Smart Question Answer Method, Apparatus, Device, and Storage Medium

§101 §102 §103

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 . 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 1-9, 11-12, 15-23 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Independent claims 1, 12 similarly recite analyzing preset text information to obtain a structured information set corresponding to the preset text information; acquiring question information; querying target structured information matching the question information from the structured information set; and generating response information corresponding to the question information based on the target structured information. The limitations of analyzing preset text information to obtain a structured information set corresponding to the preset text information, …matching the question information…, as drafted, are processes that, under their broadest reasonable interpretation, cover mental processes but from the recitation of implementing them on generic computer components. That is, nothing in the claim elements preclude the steps from practically being performed in the mind. For example, the limitation pertaining to “analyzing” in the context of this claim encompasses the user analyzing preset text information. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, claims recite an abstract idea (Step 2A, Prong 1). This judicial exception is not integrated into a practical application. In particular, the claim recites the additional elements of – acquiring question information; querying target structured information … from the structured information set; and generating response information corresponding to the question information based on the target structured information. The electronic device, processor, computer program, and memory are recited at a high-level of generality (i.e., as generic computer devices performing generic computer functions) and do not meaningfully limit the claim. The additional elements pertaining to “acquiring”, “querying”, and “generating” represent insignificant extra-solution activities to the judicial exception and are also mere data gathering steps. Accordingly, these additional elements, individually and in combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea (Step 2A, Prong 2). The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements pertaining to “acquiring”, “querying”, and “generating” represent insignificant extra-solution activities that are well-understood, routine, and conventional activities previously known to the industry. That is, these limitations represent well-understood, routine, conventional activities in the fields of data processing and/or data storage and retrieval and are merely directed to the well-understood, routine, conventional activity of storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015) and/or receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information). Therefore, these limitations, both individually and in combination, fail to amount to an inventive concept because they merely append well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception, and thus, do not cause the claim to amount to significantly more than the judicial exception. (Step 2B). Accordingly, claims 1 ,12 are not patent eligible. Independent claim 11 recites a smart question answer method, wherein the method comprises: receiving question information from a terminal; querying target structured information matching the question information from a structured information set corresponding to preset text information; generating response information corresponding to the question information based on the target structured information; and feeding back the response information to the terminal. The limitation of matching the question information, as drafted, are processes that, under their broadest reasonable interpretation, cover mental processes but from the recitation of implementing them on generic computer components. That is, nothing in the claim elements preclude the steps from practically being performed in the mind. For example, the limitation pertaining to “analyzing” in the context of this claim encompasses the user analyzing preset text information. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, claims recite an abstract idea (Step 2A, Prong 1). This judicial exception is not integrated into a practical application. In particular, the claim recites the additional elements of – receiving question information from a terminal; querying target structured information … from a structured information set corresponding to preset text information; generating response information corresponding to the question information based on the target structured information; and feeding back the response information to the terminal. The terminal is recited at a high-level of generality (i.e., as generic computer devices performing generic computer functions) and do not meaningfully limit the claim. The additional elements pertaining to “receiving”, “querying”, “generating” and “feeding back” represent insignificant extra-solution activities to the judicial exception and are also mere data gathering steps. Accordingly, these additional elements, individually and in combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea (Step 2A, Prong 2). The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements pertaining to “receiving”, “querying”, “generating” and “feeding back” represent insignificant extra-solution activities that are well-understood, routine, and conventional activities previously known to the industry. That is, these limitations represent well-understood, routine, conventional activities in the fields of data processing and/or data storage and retrieval and are merely directed to the well-understood, routine, conventional activity of storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015) and/or receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information). Therefore, these limitations, both individually and in combination, fail to amount to an inventive concept because they merely append well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception, and thus, do not cause the claim to amount to significantly more than the judicial exception. (Step 2B). Accordingly, claim 11 is not patent eligible. Claims 2-9, 15-23 depend on claim 1 and include all the limitations of these claims. Therefore, these claims are directed to the same abstract idea and the analysis must proceed to (Step 2A, Prong 2). Claims 2-3, 20 recite additional limitations pertaining to analyzing the preset text information comprising inputting and wherein the inputting comprises, and what the first task is decomposed into. This judicial exception is not integrated into a practical application. The additional elements represent further mental process steps of extracting target content matching, determining the target content, judging text, judging structure generation, analyzing/judging context, and judging text for generation. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Further, the limitations pertaining to analyzing, extracting, and determining are recited as being performed using generic computing components at a high level of generality. In the limitation pertaining to inputting, computing components are used as a tool to perform the generic computer function of (inputting/receiving data). See MPEP 2106.05(f). In the limitations pertaining to analyzing, extracting and determining, the computer is used to perform an abstract idea, as discussed above in Step 2A, Prong One, such that it amounts to no more than mere instructions to apply the exception using a generic computer. See MPEP 2106.05(f). The judicial exception of analyzing, extracting and determining are performed responsive to “inputting the preset text information and a preset type into a machine learning model” which comprises “inputting a representation vector of a sequence”. The machine learning model is used to generally apply the abstract idea without placing any limits on how the trained machine learning model functions. Rather, these limitations only recite the outcome of “extracts target content matching”, “for information extraction”, and “determining the target content in the preset text information matching” and do not include any details about how the analyzing, extracting and determining are accomplished. See MPEP 2106.05(f). The recitation of “inputting the preset text information and a preset type into a machine learning model” which comprises “inputting a representation vector of a sequence” in the limitations also merely indicates a field of use or technological environment in which the judicial exception is performed. Although the additional elements“ inputting the preset text information and a preset type into a machine learning model” which comprises “inputting a representation vector of a sequence” limits the identified judicial exceptions this type of limitation merely confines the use of the abstract idea to a particular technological environment (machine learning models) and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h). This additional step is considered an abstract idea (mental process step) and does not integrate the judicial exception into a practical application. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements represent further mental process steps. The additional element of “inputting the preset text information and a preset type into a machine learning model” which comprises “inputting a representation vector of a sequence” are at best mere instructions to “apply” the abstract ideas, which cannot provide an inventive concept. See MPEP 2106.05(f). The additional elements pertaining to “inputting” were found to be insignificant extra-solution activity in Step 2A, Prong Two, because they were determined to be insignificant limitations as necessary data gathering (for the machine learning model). Therefore, these additional limitations are not sufficient to amount to significantly more than the judicial exception. Claims 2-3, 20 are not patent eligible. Claims 4-5, 21 recite additional limitations pertaining to analyzing the preset text information comprising inputting and wherein the inputting comprises, and what the second task is decomposed into. This judicial exception is not integrated into a practical application. The additional elements represent further mental process steps of mining question-answer information, generating the interrogative sentence information corresponding to the key fields, judging text, judging a structure to generate, judging a context, and judging a generation of text . If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Further, the limitations pertaining to mining question-answer information and generating the interrogative sentence information are recited as being performed using generic computing components at a high level of generality. In the limitation pertaining to inputting, computing components are used as a tool to perform the generic computer function of (inputting/receiving data). See MPEP 2106.05(f). In the limitations pertaining to mining question-answer information and generating the interrogative sentence information, the computer is used to perform an abstract idea, as discussed above in Step 2A, Prong One, such that it amounts to no more than mere instructions to apply the exception using a generic computer. See MPEP 2106.05(f). The judicial exception of mining question-answer information and generating the interrogative sentence information are performed responsive to “inputting the preset text information and a preset type into a machine learning model” which comprises “inputting a representation vector of a sequence consisting of the name of a second task, the preset text information, and the key fields in the preset text information into the machine learning model”. The machine learning model is used to generally apply the abstract idea without placing any limits on how the trained machine learning model functions. Rather, these limitations only recite the outcomes of for mining question-answer information and generating the interrogative sentence information, and do not include any details about how the mining and generating are accomplished. See MPEP 2106.05(f). The recitation of “inputting the preset text information and a preset type into a machine learning model” which comprises “inputting a representation vector of a sequence consisting of the name of a second task, the preset text information, and the key fields in the preset text information into the machine learning model” in the limitations also merely indicates a field of use or technological environment in which the judicial exception is performed. Although the additional elements “inputting the preset text information and a preset type into a machine learning model” which comprises “inputting a representation vector of a sequence consisting of the name of a second task, the preset text information, and the key fields in the preset text information into the machine learning model” limits the identified judicial exceptions this type of limitation merely confines the use of the abstract idea to a particular technological environment (machine learning models) and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h). This additional step is considered an abstract idea (mental process step) and does not integrate the judicial exception into a practical application. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements represent further mental process steps. The additional element of “inputting the preset text information and a preset type into a machine learning model” which comprises “inputting a representation vector of a sequence consisting of the name of a second task, the preset text information, and the key fields in the preset text information into the machine learning model” are at best mere instructions to “apply” the abstract ideas, which cannot provide an inventive concept. See MPEP 2106.05(f). The additional elements pertaining to “inputting” were found to be insignificant extra-solution activity in Step 2A, Prong Two, because they were determined to be insignificant limitations as necessary data gathering (for the machine learning model). Therefore, these additional limitations are not sufficient to amount to significantly more than the judicial exception. Claims 4-5, 21 are not patent eligible. Claims 6-7, 22 recite additional limitations pertaining inputting to determine whether the any structured information matches and wherein the inputting comprises inputting a representation vector of a sequence and what the third task is decomposed into. This judicial exception is not integrated into a practical application. The additional elements represent further mental process steps of determining whether the any structured information matches, selecting structured information matching, judging an understanding of text, judging understanding of dialogue, judging matching, and judging a generation of text. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Further, the limitations pertaining to determining whether the any structured information matches and selecting structured information matching are recited as being performed using generic computing components at a high level of generality. In the limitation pertaining to inputting, computing components are used as a tool to perform the generic computer function of (inputting/receiving data). See MPEP 2106.05(f). In the limitations pertaining to determining whether the any structured information matches and selecting structured information matching, the computer is used to perform an abstract idea, as discussed above in Step 2A, Prong One, such that it amounts to no more than mere instructions to apply the exception using a generic computer. See MPEP 2106.05(f). The judicial exception of determining whether the any structured information matches and selecting structured information matching are performed responsive to “inputting any structured information in the structured information set and the question information into a machine learning model” which comprises “inputting a representation vector of a sequence consisting of the name of a third task, the any structured information in the structured information set, and the question information into the machine learning model”. The machine learning model is used to generally apply the abstract idea without placing any limits on how the trained machine learning model functions. Rather, these limitations only recite the outcomes of determining whether the any structured information matches and selecting structured information matching, and do not include any details about how the mining and generating are accomplished. See MPEP 2106.05(f). The recitation of “inputting any structured information in the structured information set and the question information into a machine learning model” which comprises “inputting a representation vector of a sequence consisting of the name of a third task, the any structured information in the structured information set, and the question information into the machine learning model” in the limitations also merely indicates a field of use or technological environment in which the judicial exception is performed. Although the additional elements “inputting any structured information in the structured information set and the question information into a machine learning model” which comprises “inputting a representation vector of a sequence consisting of the name of a third task, the any structured information in the structured information set, and the question information into the machine learning model” limits the identified judicial exceptions this type of limitation merely confines the use of the abstract idea to a particular technological environment (machine learning models) and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h). This additional step is considered an abstract idea (mental process step) and does not integrate the judicial exception into a practical application. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements represent further mental process steps. The additional element of “inputting any structured information in the structured information set and the question information into a machine learning model” which comprises “inputting a representation vector of a sequence consisting of the name of a third task, the any structured information in the structured information set, and the question information into the machine learning model” are at best mere instructions to “apply” the abstract ideas, which cannot provide an inventive concept. See MPEP 2106.05(f). The additional elements pertaining to “inputting” were found to be insignificant extra-solution activity in Step 2A, Prong Two, because they were determined to be insignificant limitations as necessary data gathering (for the machine learning model). Therefore, these additional limitations are not sufficient to amount to significantly more than the judicial exception. Claims 6-7, 22 are not patent eligible. Claims 8-9, 23 recite additional limitations pertaining inputting to determine whether the any structured information matches and wherein the inputting comprises inputting a representation vector of a sequence and what the third task is decomposed into. This judicial exception is not integrated into a practical application. The additional elements represent further mental process steps of judging response information, judging dialogue understanding, judging dialogue management, judging text generation, judging utilization of knowledge, and judging a text understanding. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Further, the limitations pertaining to generating the response information are recited as being performed using generic computing components at a high level of generality. In the limitation pertaining to inputting, computing components are used as a tool to perform the generic computer function of (inputting/receiving data). See MPEP 2106.05(f). In the limitations pertaining to generating the response information, the computer is used to perform an abstract idea, as discussed above in Step 2A, Prong One, such that it amounts to no more than mere instructions to apply the exception using a generic computer. See MPEP 2106.05(f). The judicial exception of generating the response information is performed responsive to “inputting a representation vector of a sequence consisting of the name of a fourth task, the target structured information, and the question information into the machine learning model”. The machine learning model is used to generally apply the abstract idea without placing any limits on how the trained machine learning model functions. Rather, these limitations only recite the outcome of determining generating the response information, and do not include any details about how the mining and generating are accomplished. See MPEP 2106.05(f). The recitation of “inputting a representation vector of a sequence consisting of the name of a fourth task, the target structured information, and the question information into the machine learning model” in the limitations also merely indicates a field of use or technological environment in which the judicial exception is performed. Although the additional elements “inputting a representation vector of a sequence consisting of the name of a fourth task, the target structured information, and the question information into the machine learning model” limits the identified judicial exceptions this type of limitation merely confines the use of the abstract idea to a particular technological environment (machine learning models) and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h). This additional step is considered an abstract idea (mental process step) and does not integrate the judicial exception into a practical application. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements represent further mental process steps. The additional element of “inputting a representation vector of a sequence consisting of the name of a fourth task, the target structured information, and the question information into the machine learning model” are at best mere instructions to “apply” the abstract ideas, which cannot provide an inventive concept. See MPEP 2106.05(f). The additional elements pertaining to “inputting” were found to be insignificant extra-solution activity in Step 2A, Prong Two, because they were determined to be insignificant limitations as necessary data gathering (for the machine learning model). Therefore, these additional limitations are not sufficient to amount to significantly more than the judicial exception. Claims 8-9, 23 are not patent eligible. Claim 15 recites additional limitations pertaining to the structured information. This judicial exception is not integrated into a practical application. The additional elements represent further mental process steps of analyzing preset text information to obtain a structured information set. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. This additional step is considered an abstract idea (mental process step) and does not integrate the judicial exception into a practical application. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements represent further mental process steps. Therefore, these additional limitations are not sufficient to amount to significantly more than the judicial exception. Claim 15 is not patent eligible. Claim 16 recites additional limitations pertaining to the preset text information. These additional limitations do not integrate the abstract idea into a practical application and merely represent insignificant extra-solution activities to the judicial exception and are mere data gathering steps. Accordingly, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements represent well-understood, routine, conventional activity previously known to the industry. That is, these limitations represent well-understood, routine, conventional activity in the fields of data processing and/or data storage and retrieval and are merely directed to the well-understood, routine, conventional activity of storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015). Therefore, these additional elements do not cause the claim to amount to significantly more than the judicial exception. Claim 17 recites additional limitations pertaining to the server. These devices and/or components are recited at a high-level of generality (i.e., as generic computer devices performing generic computer functions) and do not meaningfully limit the claim. Therefore, these additional elements do not integrate the judicial exception into a practical application. As aforementioned, these devices are recited at a high-level of generality (i.e., as generic computer devices performing generic computer functions) and do not meaningfully limit the claim. Therefore, these additional elements do not cause the claim to amount to significantly more than the judicial exception. Claim 18 recites additional limitations pertaining to the question information. This judicial exception is not integrated into a practical application. The additional elements represent further mental process steps of question information being a user interrogative sequence or declarative question. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. This additional step is considered an abstract idea (mental process step) and does not integrate the judicial exception into a practical application. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements represent further mental process steps. Therefore, these additional limitations are not sufficient to amount to significantly more than the judicial exception. Claim 18 is not patent eligible. Claim 19 recites additional limitations pertaining to the target structured information. This judicial exception is not integrated into a practical application. The additional elements represent further mental process steps of judging a matching of the question information. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. This additional step is considered an abstract idea (mental process step) and does not integrate the judicial exception into a practical application. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements represent further mental process steps. Therefore, these additional limitations are not sufficient to amount to significantly more than the judicial exception. Claim 19 is not patent eligible. Claim Rejections - 35 USC § 102 The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action: A person shall be entitled to a patent unless – (a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale or otherwise available to the public before the effective filing date of the claimed invention. Claims 1-9, 11-12, 18 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Zhang (US 2020/0242444). Regarding claim 1, Zhang discloses: A smart question answer method, wherein the method comprises: analyzing preset text information to obtain a structured information set corresponding to the preset text information ([0039] The KG PNG media_image1.png 38 21 media_image1.png Greyscale 160 is already embedded into two low-dimensional spaces (Predicate Embedding pace 140 and Entity Embedding Space 150), and each fact (h, PNG media_image2.png 38 21 media_image2.png Greyscale ,t) may be represented as three latent vectors, i.e., (eh, PNG media_image3.png 38 38 media_image3.png Greyscale , et). [0067] Given