Information processing method and device, electronic equipment and computer readable medium

Abstract

The invention discloses an information processing method and device, electronic equipment and a computer readable medium, and the method comprises the steps: after obtaining demand description information, generating an analysis result according to whether the semantics of the demand description information is clear or not; determining whether the demand description information belongs to a fuzzy demand class or not according to the analysis result; in response to the condition that the demand description information belongs to a fuzzy demand class, determining a demand identification result of the demand description information according to the demand number determined from the demand description information; and in response to the condition that the demand description information does not belong to the fuzzy demand class, determining a demand identification result of the demand description information according to the predicted output type of the demand description information, so that the user demand can be automatically identified, and a target module can be subsequently selected from at least one candidate processing module according to the demand identification result. And the target module is used for determining a feedback result of the demand description information, so that the user demand can be automatically processed.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an information processing method, device, electronic device, and computer-readable medium. Background Art

[0002] For some application scenarios, such as free conversation, after obtaining the question (Query) input by the user, the user's needs are identified from the Query so that the needs can be met with the help of a plug-in or a model.

[0003] However, how to identify user needs is a technical problem that needs to be solved urgently. Summary of the Invention

[0004] In order to solve the above technical problems, the present application provides an information processing method, device, electronic device, and computer-readable medium.

[0005] In order to achieve the above objectives, the technical solutions provided by this application are as follows:

[0006] The present application provides an information processing method, the method comprising:

[0007] After obtaining the requirement description information, generating an analysis result based on whether the semantics of the requirement description information is clear;

[0008] Determining whether the requirement description information belongs to a fuzzy requirement class based on the analysis result;

[0009] In response to the requirement description information belonging to a fuzzy requirement class, determining a requirement identification result of the requirement description information based on the number of requirements determined from the requirement description information;

[0010] In response to the requirement description information not belonging to the fuzzy requirement class, determining a requirement recognition result of the requirement description information according to a predicted output type of the requirement description information; the predicted output type is obtained by performing output type prediction processing on the requirement description information;

[0011] A target module is selected from at least one candidate processing module according to a requirement identification result of the requirement description information; the target module is used to determine a feedback result of the requirement description information.

[0012] In one possible implementation manner, determining a requirement identification result of the requirement description information according to a predicted output type corresponding to the requirement description information includes:

[0013] If the predicted output type is a text type, analyzing the query object of the demand description information;

[0014] Determine the requirement recognition result of the requirement description information based on the query object.

[0015] In a possible implementation manner, the determining the requirement recognition result of the requirement description information based on the query object includes:

[0016] If the query object is a preset object, determine the preset status Q&A label as the requirement recognition result of the requirement description information;

[0017] If the query object is not a preset object, determine the preset comprehensive Q&A label as the requirement recognition result of the requirement description information.

[0018] In a possible implementation manner, the determining the requirement recognition result of the requirement description information based on the predicted output type corresponding to the requirement description information includes:

[0019] If the predicted output type is an instruction type, perform operation entity recognition processing on the requirement description information to obtain an entity recognition result;

[0020] Determine the requirement recognition result of the requirement description information based on the entity recognition result.

[0021] In a possible implementation manner, the determining the requirement recognition result of the requirement description information based on the entity recognition result includes:

[0022] If the entity recognition result indicates that there is an operation entity in the requirement description information, determine the preset requirement label corresponding to the operation entity as the requirement recognition result of the requirement description information;

[0023] If the entity recognition result indicates that there is no operation entity in the requirement description information, determine the preset instruction label as the requirement recognition result of the requirement description information.

[0024] In a possible implementation manner, the determining the requirement recognition result of the requirement description information based on the number of requirements determined from the requirement description information includes:

[0025] If the number of requirements is 0, determine the preset no-requirement label as the requirement recognition result of the requirement description information;

[0026] If the number of requirements is greater than 1, determine the preset multi-requirement label as the requirement recognition result of the requirement description information.

[0027] In a possible implementation manner, after determining the requirement recognition result of the requirement description information, the method further includes:

[0028] If the correlation characterization data between the requirement description information and the historical requirement information exceeds a preset correlation threshold, update the requirement recognition result of the requirement description information according to the requirement recognition result of the historical requirement information.

[0029] In a possible implementation manner, the method further includes:

[0030] Compare and process the requirement recognition result of the requirement description information and the requirement recognition result of the historical requirement information to obtain a comparison result;

[0031] If the comparison result indicates that the requirement recognition result of the requirement description information is different from the requirement recognition result of the historical requirement information, determine whether the correlation characterization data between the requirement description information and the historical requirement information exceeds a preset correlation threshold.

[0032] In a possible implementation manner, the method further includes:

[0033] Select a target module from at least one candidate processing module according to the requirement recognition result of the requirement description information; the target module is used to determine the feedback result of the requirement description information;

[0034] Display the feedback result.

[0035] In a possible implementation manner, the target module is a plug-in;

[0036] The target module is specifically used for: performing configuration analysis on the requirement description information by using a plug-in model deployed in the target module to obtain a configuration analysis result; processing the requirement description information according to the configuration analysis result to obtain the feedback result.

[0037] In a possible implementation manner, the configuration analysis result includes a tool selection result and a parameter extraction result.

[0038] This application provides an information processing device, including:

[0039] A semantic analysis unit, configured to generate an analysis result according to whether the semantics of the requirement description information is clear after obtaining the requirement description information;

[0040] A first determination unit, configured to determine whether the requirement description information belongs to the fuzzy requirement category according to the analysis result;

[0041] A second determination unit, configured to, in response to the requirement description information belonging to the fuzzy requirement category, determine the requirement recognition result of the requirement description information according to the number of requirements determined from the requirement description information;

[0042] A third determination unit, configured to determine a requirement recognition result of the requirement description information according to a predicted output type of the requirement description information in response to the requirement description information not belonging to the fuzzy requirement category; the predicted output type is obtained by performing an output type prediction process on the requirement description information.

[0043] A module selection unit, configured to select a target module from at least one candidate processing module according to the requirement recognition result of the requirement description information; the target module is configured to determine a feedback result of the requirement description information.

[0044] This application provides an electronic device, which includes: a processor and a memory;

[0045] The memory is configured to store instructions or computer programs;

[0046] The processor is configured to execute the instructions or computer programs in the memory, so that the electronic device executes the information processing method provided by this application.

[0047] This application provides a computer-readable medium, in which instructions or computer programs are stored. When the instructions or computer programs run on a device, the device is caused to execute the information processing method provided by this application.

[0048] This application provides a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes program codes for executing the information processing method provided by this application.

[0049] Compared with the related art, this application has at least the following advantages:

[0050] In the technical solution provided by this application, after obtaining requirement description information, such as a Query input by a user, an analysis result is first generated based on whether the semantics of the requirement description information is clear; then, based on the analysis result, it is determined whether the requirement description information belongs to the fuzzy requirement category; if the requirement description information belongs to the fuzzy requirement category, the requirement recognition result of the requirement description information is determined based on the number of requirements determined from the requirement description information, so that the requirement recognition result can indicate which subtype of the fuzzy requirement category the requirement description information belongs to; however, if the requirement description information does not belong to the fuzzy requirement category, it can be determined that the requirement description information belongs to the clear requirement category, so the requirement recognition result of the requirement description information is determined based on the predicted output type of the requirement description information, so that the requirement recognition result can indicate which subtype of the clear requirement category the requirement description information belongs to. In this way, automatic recognition of user requirements can be achieved, so that subsequently, based on the requirement recognition result, a target module can be selected from at least one candidate processing module, so that the target module is used to determine the feedback result of the requirement description information, and in this way, automatic processing of user requirements can be achieved.

