the threshold, sets the passage data as the answer data."
In the same field of endeavor, Gupta teaches "when the number of characters is smaller than the threshold, sets the passage data as the answer data." Gupta teaches determining a passage coverage ratio that represents a count of the characters in the candidate answer passage and determining whether that measure is smaller than a threshold value (Gupta, col. 11, lines 15–35; Fig. 8, operations 802–804), and selecting an answer passage from the candidate answer passages and providing that selected passage directly as the answer in an answer box presented to the user (Gupta, Abstract; col. 3, lines 36–60; Fig. 2), such that the passage data is set as the answer data.
Kim and Gupta are analogous to the claimed invention as both are from the same field of endeavor of automated question answering that produces an answer to a user question from retrieved text. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to configure the device of Kim so that, when the number of characters in the passage data is smaller than the threshold, the passage data is set as the answer data, as taught by Gupta. The motivation to combine Kim and Gupta is that returning a sufficiently short retrieved passage directly as the answer (Gupta, col. 3, lines 36–60) avoids expending machine-reading computation on a passage that is already a concise answer, consistent with Kim's stated aim of improving searching speed and efficiency (Kim, col. 7, lines 8–13), yielding the predictable result of a responsive answer with reduced computation (see KSR; MPEP § 2143(A), (G)).
Regarding Claim 4, Claim 4 recites limitations commensurate in scope with those of claim 1 and is rejected under the same rationale as claim 1.
Regarding Claim 5, Claim 5 depends from claim 1 and is rejected under the same rationale as claim 1. Regarding the further limitation of claim 5, Kim teaches:
"A program for causing a computer to function as the question and answer device according to claim 1"
This limitation is taught by Kim. Kim teaches that the method is embodied in the form of program instructions executable by a computer and recorded on a computer-readable recording medium, the program instructions causing one or more processors of the computer to perform the operations of the question and answer device (Kim, col. 11, lines 43–67; col. 4, lines 11–26; Fig. 1).
Regarding Claim 6, Claim 6 depends from claim 2 and is rejected under the same rationale as claim 3.
Regarding Claim 7, Claim 7 depends from claim 2 and is rejected under the same rationale as claim 2. Regarding the further limitations of claim 7, Kim teaches:
"the question determination circuitry determines machine reading is necessary"
This sub-limitation is taught by Kim. Kim teaches that the circuitry determines whether to apply the MRC question and answer algorithm to the retrieved passage, applying the algorithm when a stored best answer value corresponding to a similar question does not already exist (Kim, col. 6, line 65 – col. 7, line 5; col. 13, lines 55–67).
However, Kim does not teach:
"based on a predetermined question type acquired from the question classification circuitry"
In the same field of endeavor, Pires teaches this limitation. Pires teaches a Question Classification module that produces the question's expected answer type (Pires, p. 2, § 3.1, Question Interpretation), and teaches that, based on that question category, the system determines the answer-production strategy, applying extraction for factoid-type questions and returning a source passage for non-factoid-type questions (Pires, p. 4, § 3.2, Passage Retrieval; p. 10, § 5.3, Answer Extraction), such that the determination of whether reading/extraction is necessary is made based on the predetermined question type acquired from the classification module.
Kim and Pires are analogous to the claimed invention as both are from the same field of endeavor of automated question answering that produces an answer to a user question from retrieved text. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to configure the device of Kim so that the determination of whether machine reading is necessary is made based on a predetermined question type acquired from the question classification circuitry, as taught by Pires. The motivation to combine Kim and Pires is that Kim seeks to provide the best answer most suitable for the user question (Kim, col. 1, lines 40–42) and to improve searching speed and efficiency by limiting unnecessary processing (Kim, col. 5, lines 52–58; col. 7, lines 8–13), and Pires provides a known technique of selecting the answer-production strategy on the basis of the question type produced by the classification module (Pires, p. 4, § 3.2), such that the combination would apply the appropriate answer-production strategy to each question type and avoid expending machine-reading computation where the question type does not require it, yielding the predictable result of an accurate answer with reduced computation (see KSR; MPEP § 2143(A), (C), (G)).
Regarding Claim 9, Claim 9 depends from claim 1 and is rejected under the same rationale as claim 1. Regarding the further limitation of claim 9, Kim teaches:
"the question data used as the input comprises a plurality of pieces of question data"
This limitation is taught by Kim. Kim teaches "a method of providing MRC question and answer, a user question is received and analyzed" and that "a best answer value or no result is provided as an answer to a user" (Kim, col. 16, lines 9–20; Fig. 8, operations S810–S860), and Kim teaches that the receiver receives a question "where a person asks a question by using a natural language-based sentence through a speech recognition speaker or a case where a person asks a question by typing through a chatbot" (Kim, col. 4, lines 18–26). One of ordinary skill in the art would interpret this teaching to derive that the question data used as the input comprises a plurality of pieces of question data, since the device interacts with the user through a chatbot or speech recognition speaker to receive and answer questions and thereby receives and processes multiple pieces of question data as input, yielding the predictable result of providing an answer to each of the plurality of pieces of question data.