a KG PNG media_image4.png 38 21 media_image4.png Greyscale and a question set Q with corresponding answers, a predicate learning model, a head entity learning model, and a HED model are trained, as shown from line 1 to line 9.); acquiring question information ([0039] given a question 110, [0067] for any new simple question Q); querying target structured information matching the question information from the structured information set ([0039] Instead of inferring the head entity and predicate directly, KEQA embodiments target jointly recovering the question's head entity, predicate, and tail entity representations (êh, PNG media_image3.png 38 38 media_image3.png Greyscale , êt) in the knowledge graph embedding spaces [0067] Then, for any new simple question Q, it is input into the trained predicate learning model, head entity learning model, and HED model to learn its predicted predicate representation PNG media_image5.png 38 29 media_image5.png Greyscale , head entity representation êh, entity name tokens HEDentity, and non-entity name tokens HEDnon. Based on the learned entity name/names in HEDentity, the entire PNG media_image4.png 38 21 media_image4.png Greyscale is searched to find the candidate fact set C. For all facts in C, their joint distances to the predicted representations (êh, PNG media_image5.png 38 29 media_image5.png Greyscale ,êt) are computed based on the objective function in Eq. (9).); and generating response information corresponding to the question information based on the target structured information ([0039] Instead of inferring the head entity and predicate directly, KEQA embodiments target jointly recovering the question's head entity, predicate, and tail entity representations (êh, PNG media_image3.png 38 38 media_image3.png Greyscale , êt) in the knowledge graph embedding spaces [0067] The fact (h*, PNG media_image6.png 38 21 media_image6.png Greyscale ,t*) with the minimum distance is selected. Finally, the head entity h* and predicate PNG media_image6.png 38 21 media_image6.png Greyscale * are returned as the answer of Q). As per claim 2, claim 1 is incorporated, Zhang further discloses: wherein analyzing the preset text information comprises: inputting the preset text information and a preset type into a machine learning model, so that the machine learning model extracts target content matching the preset type from the preset text information, the preset type and the target content constituting structured information ([0067] Given a KG PNG media_image4.png 38 21 media_image4.png Greyscale and a question set Q with corresponding answers, a predicate learning model, a head entity learning model, and a HED model are trained, as shown from line 1 to line 9. Then, for any new simple question Q, it is input into the trained predicate learning model, head entity learning model, and HED model to learn its predicted predicate representation PNG media_image5.png 38 29 media_image5.png Greyscale , head entity representation êh, entity name tokens HEDentity, and non-entity name tokens HEDnon. Based on the learned entity name/names in HEDentity, the entire PNG media_image4.png 38 21 media_image4.png Greyscale is searched to find the candidate fact set C. For all facts in C, their joint distances to the predicted representations (êh, PNG media_image5.png 38 29 media_image5.png Greyscale ,êt) are computed based on the objective function in Eq. (9).). As per claim 3, claim 2 is incorporated, Zhang further discloses: wherein inputting the preset text information and the preset type into the machine learning model, so that the machine learning model extracts the target content matching the preset type from the preset text information comprises: inputting a representation vector of a sequence consisting of the name of a first task, the preset text information, and the preset type into the machine learning model, the first task being used for information extraction; and determining the target content in the preset text information matching the preset type based on outputs of a plurality of atomic capabilities corresponding to the first task within the machine learning model ([0040], [0067] Given a KG PNG media_image4.png 38 21 media_image4.png Greyscale and a question set Q with corresponding answers, a predicate learning model, a head entity learning model, and a HED model are trained, as shown from line 1 to line 9. Then, for any new simple question Q, it is input into the trained predicate learning model, head entity learning model, and HED model to learn its predicted predicate representation PNG media_image5.png 38 29 media_image5.png Greyscale , head entity representation êh, entity name tokens HEDentity, and non-entity name tokens HEDnon. Based on the learned entity name/names in HEDentity, the entire PNG media_image4.png 38 21 media_image4.png Greyscale is searched to find the candidate fact set C. For all facts in C, their joint distances to the predicted representations (êh, PNG media_image5.png 38 29 media_image5.png Greyscale ,êt) are computed based on the objective function in Eq. (9).). As per claim 4, claim 1 is incorporated, Zhang further discloses: wherein analyzing the preset text information comprises: inputting the preset text information and key fields in the preset text information into a machine learning model, so that the machine learning model generates interrogative sentence information corresponding to the key fields, the key fields and the interrogative sentence information constituting structured information ([0040], [0067] Given a KG PNG media_image4.png 38 21 media_image4.png Greyscale and a question set Q with corresponding answers, a predicate learning model, a head entity learning model, and a HED model are trained, as shown from line 1 to line 9. Then, for any new simple question Q, it is input into the trained predicate learning model, head entity learning model, and HED model to learn its predicted predicate representation PNG media_image5.png 38 29 media_image5.png Greyscale , head entity representation êh, entity name tokens HEDentity, and non-entity name tokens HEDnon. Based on the learned entity name/names in HEDentity, the entire PNG media_image4.png 38 21 media_image4.png Greyscale is searched to find the candidate fact set C. For all facts in C, their joint distances to the predicted representations (êh, PNG media_image5.png 38 29 media_image5.png Greyscale ,êt) are computed based on the objective function in Eq. (9).). As per claim 5, claim 4 is incorporated, Zhang further discloses: wherein inputting the preset text information and the key fields in the preset text information into the machine learning model, so that the machine learning model generates the interrogative sentence information corresponding to the key fields comprises: inputting a representation vector of a sequence consisting of the name of a second task, the preset text