[0051] In addition, this application determines the requirement recognition result of the requirement description information through a multi-level division method for the requirement description information, so that the requirement recognition result can better represent the user requirements described by the requirement description information. This is beneficial to improving the recognition accuracy, thus beneficial to improving the feedback accuracy for user problems, and further beneficial to improving the user's question-and-answer experience.

[0052] Furthermore, since the requirement recognition result of the above-mentioned requirement description information can accurately represent the user requirements described by the requirement description information, the target module selected based on the requirement recognition result is more suitable for processing the user request, so that the feedback result determined by using the target module for the requirement description information is more accurate. This is beneficial to improving the requirement feedback effect. Brief Description of the Drawings

[0053] In order to more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, the drawings described below are only some embodiments recorded in this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0054] Figure 1 It is a flowchart of an information processing method provided by an embodiment of this application;

[0055] Figure 2 It is a flowchart of a multi-level division method provided by an embodiment of this application;

[0056] Figure 3 A schematic diagram of a requirement recognition process in a multi-round dialogue scenario provided by an embodiment of the present application;

[0057] Figure 4 A schematic diagram of a requirement processing process provided by an embodiment of the present application;

[0058] Figure 5 A schematic diagram of the structure of an information processing device provided by an embodiment of the present application;

[0059] Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0060] In order to enable those skilled in the art of the present technology to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0061] To better understand the technical solutions provided by the present application, the information processing method provided by the present application will be described below with reference to some accompanying drawings first. As Figure 1 shown, the information processing method provided by the embodiments of the present application includes S1-S4 below. Among them, the Figure 1 is a flowchart of an information processing method provided by an embodiment of the present application.

[0062] S1: After obtaining the requirement description information, generate an analysis result according to whether the semantics of the requirement description information is clear. <>

[0063] Among them, the requirement description information refers to the information that needs to be processed for requirement recognition; moreover, the present application does not limit the requirement description information. For example, the requirement description information can be implemented using a Query input by the user, so that the requirement description information can describe the user's requirements. In addition, the present application does not limit the acquisition method of the requirement description information.

[0064] The analysis result is used to indicate whether the semantics of the above-mentioned requirement description information is clear; moreover, this application does not limit the determination process of this analysis result. For example, this analysis result can be determined by means of pre-constructed fuzzy semantics recognition rules. Among them, the fuzzy semantics recognition rule refers to a rule set in advance for describing the characteristics of fuzzy semantics; moreover, this fuzzy semantics recognition rule can be set in advance according to the application scenario. For another example, this analysis result can be determined by means of a pre-constructed fuzzy semantics recognition model. Among them, this fuzzy semantics recognition model is used to determine whether the input data of this fuzzy semantics recognition model carries clear semantics.

[0065] It should be noted that this application does not limit the fuzzy semantics in the above paragraph. For example, for a Query, if the sentence components in the Query are missing, it will cause the semantics expressed by the Query to be unclear, so it can be determined that the Query carries fuzzy semantics; if there are some errors in the Query, such as word order errors, grammar errors or word usage errors, etc., it will cause the semantics expressed by the Query to be unclear, so it can be determined that the Query carries fuzzy semantics.

[0066] Based on the relevant content of S1 above, for some application scenarios, after obtaining the requirement description information, such as Figure 2 the user question shown, analyze whether the semantics of this requirement description information is clear, and obtain an analysis result, so that this analysis result can indicate whether the semantics of this requirement description information is clear, so as to be able to determine the category to which this requirement description information belongs based on this analysis result subsequently.

[0067] S2: According to the analysis result, determine whether the requirement description information belongs to the fuzzy requirement category.

[0068] In this application, after obtaining the above-mentioned analysis result, if this analysis result indicates that the semantics of the above-mentioned requirement description information is clear, it can be determined that this requirement description information can clearly represent the user requirement, so it can be determined that this requirement description information belongs to the clear requirement category and does not belong to the fuzzy requirement category; however, if this analysis result indicates that the semantics of this requirement description information is fuzzy, it can be determined that this requirement description information cannot clearly represent the user requirement, so it can be determined that this requirement description information belongs to the fuzzy requirement category.

[0069] S3: In response to the requirement description information belonging to the fuzzy requirement category, determine the requirement recognition result of this requirement description information according to the number of requirements determined from this requirement description information.

[0070] In this application, for the above-mentioned requirement description information, if the requirement description information belongs to the fuzzy requirement category, it can be determined that the requirement description information cannot clearly represent the user's requirements. Thus, it can be speculated that there may be multiple requirements or no requirements when performing requirement analysis on the requirement description information. Therefore, in order to further improve the recognition accuracy, the requirement recognition result of the requirement description information can be determined based on the number of requirements determined from the requirement description information, so that the requirement recognition result can indicate which subclass of the fuzzy requirement category the requirement description information belongs to, such as Figure 2 the multi-requirement category or the no-requirement category shown, etc. Among them, the requirement recognition result is used to represent the result obtained by performing requirement recognition on the requirement description information, so that the requirement recognition result can represent the user's requirements. The number of requirements is used to represent the number of requirements corresponding to the requirement description information; moreover, this application does not limit the determination method of the number of requirements. For example, the number of requirements can be determined by means of a pre-constructed number-of-requirements analysis model. Among them, the number-of-requirements analysis model is used to perform number-of-requirements analysis on the input data of the number-of-requirements analysis model; moreover, this application does not limit the implementation manner of the number-of-requirements analysis model. Again, the determination process of the number of requirements can be: first, analyze the requirements corresponding to the requirement description information by means of a preset analysis method; then count the number of requirements corresponding to the requirement description information to obtain the number of requirements. Among them, the preset analysis method is used to analyze the requirements corresponding to a Query; moreover, this application does not limit the implementation manner of the preset analysis method. For example, it can be implemented by using any existing or future method that can analyze which requirements a Query corresponds to.

[0071] In addition, this application does not limit the implementation manner of S3 above. For example, it can specifically include steps 11 - step 12 below.

[0072] Step 11: If the above-mentioned number of requirements is 0, then determine the preset no-requirement label as the requirement recognition result of the above-mentioned requirement description information.

[0073] Among them, the preset no-requirement label refers to the requirement label preset for the no-requirement category; moreover, this application does not limit the implementation manner of the preset no-requirement label. For example, the preset no-requirement label can be used to represent the no-requirement category. Again, the preset no-requirement label can be used to represent what means to use to process the Query under the no-requirement category. It can be seen that in a possible implementation manner, the preset no-requirement label can be implemented by using the string "model corresponding to the no-requirement category" so that it can be known later based on the preset no-requirement label that the model corresponding to the no-requirement category can be used to process the current Query.

[0074] Based on the relevant content of step 11 above, for the above demand description information, such as declarative sentences similar to "The weather is really nice", if the demand description information belongs to the fuzzy demand category and the number of demands determined from the demand description information is 0, it can be determined that the demand description information cannot represent any demand, and thus it can be determined that the demand description information belongs to the no-demand category. Therefore, the preset no-demand label corresponding to the no-demand category can be directly determined as the demand recognition result of the demand description information, so that the demand recognition result can represent the information that the demand description information belongs to the no-demand category and / or the means of processing the demand description information.

[0075] Step 12: If the number of demands above is greater than 1, then determine the preset multi-demand label as the demand recognition result of the above demand description information.

[0076] Among them, the preset multi-demand label refers to the demand label preset for the multi-demand category; and this application does not limit the implementation manner of the preset multi-demand label. For example, the preset multi-demand label can be used to represent the multi-demand category. Another example is that the preset multi-demand label can be used to represent the means of processing the Query under the multi-demand category. It can be seen that in a possible implementation manner, the preset multi-demand label can be implemented by using the string "model corresponding to the multi-demand category", so that subsequently, based on the preset multi-demand label, it can be known that the model corresponding to the multi-demand category can be used to process the current Query.