Regarding Claim 11, Claim 11 depends from claim 4 and is rejected under the same rationale as claim 4. Regarding the further limitations of claim 11, Kim teaches:
“the question data used as the input comprises a plurality of pieces of question data”
This limitation is taught by Kim. Kim teaches “a method of providing MRC question and answer, a user question is received and analyzed” and that “a best answer value or no result is provided as an answer to a user” (Kim, col. 16, lines 9-20; Fig. 8, operations S810-S860), and Kim teaches that the receiver receives a question “where a person asks a question by using a natural language-based sentence through a speech recognition speaker or a case where a person asks a question by typing through a chatbot” (Kim, col. 4, lines 18-26). One of ordinary skill in the art would interpret this teaching to derive that the question data used as the input comprises a plurality of pieces of question data, since the device interacts with the user through a chatbot or speech recognition speaker to receive and answer questions and thereby receives an process multiple pieces of question data as input, yielding the predictable result of providing an answer to each of the plurality of pieces of question data.
Claims 8 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Kim, in view of Pires, and further in view of Gupta, and further in view of Tan et al. (Tan), Non-Patent Literature, "S-Net: From Answer Extraction to Answer Generation for Machine Reading Comprehension," published June 2, 2018, pp. 1, 2, 3, 6, and 9.
Regarding Claim 8, Claim 8 depends from claim 1 and is rejected under the same rationale as claim 1. Regarding the further limitations of claim 8:
Kim teaches subject matter related to the machine reading of claim 8, in that Kim teaches an MRC question and answer algorithm that comprehends the user question and automatically reads and derives an answer from the passages (Kim, col. 7, lines 43–56). However, Kim does not teach:
"a generative machine reading is used"
In the same field of endeavor, Tan teaches this sub-limitation. Tan teaches an answer synthesis model that applies a sequence-to-sequence neural network to generate the answer from the question and the extracted evidence (Tan, p. 1, Abstract; p. 6, § 3.3, Answer Synthesis), thereby providing a generative machine reading.
"when the answer data exceeds the range of passage data"
In the same field of endeavor, Tan teaches this sub-limitation. Tan teaches that, unlike exact-span extraction, the words in the answer are not necessarily present in the passages, and that the answer is synthesized from the extraction results in an extraction-then-synthesis framework (Tan, p. 1, Abstract; p. 2, Fig. 1; p. 3, § contributions), such that the generated answer data exceeds the range of the passage data.
Kim, Pires, Gupta, and Tan are analogous to the claimed invention as all are from the same field of endeavor of automated question answering that produces an answer to a user question from retrieved text. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to configure the machine reading of Kim so that a generative machine reading is used when the answer data exceeds the range of passage data, as taught by Tan. The motivation to combine Kim, Pires, Gupta, and Tan is that Tan teaches that synthesizing the answer, rather than restricting it to an exact span of the passage, produces answers whose words are not confined to the passage (Tan, p. 1, Abstract) and outperforms pure extraction (Tan, p. 9, Tables 2–3), such that applying Tan's answer generation to Kim would enable a responsive answer to be produced where the correct answer is not a verbatim span of the retrieved passage, yielding the predictable result of improved answer accuracy (see KSR; MPEP § 2143(A), (G)).
Regarding Claim 10, Claim 10 depends from claim 4 and is rejected under the same rationale as claim 4. Regarding the further limitations of claim 10:
Kim teaches subject matter related to the machine reading of claim 10, in that Kim teaches an MRC question and answer algorithm that comprehends the user question and automatically reads and derives an answer from the passages (Kim, col. 7, lines 43–56). However, Kim does not teach:
"a generative machine reading is used"
In the same field of endeavor, Tan teaches this sub-limitation. Tan teaches "we use it to generate the synthetic answer" by applying "the sequence-to-sequence model to synthesize the answer with the extracted evidences as features" (Tan, p. 3, § 3; p. 6, § 3.3, Answer Synthesis), thereby providing a generative machine reading.
"when the answer data exceeds the range of passage data"
Tan further teaches this sub-limitation. Tan teaches that "the words in the answer are not necessary in the passages" and that the model is to "synthesize answers from extraction results" in an extraction-then-synthesis framework (Tan, p. 1, Abstract; p. 2, Fig. 1), such that the generated answer data exceeds the range of the passage data.
Kim and Tan are analogous to the claimed invention as both are from the same field of endeavor of automated question answering that produces an answer to a user question from retrieved text. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to configure the machine reading of Kim so that a generative machine reading is used when the answer data exceeds the range of passage data, as taught by Tan. The motivation to combine Kim and Tan is that Tan teaches that synthesizing the answer, rather than restricting it to an exact span of the passage, produces answers whose words are not confined to the passage (Tan, p. 1, Abstract) and that the extraction-then-synthesis method "outperforms state-of-the-art methods" (Tan, p. 1, Abstract), such that applying Tan's answer generation to Kim would enable a responsive answer to be produced where the correct answer is not a verbatim span of the retrieved passage, yielding the predictable result of improved answer accuracy (see KSR; MPEP § 2143(A), (G)).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HUNG VAN LE whose telephone number is (571)270-0164. The examiner can normally be reached 8 a.m. - 5 p.m..
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, Cesar Paula can be reached at (571) 272-4128. 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.
/HUNG VAN LE/Examiner, Art Unit 2145
/CESAR B PAULA/Supervisory Patent Examiner, Art Unit 2145
Prosecution Insights