information, and the key fields in the preset text information into the machine learning model, the second task being used for mining question-answer information; and generating the interrogative sentence information corresponding to the key fields based on outputs of a plurality of atomic capabilities corresponding to the second task within the machine learning model ([0040], [0067] Given a KG PNG media_image4.png 38 21 media_image4.png Greyscale and a question set Q with corresponding answers, a predicate learning model, a head entity learning model, and a HED model are trained, as shown from line 1 to line 9. Then, for any new simple question Q, it is input into the trained predicate learning model, head entity learning model, and HED model to learn its predicted predicate representation PNG media_image5.png 38 29 media_image5.png Greyscale , head entity representation êh, entity name tokens HEDentity, and non-entity name tokens HEDnon. Based on the learned entity name/names in HEDentity, the entire PNG media_image4.png 38 21 media_image4.png Greyscale is searched to find the candidate fact set C. For all facts in C, their joint distances to the predicted representations (êh, PNG media_image5.png 38 29 media_image5.png Greyscale ,êt) are computed based on the objective function in Eq. (9).). As per claim 6, claim 1 is incorporated, Zhang further discloses: wherein querying the target structured information matching the question information from the structured information set comprises: inputting any structured information in the structured information set and the question information into a machine learning model, so that the machine learning model determines whether the any structured information matches the question information, and in response to the any structured information matching the question information, the any structured information is the target structured information matching the question information ([0040], [0067] Given a KG PNG media_image4.png 38 21 media_image4.png Greyscale and a question set Q with corresponding answers, a predicate learning model, a head entity learning model, and a HED model are trained, as shown from line 1 to line 9. Then, for any new simple question Q, it is input into the trained predicate learning model, head entity learning model, and HED model to learn its predicted predicate representation PNG media_image5.png 38 29 media_image5.png Greyscale , head entity representation êh, entity name tokens HEDentity, and non-entity name tokens HEDnon. Based on the learned entity name/names in HEDentity, the entire PNG media_image4.png 38 21 media_image4.png Greyscale is searched to find the candidate fact set C. For all facts in C, their joint distances to the predicted representations (êh, PNG media_image5.png 38 29 media_image5.png Greyscale ,êt) are computed based on the objective function in Eq. (9).). As per claim 7, claim 6 is incorporated, Zhang further discloses: wherein inputting the any structured information in the structured information set and the question information into the machine learning model, so that the machine learning model determines whether the any structured information matches the question information comprises: inputting a representation vector of a sequence consisting of the name of a third task, the any structured information in the structured information set, and the question information into the machine learning model, the third task being used for selecting structured information matching the question information; and generating information indicating whether the any structured information matches the question information based on outputs of a plurality of atomic capabilities corresponding to the third task within the machine learning model ([0040], [0062]-[0065], [0067], [0070]). As per claim 8, claim 1 is incorporated, Zhang further discloses: wherein generating the response information corresponding to the question information based on the target structured information comprises: generating the response information corresponding to the question information based on the target structured information and the question information ([0040], [0067] Given a KG PNG media_image4.png 38 21 media_image4.png Greyscale and a question set Q with corresponding answers, a predicate learning model, a head entity learning model, and a HED model are trained, as shown from line 1 to line 9. Then, for any new simple question Q, it is input into the trained predicate learning model, head entity learning model, and HED model to learn its predicted predicate representation PNG media_image5.png 38 29 media_image5.png Greyscale , head entity representation êh, entity name tokens HEDentity, and non-entity name tokens HEDnon. Based on the learned entity name/names in HEDentity, the entire PNG media_image4.png 38 21 media_image4.png Greyscale is searched to find the candidate fact set C. For all facts in C, their joint distances to the predicted representations (êh, PNG media_image5.png 38 29 media_image5.png Greyscale ,êt) are computed based on the objective function in Eq. (9). The fact (h*, PNG media_image6.png 38 21 media_image6.png Greyscale ,t*) with the minimum distance is selected. Finally, the head entity h* and predicate PNG media_image6.png 38 21 media_image6.png Greyscale * are returned as the answer of Q). As per claim 9, claim 8 is incorporated, Zhang further discloses: wherein generating the response information corresponding to the question information based on the target structured information and the question information comprises: inputting a representation vector of a sequence consisting of the name of a fourth task, the target structured information, and the question information into the machine learning model, the fourth task being used for generating the response information; and generating the response information corresponding to the question information based on outputs of a plurality of atomic capabilities corresponding to the fourth task within the machine learning model (, [0040], [0062]-[0065], [0067], [0070]). Regarding claim 11, Zhang discloses: A smart question answer method, wherein the method comprises: receiving question information from a terminal (([0039] given a question 110, [0067] for any new simple question Q, [0114], [0119]-[0120]) querying target structured information matching the question information from a structured information set corresponding to preset text information ([0039] Instead of inferring the head entity and predicate directly, KEQA embodiments target jointly recovering the question's head entity, predicate, and tail entity representations (êh, PNG media_image3.png 38 38 media_image3.png Greyscale , êt) in the knowledge graph embedding spaces [0067] Then, for any new simple question Q, it is input into the trained predicate learning model, head entity learning model, and HED model to learn its