[0077] Based on the relevant content of step 12 above, for the above demand description information, such as a single noun similar to "XXX document", if the demand description information belongs to the fuzzy demand category and the number of demands determined from the demand description information ≥ 2, it can be determined that the demand description information corresponds to multiple demands, such as question-and-answer demands, search demands, document processing demands, etc., and thus it can be determined that the demand description information belongs to the multi-demand category. Therefore, the preset multi-demand label corresponding to the multi-demand category can be directly determined as the demand recognition result of the demand description information, so that the demand recognition result can represent the information that the demand description information belongs to the multi-demand category and / or the means of processing the demand description information.

[0078] Based on the relevant content of steps 11 to 12 above, for the above demand description information, after determining that the demand description information belongs to the fuzzy demand category, first analyze the number of demands corresponding to the demand description information; then determine the demand recognition result of the demand description information according to the number, so that the demand recognition result can at least represent which subclass under the fuzzy demand category the demand description information belongs to, such as the no-demand category or the multi-demand category, etc., so that the demand recognition result can better represent the user demand.

[0079] S4: In response to the requirement description information not belonging to the fuzzy requirement category, determine the requirement recognition result of the requirement description information according to the predicted output type of the requirement description information; the predicted output type is obtained by performing output type prediction processing on the requirement description information.

[0080] Among them, the predicted output type is used to represent the type to which the feedback for the above requirement description information belongs, such as the text type or the instruction type, etc.; and the predicted output type is obtained by performing output type prediction processing on the requirement description information. It should be noted that this application does not limit the determination method of the predicted output type. For example, the predicted output type can be implemented using a pre-constructed type prediction model. Among them, the type prediction model is used to perform output type prediction processing on the input data of the type prediction model; and this application does not limit the implementation manner of the type prediction model.

[0081] In addition, this application does not limit the implementation manner of S4 above. For example, it can specifically be: If the above requirement description information does not belong to the fuzzy requirement category, then according to the predicted output type of the requirement description information, look up the requirement label corresponding to the predicted output type in the pre-constructed first mapping relationship as the requirement recognition result of the requirement description information. Among them, the first mapping relationship is used to record the requirement labels corresponding to some output types; and the first mapping relationship can be set according to the actual application scenario.

[0082] In addition, in order to better improve the recognition accuracy, this application also provides a possible implementation manner of S4 above. In this implementation manner, S4 can specifically include the following steps 21 - step 22.

[0083] Step 21: If the above requirement description information does not belong to the fuzzy requirement category and the predicted output type of the requirement description information is the text type, then analyze the query object of the requirement description information.

[0084] Among them, the query object is used to represent which object the above requirement description information requests to perform query processing on. It should be noted that this application does not limit the implementation manner of the object. It can adopt any object, such as a question and answer system, a question and answer robot, weather, etc.

[0085] In addition, this application does not limit the acquisition method of the above query object. For example, it can specifically be implemented using any existing or future method that can perform object analysis processing on the requirement description information.

[0086] Based on the relevant content of Step 21 above, for the above demand description information, after determining that the demand description information does not belong to the fuzzy demand category, if the predicted output type of the demand description information is text type, it can be determined that the demand description information belongs to the Q&A subclass under the clear demand category. Therefore, the query object of the demand description information can be further analyzed so that the subclass under the Q&A category to which the demand description information belongs can be determined based on the query object later, such as Figure 2 the comprehensive Q&A category or the preset object Q&A category shown.

[0087] Step 22: Determine the demand recognition result of the above demand description information based on the above query object.

[0088] It should be noted that the present application does not limit the implementation manner of Step 22 above. For example, specifically, after obtaining the above query object, the demand label corresponding to the query object is searched from the pre-set second mapping relationship as the demand recognition result of the above demand description information. Among them, the second mapping relationship is used to record the demand labels corresponding to multiple objects; and the second mapping relationship can be set according to the actual application scenario.

[0089] In fact, in order to better improve the recognition accuracy, the present application also provides a possible implementation manner of Step 22 above. In this implementation manner, Step 22 specifically may include Step 221 - Step 222 below.

[0090] Step 221: If the above query object is a preset object, the preset status Q&A label is determined as the demand recognition result of the demand description information.

[0091] Among them, the preset object refers to an object that is pre-set and configured with an independent query tool; and the present application does not limit the preset object. For example, the preset object may be a Q&A system or a Q&A robot.

[0092] The preset status Q&A label refers to the demand label pre-set for the preset object Q&A category; and the present application does not limit the implementation manner of the preset status Q&A label. For example, the preset status Q&A label can be used to represent the preset object Q&A category. Another example is that the preset status Q&A label can be used to represent what means are used to process the Query under the preset object Q&A category. It can be seen that in a possible implementation manner, the preset status Q&A label can be implemented by using the string "model corresponding to the preset object Q&A category" so that it can be known later based on the preset status Q&A label that the model corresponding to the preset object Q&A category can be used to process the current Query.

[0093] Based on the relevant content of step 221 above, for the above demand description information, such as questions like "Who are you", after determining that the demand description information belongs to the subclass of Q&A under the category of clear demands, if the query object of the demand description information is a preset object, it can be determined that the demand description information belongs to the preset object Q&A category. Therefore, the preset status Q&A label corresponding to the preset object Q&A category can be determined as the demand recognition result of the demand description information, so that the demand recognition result can represent the information that the demand description information belongs to the preset object Q&A category and / or the means of processing the demand description information.

[0094] Step 222: If the above query object is not a preset object, then determine the preset comprehensive Q&A label as the demand recognition result of the demand description information.

[0095] Among them, the preset comprehensive Q&A label refers to the demand label preset for the comprehensive Q&A category; moreover, the present application does not limit the implementation manner of the preset comprehensive Q&A label. For example, the preset comprehensive Q&A label can be used to represent the comprehensive Q&A category. Another example is that the preset comprehensive Q&A label can be used to represent the means of processing the Query under the comprehensive Q&A category. It can be seen that in a possible implementation manner, the preset comprehensive Q&A label can be implemented using the string "model corresponding to the comprehensive Q&A category" so that subsequently, based on the preset comprehensive Q&A label, it can be known that the model corresponding to the comprehensive Q&A category can be used to process the current Query.

[0096] Based on the relevant content of step 222 above, for the above demand description information, such as questions like "What day is a certain festival", after determining that the demand description information belongs to the subclass of Q&A under the category of clear demands, if the query object of the demand description information does not belong to the preset object, it can be determined that the demand description information is used to query comprehensive knowledge, and thus it can be determined that the demand description information belongs to the comprehensive Q&A category. Therefore, the preset comprehensive Q&A label corresponding to the comprehensive Q&A category can be determined as the demand recognition result of the demand description information, so that the demand recognition result can represent the information that the demand description information belongs to the comprehensive Q&A category and / or the means of processing the demand description information.

[0097] Based on the relevant content of the above steps 221 to 222, for the above demand description information, if the demand description information belongs to the subclass of Q&A under the clear demand category, after determining whether the query object of the demand description information is a preset object, according to the judgment result, determine the demand recognition result of the above demand description information, so that the demand recognition result can represent which subclass under this Q&A category the demand description information belongs to, such as the comprehensive Q&A category or the preset object Q&A category, etc., so that the demand recognition result can better represent the user demand.

[0098] Based on the relevant content of the above steps 21 to 22, for the above demand description information, if the demand description information does not belong to the fuzzy demand category, it can be determined that the demand description information belongs to the clear demand category. Therefore, when it is determined that the predicted output type of the demand description information is the text type, it can be determined that the demand description information belongs to the subclass of Q&A under this clear demand category. Moreover, in order to better improve the recognition accuracy, it can be further determined whether the query object of the demand description information is a preset object to obtain a judgment result; then, according to the judgment result, determine the demand recognition result of the above demand description information, so that the demand recognition result can more accurately represent which subclass under this Q&A category the demand description information belongs to, such as the comprehensive Q&A category or the preset object Q&A category, etc., so that the demand recognition result can better represent the user demand.

[0099] In addition, in order to better improve the recognition accuracy, the present application also provides a possible implementation manner of the above S4. Under this implementation manner, S4 can specifically include the following steps 31 - step 32.

[0100] Step 31: If the above demand description information does not belong to the fuzzy demand category and the predicted output type of the demand description information is the instruction type, perform operation entity recognition processing on the demand description information to obtain an entity recognition result.

[0101] Among them, the entity recognition result is used to represent whether there is an operation entity in the above demand description information, such as a document, a table, a schedule, etc. It should be noted that the present application does not limit the implementation manner of the operation entity. For example, the operation entity can be set in advance according to the actual application scenario, so that the operation entity can represent a pre-configured functional entity.

[0102] In addition, the present application does not limit the determination method of the above entity recognition result. For example, the entity recognition result can be implemented by means of a pre-constructed operation entity recognition model. Among them, the operation entity recognition model is used to perform operation entity recognition processing on the input data of the operation entity recognition model; and the present application does not limit the implementation manner of the operation entity recognition model.

[0103] Based on the relevant content of step 31 above, for the requirement description information above, if the requirement description information does not belong to the fuzzy requirement category and the predicted output type of the requirement description information is the instruction type, it can be determined that the requirement description information belongs to the instruction subclass under the clear requirement category. Therefore, the operation entity recognition process can be further performed on the requirement description information to obtain the entity recognition result, so that subsequently, based on the entity recognition result, it can be determined which subclass under the instruction class the requirement description information belongs to, such as Figure 2 the operation instruction class or other instruction classes shown.

[0104] Step 32: Determine the requirement recognition result of the requirement description information according to the entity recognition result above.

[0105] It should be noted that the present application does not limit the implementation manner of step 32 above. For example, this step 32 can be implemented by means of a pre-constructed third mapping relationship. Among them, the third mapping relationship is used to record the requirement labels corresponding to different entity recognition results; and the third mapping relationship can be set in advance according to the actual application scenario.

[0106] In fact, in order to better improve the recognition accuracy, the present application also provides a possible implementation manner of step 32 above. In this implementation manner, this step 32 can specifically include the following steps 321-step 322.

[0107] Step 321: If the entity recognition result above indicates that there is an operation entity in the requirement description information above, determine the preset requirement label corresponding to the operation entity as the requirement recognition result of the requirement description information.

[0108] Among them, the preset requirement label corresponding to the operation entity refers to the requirement label preset for the requirement type corresponding to the operation entity. The requirement type corresponding to the operation entity is used to represent the type to which the requirement related to the operation entity belongs. For example, if the operation entity is a schedule, the requirement type corresponding to the operation entity can be a calendar processing class. If the operation entity is a document, the requirement type corresponding to the operation entity can be a document processing class.

[0109] In addition, the present application does not limit the implementation manner of the preset requirement label corresponding to the operation entity above. For example, the preset requirement label corresponding to the operation entity can be used to represent the requirement type corresponding to the operation entity, such as Figure 2Document processing classes and the like as shown. Also, for example, the preset requirement label corresponding to the operation entity can be used to indicate what means to use to process the Query under the requirement type corresponding to the operation entity. It can be seen that in one possible implementation, the preset requirement label corresponding to the operation entity can be implemented using the string "plugin for implementing the relevant processing of the operation entity", so that subsequently, based on the preset requirement label corresponding to the operation entity, it can be known that this plugin can be used to process the current Query. For example, if the operation entity is a document, the preset requirement label corresponding to the operation entity can be the string "document processing plugin".

[0110] Based on the relevant content of step 321 above, for the above requirement description information, such as instruction information similar to "create a document with the title XXX", after determining that the requirement description information belongs to the instruction category, if there is an operation entity in the requirement description information, such as an entity like a document, it can be determined that the requirement description information belongs to the operation instruction category. Therefore, the requirement recognition result of the requirement description information can be determined according to the preset requirement label corresponding to the operation entity, so that the requirement recognition result can indicate which subclass of the operation instruction category the requirement description information belongs to and / or what means to use to process the requirement description information.

[0111] Step 322: If the above entity recognition result indicates that there is no operation entity in the above requirement description information, then determine the preset instruction label as the requirement recognition result of the requirement description information.

[0112] Among them, the preset instruction label refers to the requirement label preset for other instruction categories; moreover, the present application does not limit the implementation manner of the preset instruction label. For example, the preset instruction label can be used to represent the other instruction category. Also, for example, the preset instruction label can be used to indicate what means to use to process the Query under the other instruction category. It can be seen that in one possible implementation, the preset instruction label can be implemented using the string "plugin for processing the Query under other instruction categories", so that subsequently, based on the preset instruction label, it can be known that this plugin can be used to process the current Query. Among them, the other instruction category is used to record other instructions except all instructions under the operation instruction category, such as printing and other instructions. And in some application scenarios, all instructions under the other instruction category can be processed by the same plugin; moreover, the present application does not limit this plugin. For example, the implementation manner of this plugin is similar to the implementation manner of the following document processing plugin or calendar processing plugin, etc.

[0113] Based on the relevant content of the above step 322, for the above demand description information, such as instruction information similar to "print XXX", after determining that the demand description information belongs to the instruction category, if there is no operation entity in the demand description information, it can be determined that the demand description information does not belong to the operation instruction category, and thus it can be determined that the demand description information belongs to other instruction categories. Therefore, the preset instruction tag corresponding to the other instruction category can be determined as the demand recognition result of the demand description information, so that the demand recognition result can represent the information that the demand description information belongs to other instruction categories and / or the means of processing the demand description information.

[0114] Based on the relevant content of the above steps 321 to 322, for the above demand description information, if the demand description information belongs to the instruction category, after obtaining the entity recognition result of the demand description information, the demand recognition result of the demand description information is determined according to the entity recognition result, so that the demand recognition result can at least represent which subclass of the instruction category the demand description information belongs to, such as Figure 2 the operation instruction category or other instruction categories shown, etc., so that the demand recognition result can better represent the user demand carried by the demand description information.

[0115] Based on the relevant content of the above steps 31 to 32, for the above demand description information, if the demand description information does not belong to the fuzzy demand category, it can be determined that the demand description information belongs to the clear demand category. Therefore, when determining that the predicted output type of the demand description information is the instruction type, it can be determined that the demand description information belongs to the instruction category subclass under the clear demand category. Moreover, in order to better improve the recognition accuracy, the operation entity recognition process can be further performed on the demand description information to obtain the entity recognition result; then, according to the entity recognition result, the demand recognition result of the demand description information is determined, so that the demand recognition result can represent which subclass of the instruction category the demand description information belongs to, so that the demand recognition result can better represent the user demand carried by the demand description information.

[0116] S5: Select a target module from at least one candidate processing module according to the demand recognition result of the demand description information; the target module is used to determine the feedback result of the demand description information.

[0117] Among them, the candidate processing module is used to process a certain user demand; moreover, the implementation manner of the above at least one candidate processing module is not limited in this application. For example, the at least one candidate processing module may include one or more of at least one model and at least one plug-in.