predicted predicate representation PNG media_image5.png 38 29 media_image5.png Greyscale , head entity representation êh, entity name tokens HEDentity, and non-entity name tokens HEDnon. Based on the learned entity name/names in HEDentity, the entire PNG media_image7.png 38 21 media_image7.png Greyscale is searched to find the candidate fact set C. For all facts in C, their joint distances to the predicted representations (êh, PNG media_image5.png 38 29 media_image5.png Greyscale ,êt) are computed based on the objective function in Eq. (9).); generating response information corresponding to the question information based on the target structured information; and feeding back the response information to the terminal ([0039] Instead of inferring the head entity and predicate directly, KEQA embodiments target jointly recovering the question's head entity, predicate, and tail entity representations (êh, PNG media_image3.png 38 38 media_image3.png Greyscale , êt) in the knowledge graph embedding spaces, [0040], [0067] The fact (h*, PNG media_image6.png 38 21 media_image6.png Greyscale ,t*) with the minimum distance is selected. Finally, the head entity h* and predicate PNG media_image6.png 38 21 media_image6.png Greyscale * are returned as the answer of Q, [0114], [0119]-[0120]). Regarding claim 12, Zhang discloses: An electronic device, comprising: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and is arranged for being executed by the processor to implement the following steps: analyzing preset text information to obtain a structured information set corresponding to the preset text information ([0039] The KG PNG media_image1.png 38 21 media_image1.png Greyscale 160 is already embedded into two low-dimensional spaces (Predicate Embedding pace 140 and Entity Embedding Space 150), and each fact (h, PNG media_image2.png 38 21 media_image2.png Greyscale ,t) may be represented as three latent vectors, i.e., (eh, PNG media_image3.png 38 38 media_image3.png Greyscale , et). [0067] Given a KG PNG media_image4.png 38 21 media_image4.png Greyscale and a question set Q with corresponding answers, a predicate learning model, a head entity learning model, and a HED model are trained, as shown from line 1 to line 9.); acquiring question information ([0039] given a question 110, [0067] for any new simple question Q); querying target structured information matching the question information from the structured information set ([0039] Instead of inferring the head entity and predicate directly, KEQA embodiments target jointly recovering the question's head entity, predicate, and tail entity representations (êh, PNG media_image3.png 38 38 media_image3.png Greyscale , êt) in the knowledge graph embedding spaces [0067] Then, for any new simple question Q, it is input into the trained predicate learning model, head entity learning model, and HED model to learn its predicted predicate representation PNG media_image5.png 38 29 media_image5.png Greyscale , head entity representation êh, entity name tokens HEDentity, and non-entity name tokens HEDnon. Based on the learned entity name/names in HEDentity, the entire PNG media_image7.png 38 21 media_image7.png Greyscale is searched to find the candidate fact set C. For all facts in C, their joint distances to the predicted representations (êh, PNG media_image5.png 38 29 media_image5.png Greyscale ,êt) are computed based on the objective function in Eq. (9).); and generating response information corresponding to the question information based on the target structured information ([0039] Instead of inferring the head entity and predicate directly, KEQA embodiments target jointly recovering the question's head entity, predicate, and tail entity representations (êh, PNG media_image3.png 38 38 media_image3.png Greyscale , êt) in the knowledge graph embedding spaces [0067] The fact (h*, PNG media_image8.png 38 21 media_image8.png Greyscale ,t*) with the minimum distance is selected. Finally, the head entity h* and predicate PNG media_image6.png 38 21 media_image6.png Greyscale * are returned as the answer of Q). As per claim 18, claim 1 is incorporated, Zhang further discloses: wherein the question information is a user interrogative sentence, or the question information is a declarative question ([0039] given a question 110, [0067] for any new simple question Q). Claim Rejections - 35 USC § 103 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 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. Claims 15, 19 are rejected under 35 U.S.C. 103 as being unpatentable over Zhang (US 2020/0242444) in view of Seonwoo (US 2023/0418868). As per claim 15, claim 1 is incorporated, Zhang fails to disclose “wherein the structured information set is a Document Graph” However, Seonwoo teaches the above limitation ([0066] document VKGs, [0082], [0088]-[0089], Fig. 7). Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the claimed invention to incorporate the teaching of Seonwoo into the teaching of Zhang because the references similarly disclose queries or question answering and graph data. Consequently, one of ordinary skill in the art would be motivated to further modify the system as in Zhang to further include the document graphs as in Seonwoo in order to “efficiently identify a document corresponding to the query for users” (Seonwoo, [0026]). As per claim 19, claim 1 is incorporated, Zhang fails to disclose “wherein the target structured information matching the question information is selected from a Document Graph through knowledge grounding” However, Seonwoo teaches the above limitation ([0047] training component 220 receives ground truth training data associating the pair of the set of mentions with a common entity. Training component 220 computes a cross-entropy loss based on the relation vector and the ground truth training data, where the parameters of relation encoder network 235 are updated based on the cross-entropy loss, [0117], [0141] ). Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the claimed invention to incorporate the teaching of Seonwoo into the teaching of Zhang because the references similarly disclose queries or question answering and graph data. Consequently, one of ordinary skill in the art would be motivated to further modify the system as in Zhang to further include the ground truth in training and document retrieval as in Seonwoo in order to increase accuracy and truth of the model. Claim 16 is rejected under 35 U.S.C. 103 as being unpatentable over Zhang (US 2020/0242444) in view Seonwoo (US 2023/0418868) and further in view of Mitra (US 2023/0169361). As per claim 16, claim 15 is incorporated, Zhang, Seonwoo fail to disclose “wherein the preset text information is pre-stored information in a server” However, Mitra teaches the above limitation ([0031] KG database, [0032]). Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the claimed invention to incorporate the teaching of Mitra into the teaching of Zhang, Seonwoo because the references similarly disclose queries or question answering and graph data. Consequently, one of ordinary skill in the art would be motivated to further modify the system as in the combination of references to further include the stored knowledge graphs as in Mitra in order to “improve speed and efficiency of the process and conserve computing resources” (Mitra, [0035]). Claim 17 is rejected under 35 U.S.C. 103 as being unpatentable over Zhang (US 2020/0242444) in view of Seonwoo (US 2023/0418868) and Mitra (US 2023/0169361) ) and further in view of Galitsky (US 2023/0043849). As per claim 17 claim 16 is incorporated, Zhang, Seonwoo, Mitra fail to disclose “wherein the server is arranged for performing information extraction and question answer mining on the preset text information in an offline state to obtain the Document Graph”. However, Galitsky teaches the above limitation ([0047] The computing device 102 may include a lattice manager 122. The lattice manager 122 may be configured to generate one or more decision trees (e.g., decision trees, supported decision trees, etc.) from the corpus of documents within database 110. In some embodiments the lattice manager 122 may utilize the techniques discussed herein in connection with FIGS. 2-11 to generate these decision trees which may then be stored in a data store (e.g., decision trees 116) for subsequent use. The lattice manager 122 may be configured to generate a lattice from the supported decision trees of decision trees 116. The generated lattice may be stored in a data store (e.g., lattices 118) for subsequent use. In some embodiments, the supported decision trees and/or the lattice may be generated in an offline process). Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the claimed invention to incorporate the teaching of Galitsky into the teaching of Zhang, Seonwoo because the references similarly disclose queries or question answering and graph data. Consequently, one of ordinary skill in the art would be motivated to further modify the system as in the combination of references to further include the offline generation of trees or latices as in Galitsky in order to reduce or eliminate runtime overhead. Claims 20-23 are rejected under 35 U.S.C. 103 as being unpatentable over Zhang (US 2020/0242444) in view of Sengupta (US 2022/0156582). As per claim 20, claim 3 is incorporated, Zhang fails to disclose “wherein information extraction is denoted as the first task, and the first task is decomposed into four atomic capabilities including text understanding, structure generation, context understanding, and text generation However, Sengupta teaches the above limitation ([0027] generation of KGs, [0031], [0038]-[0042], [0077]). Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the claimed invention to incorporate the teaching of Sengupta into the teaching of Zhang because the references similarly disclose queries or question answering and graph data. Consequently, one of ordinary skill in the art would be motivated to further modify the system as in Zhang to further include the capabilities as in Sengupta in order to be able to “efficiently build and/or augment knowledge graphs and improve the quality of the generated responses by a dialogue agent despite a sparsity of data” (Sengupta, [Abstract]). As per claim 21, claim 5 is incorporated, Zhang fails to disclose “wherein question answer mining is denoted as the second task, and the second task is decomposed into four atomic capabilities including text understanding, structure generation, context understanding, and text generation” However, Sengupta teaches the above limitation ([0027] generation of KGs, [0031], [0038]-[0042], [0077]). Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the claimed invention to incorporate the teaching of Sengupta into the teaching of Zhang because the references similarly disclose queries or question answering and graph data. Consequently, one of ordinary skill in the art would be motivated to further modify the system as in Zhang to further include the capabilities as in Sengupta in order to be able to “efficiently build and/or augment knowledge graphs and improve the quality of the generated responses by a dialogue agent despite a sparsity of data” (Sengupta, [Abstract]). As per claim 22, claim 7 is incorporated, Zhang fails to disclose “wherein knowledge selection is denoted as the third task, and the third task is decomposed into four atomic capabilities including text understanding, dialogue understanding, matching, and text generation” However, Sengupta teaches the above limitation ([0027] generation of KGs, [0031], [0038]-[0042], [0077]). Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the claimed invention to incorporate the teaching of Sengupta into the teaching of Zhang because the references similarly disclose queries or question answering and graph data. Consequently, one of ordinary skill in the art would be motivated to further modify the system as in Zhang to further include the capabilities as in Sengupta in order to be able to “efficiently build and/or augment knowledge graphs and improve the quality of the generated responses by a dialogue agent despite a sparsity of data” (Sengupta, [Abstract]). As per claim 23, claim 9 is incorporated, Zhang fails to disclose “wherein response generation is denoted as the fourth task, and the fourth task is decomposed into five atomic capabilities including dialogue understanding, dialogue management, text generation, knowledge utilization, and text understanding” However, Sengupta teaches the above limitation ([0027] generation of KGs, [0031], [0038]-[0042], [0077]). Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the claimed invention to incorporate the teaching of Sengupta into the teaching of Zhang because the references similarly disclose queries or question answering and graph data. Consequently, one of ordinary skill in the art would be motivated to further modify the system as in Zhang to further include the capabilities as in Sengupta in order to be able to “efficiently build and/or augment knowledge graphs and improve the quality of the generated responses by a dialogue agent despite a sparsity of data” (Sengupta, [Abstract]). Pertinent Prior Art The following prior art made of record and not relied upon is considered pertinent to applicant's disclosure: Japa (US 2022/0292262) discloses hybrid question answering over knowledge graph. Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to WILLIAM P BARTLETT whose telephone number is (469)295-9085. The examiner can normally be reached on M-Th 11:30-8:30, F 11-3. 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, Sherief Badawi can be reached on 571-272-9782. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000. /WILLIAM P BARTLETT/ Primary Examiner, Art Unit 2169