[0118] For at least one of the above models, different models are used to handle different user requirements; moreover, the present application does not limit the implementation manners of the at least one model. For example, the at least one model may include a model corresponding to multiple requirements categories, a model corresponding to no-requirement categories, and a model corresponding to question-and-answer categories. Among them, the model corresponding to the multiple requirements categories is used to handle questions under the multiple requirements categories, such as Queries with preset multiple-requirement tags, etc. The model corresponding to the no-requirement categories is used to handle questions under the no-requirement categories, such as Queries with preset no-requirement tags, etc. The model corresponding to the question-and-answer categories is used to handle questions under the question-and-answer categories, and the present application does not limit the implementation manners of the model corresponding to the question-and-answer categories. For example, the model corresponding to the question-and-answer categories may include a model corresponding to comprehensive question-and-answer categories and a model corresponding to preset object question-and-answer categories. Among them, the model corresponding to the comprehensive question-and-answer categories is used to handle questions under the comprehensive question-and-answer categories, such as Queries with preset comprehensive question-and-answer tags, etc. The model corresponding to the preset object question-and-answer categories is used to handle questions under the preset object question-and-answer categories, such as Queries with preset status question-and-answer tags, etc.

[0119] For another example, in a possible implementation manner, the working principle of the model corresponding to the question-and-answer categories may be: If the requirement recognition result of the above requirement description information indicates that the requirement description information belongs to the comprehensive question-and-answer category, the model corresponding to the question-and-answer categories may be first updated using the model parameter information corresponding to the comprehensive question-and-answer category, so that the updated model has the ability to answer questions under the comprehensive question-and-answer category, so as to subsequently use the updated model to process the requirement description information and obtain the feedback result of the requirement description information; If the requirement recognition result of the requirement description information indicates that the requirement description information belongs to the preset object question-and-answer category, the model corresponding to the question-and-answer categories may be first updated using the model parameter information corresponding to the preset object question-and-answer category, so that the updated model has the ability to answer questions under the preset object question-and-answer category, so as to subsequently use the updated model to process the requirement description information and obtain the feedback result of the requirement description information. In this way, model reuse can be achieved, which is beneficial to reducing resource consumption. Among them, the feedback result refers to the result obtained by processing the requirement description information; moreover, the present application does not limit the feedback result. For example, if the requirement description information is a Query, the feedback result may refer to the answer determined for the Query.

[0120] For at least one of the above plugins, different plugins are used to handle different user requirements; moreover, the present application does not limit the implementation manners of the at least one plugin. For example, as Figure 4 shown, the at least one plugin may include a document processing plugin, a calendar processing plugin,.... Among them, the document processing plugin is used to handle document-related requirements. The calendar processing plugin is used to handle calendar-related requirements.

[0121] The target module refers to a module selected from at least one candidate processing module and corresponding to the requirement recognition result of the above-mentioned requirement description information, so that the target module can be used to determine the feedback result of the requirement description information. Here, the feedback result refers to the information displayed on the screen determined for the requirement description information. Moreover, the implementation manner of the feedback result in this application is as follows. For example, if the requirement described by the requirement description information can be completed, the feedback result can refer to the execution result of the requirement, such as a created calendar, queried information, etc.; if the requirement cannot be completed due to incomplete parameters, the feedback result can refer to the reason why the requirement cannot be completed, such as being unable to create a calendar due to lack of time, etc., or the feedback result can refer to the suggestion given for the next round of Query, such as please provide the calendar time, etc.; if the requirement cannot be completed due to the absence of the corresponding function, the feedback result can refer to the reason why the requirement cannot be completed, such as the calendar deletion function is not supported temporarily, etc.

[0122] In addition, this application does not limit the determination process of the above-mentioned target module. For example, the target module can be determined with the help of a pre-constructed fourth mapping relationship. Here, the fourth mapping relationship is used to record the candidate processing modules corresponding to different requirements, such as Figure 4 the corresponding relationship shown, etc.

[0123] Furthermore, this application does not limit the implementation manner of the above-mentioned target module. For example, in order to better meet the user's needs, this application also provides a possible implementation manner of the target module. In this implementation manner, if the target module is a plug-in, the working principle of the target module can include Step 41 - Step 42 below.

[0124] Step 41: The target module uses the plug-in model deployed in the target module to perform configuration analysis on the requirement description information and obtains a configuration analysis result.

[0125] Here, the plug-in model is used to determine the configuration information required when the target module processes different requirements, such as tools and parameters, etc.; and the plug-in model is deployed inside the target module. In addition, this application does not limit the implementation manner of the plug-in model.

[0126] The configuration analysis result is used to represent the configuration information required when the target module processes the above-mentioned requirement description information; and this application does not limit the configuration analysis result. For example, the configuration analysis result can include a tool selection result and a parameter extraction result. Here, the tool selection result is used to represent which tool needs to be used when the target module processes the requirement description information, such as which interface to call, etc. The parameter extraction result is used to represent the parameters required when the target module processes the requirement description information, such as the parameters required to be input when calling an interface, etc.

[0127] Based on the relevant content of Step 41 above, in some application scenarios, for the above demand description information, after determining the target module based on the demand recognition result of the demand description information, if the target module is a plug-in, the target module can be called so that the target module can use the plug-in model deployed in the target module to perform configuration analysis on the demand description information and obtain a configuration analysis result, so that the configuration analysis result can be used to represent the configuration information required when the target module processes the demand description information, so that the target module can process the demand description information based on the configuration analysis result subsequently.

[0128] Step 42: The target module processes the demand description information according to the above configuration analysis result and obtains a feedback result of the demand description information.

[0129] In this application, for the above demand description information, if the demand description information needs to be processed using the target module, after the target module determines the configuration analysis result for the demand description information, the target module processes the demand description information according to the configuration analysis result and obtains a feedback result of the demand description information, so that the feedback result can represent the processing result of the target module for the demand description information.

[0130] Based on the relevant content of Step 41 to Step 42 above, for the target module determined according to the demand recognition result of the above demand description information, if the target module is a plug-in, the target module can complete tool selection processing and parameter extraction processing with the help of the plug-in module in the target module, so that the tool can subsequently complete the processing process for the demand description information based on the parameter and obtain a feedback result of the demand description information. This is conducive to better meeting user needs and thus improving the user experience.

[0131] In addition, for the above target module, in order to better improve the user experience, the target module can include a preset rejection recognition tool, such as Figure 4 the rejection recognition interface shown. Among them, the rejection recognition tool refers to a Query preset for the target module and used to respond to the situation where the target module cannot provide corresponding services; and this application does not limit the rejection recognition tool. For example, the rejection recognition tool can be implemented using the not_support interface, so that the rejection recognition tool can use the information "unable to process the current Query" as the feedback result of the above demand description information, so that the user can learn from the feedback result that the target module cannot process the Query, which can effectively avoid problems caused by giving wrong answers when the Query cannot be processed, thus being conducive to improving the user experience.

[0132] Based on the relevant content of S5 above, for the above demand description information, after obtaining the demand recognition result of the demand description information, determine the target module corresponding to the demand recognition result of the demand description information, so that the processing of the demand description information can be completed by calling the target module subsequently, and the feedback result of the demand description information can be obtained.

[0133] Based on the relevant content of S1 to S5 above, for the information processing method provided in the embodiments of the present application, after obtaining demand description information, such as a Query input by a user, first analyze whether the semantics carried by the demand description information is clear to obtain an analysis result; then, according to the analysis result, determine whether the demand description information belongs to the fuzzy demand category; if the demand description information belongs to the fuzzy demand category, determine the demand recognition result of the demand description information according to the number of demands determined from the demand description information, so that the demand recognition result can represent which subtype of the fuzzy demand category the demand description information belongs to; however, if the demand description information does not belong to the fuzzy demand category, it can be determined that the demand description information belongs to the clear demand category, so determine the demand recognition result of the demand description information according to the predicted output type of the demand description information, so that the demand recognition result can represent which subtype of the clear demand category the demand description information belongs to. In this way, the user demand can be automatically recognized, so that the target module can be selected from at least one candidate processing module according to the demand recognition result subsequently, and the target module is used to determine the feedback result of the demand description information, so that the user demand can be automatically processed.

[0134] In addition, the present application determines the demand recognition result of the demand description information through a multi-level division method for the demand description information, so that the demand recognition result can better represent the user demand described by the demand description information. This is beneficial to improving the recognition accuracy, thus beneficial to improving the feedback accuracy for user questions, and further beneficial to improving the user's question-and-answer experience.

[0135] In addition, since the demand recognition result of the above demand description information can accurately represent the user demand described by the demand description information, the target module selected based on the demand recognition result is more suitable for processing the user request, so that the feedback result determined for the demand description information using the target module is more accurate. This is beneficial to improving the demand feedback effect.

[0136] Furthermore, this application does not limit the execution subject of the information processing method provided in the embodiments of this application. For example, the information processing method provided in the embodiments of this application can be applied to a terminal device or a server. Also, for example, the information processing method provided in the embodiments of this application can also be implemented by means of the data interaction process between the terminal device and the server. Among them, the terminal device can be a smart phone, a computer, a personal digital assistant (Personal Digital Assitant, PDA), a tablet computer, etc. The server can be an independent server, a cluster server or a cloud server.

[0137] Moreover, for some application scenarios, such as a multi-round dialogue scenario, the requirements described in the user question of the current round may be the same as those involved in the historical question-and-answer process, or may be a new requirement that is completely different from those involved in the historical question-and-answer process. Based on this, in order to better improve the recognition accuracy, this application also provides a possible implementation manner of the above information processing method. In this implementation manner, the information processing method not only includes the above S1-S4, but may also include step 51 below.

[0138] Step 51: If the correlation representation data between the above requirement description information and the historical requirement information exceeds the preset correlation threshold, then update the requirement recognition result of the requirement description information according to the requirement recognition result of the historical requirement information.

[0139] The historical requirement information refers to the historical information required for requirement recognition processing of the above requirement description information, such as the previous rounds of question and answer, etc.

[0140] In addition, this application does not limit the implementation manner of the above historical requirement information. For example, if the above requirement description information refers to the question input by the user in the Mth round of dialogue, then the historical requirement information may include part or all of the questions and answers in the first round of dialogue, the questions and answers in the second round of dialogue,... (and so on), and the questions and answers in the (M - 1)th round of dialogue. Here, M is a positive integer and M≥2.

[0141] In addition, for the above historical requirement information, the requirement recognition result of the historical requirement information refers to the user requirement determined for the historical requirement information, so that the requirement recognition result of the historical requirement information can represent the historical requirement corresponding to the above requirement description information.

[0142] Furthermore, the present application does not limit the manner of obtaining the demand recognition result of the above historical demand information. For example, it may refer to the result obtained by performing demand recognition processing on the problem in the historical demand information within a historical time period. Another example is that the demand recognition result of the historical demand information can be obtained by processing the historical demand information using a pre-constructed demand recognition model. Among them, the demand recognition model is used to perform intention recognition processing on the input data of the demand recognition model; and the present application does not limit the demand recognition model. For example, the working principle of the demand recognition model is similar to the demand recognition process shown in S1-S4 above.

[0143] The relevance characterization data is used to characterize the degree of association between the demand description information and the historical demand information. For example, the relevance characterization data can be implemented using Figure 3 the shown relevance score.

[0144] In addition, the present application does not limit the determination process of the above relevance characterization data. For example, the relevance characterization data can be determined by means of a pre-constructed relevance determination model. Among them, the relevance determination model is used to perform relevance determination processing on the input data of the relevance determination model; and the present application does not limit the implementation manner of the relevance determination model.

[0145] The preset relevance threshold refers to the lowest value of relevance preset for when multiple rounds of conversations continue the same demand; and the preset relevance threshold can be set according to the actual application scenario. For example, the preset relevance threshold can be 0.5.

[0146] Based on the relevant content of step 51 above, for the above demand description information, after obtaining the demand recognition result of the demand description information, if the relevance characterization data between the demand description information and the historical demand information exceeds the preset relevance threshold, it can be determined that the demand description information continues the historical demand. Therefore, the demand recognition result of the historical demand information can be directly used to update the demand recognition result of the demand description information, so that the updated demand recognition result of the demand description information is consistent with the demand recognition result of the historical demand information, which is beneficial to improving the recognition accuracy.

[0147] In fact, in some application scenarios, in order to better improve efficiency, the present application also provides a possible implementation manner of the above information processing method. In this implementation manner, the information processing method not only includes S1-S4 above, but may also include the following steps 61-63.

[0148] Step 61: Compare and process the demand recognition result of the above demand description information and the demand recognition result of the historical demand information to obtain a comparison result.

[0149] In this application, for the above-mentioned requirement description information, after obtaining the requirement recognition result of the requirement description information, the requirement recognition result of the requirement description information can be compared with the requirement recognition result of the historical requirement information to obtain a comparison result, so that the comparison result can indicate whether the requirement recognition result of the requirement description information is the same as the requirement recognition result of the historical requirement information.

[0150] Step 62: If the above comparison result indicates that the requirement recognition result of the requirement description information is different from the requirement recognition result of the historical requirement information, then determine whether the correlation representation data between the requirement description information and the historical requirement information exceeds a preset correlation threshold.

[0151] In this application, for the above-mentioned requirement description information, if the requirement recognition result of the requirement description information is different from the requirement recognition result of the historical requirement information, then further determine whether the correlation representation data between the requirement description information and the historical requirement information exceeds a preset correlation threshold, so as to be able to determine whether the current round still continues the historical requirement based on the judgment result subsequently.

[0152] Step 63: If the above correlation representation data exceeds the preset correlation threshold, then update the requirement recognition result of the requirement description information according to the requirement recognition result of the historical requirement information.

[0153] In this application, for the above-mentioned requirement description information, if the requirement recognition result of the requirement description information is different from the requirement recognition result of the historical requirement information, and the correlation representation data between the requirement description information and the historical requirement information exceeds the preset correlation threshold, then it can be determined that the requirement description information should continue the historical requirement. Thus, it can be determined that the requirement recognition result of the requirement description information cannot accurately represent the current user requirement. Therefore, the requirement recognition result of the historical requirement information can be directly used to update the requirement recognition result of the requirement description information, so that the updated requirement recognition result of the requirement description information is consistent with the requirement recognition result of the historical requirement information. This is beneficial to improving the recognition accuracy.

[0154] Based on the relevant content of the above Steps 61 to 63, for some application scenarios, such as Figure 3For the scenario shown, after obtaining the requirement recognition result of the above-mentioned requirement description information, first determine whether the requirement recognition result of the requirement description information is the same as the requirement recognition result of the historical requirement information. If they are the same, it can be determined that the current round still continues the historical requirement, and the requirement recognition result of the requirement description information can accurately represent the current user requirement. Therefore, there is no need to perform correlation calculation, and corresponding operations can be directly performed based on the requirement recognition result of the requirement description information, which is beneficial to improving efficiency. If they are different, it can be further determined whether the correlation representation data between the requirement description information and the historical requirement information exceeds the preset correlation threshold. If it exceeds, it can be determined that the current round still continues the historical requirement, but the requirement recognition result of the requirement description information cannot accurately represent the current user requirement. Therefore, the requirement recognition result of the historical requirement information can be directly used to update the requirement recognition result of the requirement description information, so that the updated requirement recognition result of the requirement description information is consistent with the requirement recognition result of the historical requirement information, so that corresponding operations can be directly performed based on the updated requirement recognition result of the requirement description information in the future. If it does not exceed, it can be determined that a new requirement is proposed in the current round, and the requirement recognition result of the requirement description information can accurately represent the user requirement, so that corresponding operations can be directly performed based on the requirement recognition result of the requirement description information in the future, which is beneficial to improving the recognition accuracy.

[0155] Actually, in some application scenarios, in order to better improve the user experience, the present application also provides a possible implementation manner of the above-mentioned information processing method. In this implementation manner, the information processing method may at least include step 71 below.

[0156] Step 71: Display the above feedback result.

[0157] Based on the relevant content of step 71 above, for the above-mentioned requirement description information, after processing the requirement description information through the target module corresponding to the requirement recognition result of the requirement description information to obtain a feedback result, the feedback result can be displayed to the user so that the user can view the feedback result, which can meet the user requirement.

[0158] Based on the relevant content of the information processing method provided by the present application, the technical solution provided by the present application has the following advantages shown in (1) to (6).

[0159] (1) The technical solution provided by the present application obtains the requirement recognition result of the Query by means of multi-level division processing of the Query input by the user, so that the requirement recognition result can more accurately describe the user requirement, which is beneficial to better meeting the user requirement and further beneficial to improving the user experience.

[0160] (2) The technical solution provided by this application solves the problem of multi-round context relevance requirements through the comprehensive processing of multi-round Q&A information. This is beneficial to improving the recognition accuracy, thus better meeting the user's needs, and further improving the user experience.

[0161] (3) The technical solution provided by this application realizes tool selection processing and parameter extraction processing through a plug-in model. In this way, it can better determine the processing tool for the Query with the help of the semantic understanding ability of the model, which is beneficial to improving the processing effect of the Query, and further improving the user experience.

[0162] (4) In the technical solution provided by this application, the requirement recognition process and the tool selection process are decoupled from each other, so that the plug-in modules involved in the tool selection process can be dynamically plugged and unplugged. This can effectively avoid the defects caused by the mutual interference between the requirement recognition process and the tool selection process.

[0163] (5) In the technical solution provided by this application, each plug-in is an independent module, and each plug-in can realize tool selection processing and parameter extraction processing with the help of the plug-in model deployed inside it, so that each plug-in can be independently optimized. Independent optimization can effectively reduce the influence between the configuration information determination processes in different plug-ins, which is beneficial to improving the parsing correctness of the Query, and further effectively avoiding the overfitting phenomenon.

[0164] (6) In the technical solution provided by this application, each plug-in is internally configured with a rejection recognition ability, and the rejection recognition ability is abstracted into an empty interface function. This can unify the output format of each plug-in and provide the ability of dynamic rejection recognition copywriting.

[0165] Based on the content shown in (1) to (6) above, the technical solution provided by this application can accurately recognize requirements and provide guarantee for subsequent high-quality responses.

[0166] Based on the information processing method provided by the embodiments of this application, the embodiments of this application also provide an information processing device. The following is combined with Figure 5 for explanation and illustration. Among them, Figure 5 is a schematic structural diagram of an information processing device provided by the embodiments of this application. It should be noted that for the technical details of the information processing device provided by the embodiments of this application, please refer to the relevant content of the above information processing method.

[0167] As Figure 5 shown, the information processing device 500 provided by the embodiments of this application includes:

[0168] A semantic analysis unit 501, configured to generate an analysis result according to whether the semantics of the requirement description information is clear after obtaining the requirement description information;

[0169] A first determination unit 502, configured to determine whether the requirement description information belongs to the fuzzy requirement category according to the analysis result;

[0170] A second determination unit 503, configured to, in response to the requirement description information belonging to the fuzzy requirement category, determine a requirement recognition result of the requirement description information according to the number of requirements determined from the requirement description information;

[0171] A third determination unit 504, configured to, in response to the requirement description information not belonging to the fuzzy requirement category, determine a requirement recognition result of the requirement description information according to the predicted output type of the requirement description information; the predicted output type is obtained by performing output type prediction processing on the requirement description information;

[0172] A module selection unit 505, configured to select a target module from at least one candidate processing module according to the requirement recognition result of the requirement description information; the target module is used to determine a feedback result of the requirement description information.

[0173] In a possible implementation manner, the information processing device 500 is deployed on a client;

[0174] The third determination unit 504 includes:

[0175] An analysis subunit, configured to analyze a query object of the requirement description information if the predicted output type is a text type;

[0176] A first determination subunit, configured to determine a requirement recognition result of the requirement description information according to the query object.

[0177] In a possible implementation manner, the first determination subunit is specifically configured to: if the query object is a preset object, determine a preset status question-and-answer label as the requirement recognition result of the requirement description information; if the query object is not a preset object, determine a preset comprehensive question-and-answer label as the requirement recognition result of the requirement description information.

[0178] In a possible implementation manner, the third determination unit 504 includes:

[0179] An entity recognition subunit, configured to perform operation entity recognition processing on the requirement description information if the predicted output type is an instruction type, and obtain an entity recognition result;

[0180] A second determination subunit, configured to determine a requirement recognition result of the requirement description information according to the entity recognition result.

[0181] In a possible implementation manner, the second determination subunit is specifically configured to: if the entity recognition result indicates that there is an operation entity in the requirement description information, determine the preset requirement tag corresponding to the operation entity as the requirement recognition result of the requirement description information; if the entity recognition result indicates that there is no operation entity in the requirement description information, determine the preset instruction tag as the requirement recognition result of the requirement description information.

[0182] In a possible implementation manner, the second determination unit 503 is specifically configured to: if the number of requirements is 0, determine the preset no-requirement tag as the requirement recognition result of the requirement description information; if the number of requirements is greater than 1, determine the preset multi-requirement tag as the requirement recognition result of the requirement description information.

[0183] In a possible implementation manner, the information processing device 500 further includes:

[0184] A requirement update unit, configured to update the requirement recognition result of the requirement description information according to the requirement recognition result of the historical requirement information if the relevance characterization data between the requirement description information and the historical requirement information exceeds a preset relevance threshold.

[0185] In a possible implementation manner, the information processing device 500 further includes:

[0186] A requirement comparison unit, configured to perform a comparison process on the requirement recognition result of the requirement description information and the requirement recognition result of the historical requirement information to obtain a comparison result;

[0187] A relevance determination unit, configured to determine whether the relevance characterization data between the requirement description information and the historical requirement information exceeds a preset relevance threshold if the comparison result indicates that the requirement recognition result of the requirement description information is different from the requirement recognition result of the historical requirement information.

[0188] In a possible implementation manner, the information processing device 500 further includes:

[0189] A feedback display unit, configured to display the feedback result.

[0190] In a possible implementation manner, the target module is a plug-in; the target module is specifically configured to: perform configuration analysis on the requirement description information by using a plug-in model deployed in the target module to obtain a configuration analysis result; process the requirement description information according to the configuration analysis result to obtain the feedback result.

[0191] In a possible implementation manner, the configuration analysis result includes a tool selection result and a parameter extraction result.

[0192] Based on the relevant content of the above information processing device 500, for the information processing device 500 provided in the embodiments of the present application, after obtaining the requirement description information, such as the Query input by the user, first analyze whether the semantics of the requirement description information is clear to obtain an analysis result; then, according to the analysis result, determine whether the requirement description information belongs to the fuzzy requirement category; if the requirement description information belongs to the fuzzy requirement category, determine the requirement recognition result of the requirement description information according to the number of requirements determined from the requirement description information, so that the requirement recognition result can indicate which subtype the requirement description information belongs to under the fuzzy requirement category; however, if the requirement description information does not belong to the fuzzy requirement category, it can be determined that the requirement description information belongs to the clear requirement category, so determine the requirement recognition result of the requirement description information according to the predicted output type of the requirement description information, so that the requirement recognition result can indicate which subtype the requirement description information belongs to under the clear requirement category. In this way, the user requirements can be automatically recognized, so that subsequently, according to the requirement recognition result, a target module can be selected from at least one candidate processing module, so that the target module is used to determine the feedback result of the requirement description information, and thus the user requirements can be automatically processed.

[0193] In addition, since the information processing device 500 determines the requirement recognition result of the requirement description information through a multi-level division method for the requirement description information, so that the requirement recognition result can better represent the user requirements described by the requirement description information, which is beneficial to improving the recognition accuracy, and thus beneficial to improving the feedback accuracy for user questions, and further beneficial to improving the user's question-and-answer experience.

[0194] In addition, since the requirement recognition result of the above requirement description information can accurately represent the user requirements described by the requirement description information, so that the target module selected based on the requirement recognition result is more suitable for processing the user request, and thus the feedback result determined by using the target module for the requirement description information is more accurate, which is beneficial to improving the requirement feedback effect.

[0195] Furthermore, the embodiments of the present application also provide an electronic device, the device includes a processor and a memory: the memory is used to store instructions or computer programs; the processor is used to execute the instructions or computer programs in the memory, so that the electronic device executes any implementation manner of the information processing method provided in the embodiments of the present application.

[0196] See Figure 6, which shows a schematic structural diagram of an electronic device 600 suitable for implementing the embodiments of the present disclosure. The terminal devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The electronic device shown is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present disclosure.

[0197] As Figure 6 shown, the electronic device 600 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 601, which may perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 602 or the program loaded from the storage device 608 into the random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. The input / output (I / O) interface 605 is also connected to the bus 604.

[0198] Generally, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or wiredly to exchange data. Although Figure 6 the electronic device 600 with various devices is shown, it should be understood that it is not required to implement or have all the shown devices. More or fewer devices may be alternatively implemented or had.

[0199] Particularly, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, the embodiments of the present disclosure include a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the methods of the embodiments of the present disclosure are executed.

[0200] The electronic device provided by the embodiments of the present disclosure belongs to the same inventive concept as the method provided by the above embodiments. For technical details not described in detail in this embodiment, reference may be made to the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0201] An embodiment of the present application further provides a computer-readable medium, in which instructions or computer programs are stored. When the instructions or computer programs run on a device, the device is caused to execute any implementation manner of the information processing method provided by the embodiments of the present application.

[0202] It should be noted that the computer-readable medium in the present disclosure may be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or combined with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, and the computer-readable signal medium may send, propagate, or transmit a program for use by or combined with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium may be transmitted by any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0203] In some embodiments, the client and the server can communicate using any currently known or future-developed network protocol such as HTTP (Hyper Text Transfer Protocol), and can be interconnected with digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0204] The above computer-readable medium can be included in the above electronic device; it can also exist separately without being assembled into the electronic device.

[0205] The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by the electronic device, the electronic device can execute the above method.

[0206] Computer program code for performing the operations of the present disclosure can be written in one or more programming languages or combinations thereof. The above programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., by using an Internet service provider to connect through the Internet).

[0207] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur in a different order than that noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or by a combination of dedicated hardware and computer instructions.

[0208] The units involved in the embodiments described in the present disclosure can be implemented in a software manner or in a hardware manner. Among them, the name of the unit / module does not constitute a limitation on the unit itself in some cases.

[0209] The functions described above in this document can be at least partially performed by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that can be used include: Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), Systems on Chip (SOCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0210] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a Random Access Memory (RAM), a Read Only Memory (ROM), an Erasable Programmable Read Only Memory (EPROM or Flash Memory), an optical fiber, a portable Compact Disc Read Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0211] It should be noted that the embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other. For the systems or devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple. For the relevant parts, reference can be made to the description in the method section.

[0212] It should be understood that in this application, "at least one (item)" means one or more, and "multiple" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally means that the associated objects before and after are in an "or" relationship. "At least one (one)" or its similar expression means any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0213] It should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to this process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0214] The steps of the method or algorithm described in combination with the embodiments disclosed in this article can be directly implemented by hardware, software modules executed by a processor, or a combination of both. The software module can be placed in a random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0215] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not intended to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An information processing method, characterized in that: The method comprises: After obtaining the requirement description information, generating an analysis result based on whether the semantics of the requirement description information is clear; Determining whether the requirement description information belongs to a fuzzy requirement class based on the analysis result; In response to the requirement description information belonging to a fuzzy requirement class, determining a requirement identification result of the requirement description information based on the number of requirements determined from the requirement description information; In response to the requirement description information not belonging to the fuzzy requirement class, determining a requirement recognition result of the requirement description information according to a predicted output type of the requirement description information; the predicted output type is obtained by performing output type prediction processing on the requirement description information; A target module is selected from at least one candidate processing module according to a requirement identification result of the requirement description information; the target module is used to determine a feedback result of the requirement description information.

2. The method according to claim 1, characterized in that The determining, based on the predicted output type corresponding to the demand description information, a demand identification result of the demand description information includes: If the predicted output type is a text type, analyzing the query object of the demand description information; Determine a requirement identification result of the requirement description information based on the query object.

3. The method according to claim 2, characterized in that Determining the requirement identification result of the requirement description information based on the query object includes: If the query object is a preset object, the preset status question and answer tag is determined as the demand identification result of the demand description information; If the query object is not a preset object, the preset comprehensive question and answer tag is determined as the demand identification result of the demand description information.

4. The method according to claim 1, wherein The determining, based on the predicted output type corresponding to the demand description information, a demand identification result of the demand description information includes: If the predicted output type is an instruction type, performing operation entity recognition processing on the requirement description information to obtain an entity recognition result; Determine the requirement recognition result of the requirement description information based on the entity recognition result.

5. The method according to claim 4, characterized in that Determining the requirement recognition result of the requirement description information based on the entity recognition result includes: If the entity recognition result indicates that an operation entity exists in the requirement description information, determining the preset requirement tag corresponding to the operation entity as the requirement recognition result of the requirement description information; If the entity recognition result indicates that the operation entity does not exist in the requirement description information, the preset instruction tag is determined as the requirement recognition result of the requirement description information.

6. The method according to claim 1, wherein Determining the requirement identification result of the requirement description information based on the number of requirements determined from the requirement description information includes: If the number of requirements is 0, a no-requirement label is preset and determined as the requirement identification result of the requirement description information; If the number of requirements is greater than 1, multiple requirement tags will be preset and determined as the requirement identification result of the requirement description information.

7. The method according to claim 1, characterized in that After determining the requirement identification result of the requirement description information, the method further includes: If the correlation representation data between the demand description information and the historical demand information exceeds a preset correlation threshold, the demand identification result of the demand description information is updated according to the demand identification result of the historical demand information.

8. The method according to claim 7, characterized in that The method further comprises: Comparing the demand identification result of the demand description information with the demand identification result of the historical demand information to obtain a comparison result; If the comparison result indicates that the demand identification result of the demand description information is different from the demand identification result of the historical demand information, it is determined whether the correlation representation data between the demand description information and the historical demand information exceeds a preset correlation threshold.

9. The method according to claim 1, characterized in that The target module is a plug-in; The target module is specifically used to: perform configuration analysis on the requirement description information using the plug-in model deployed in the target module to obtain a configuration analysis result; and process the requirement description information according to the configuration analysis result to obtain the feedback result.

10. The method according to claim 9, characterized in that The configuration analysis results include tool selection results and parameter extraction results.

11. An information processing device, characterized in that: include: A semantic analysis unit, configured to generate an analysis result based on whether the semantics of the requirement description information are clear after obtaining the requirement description information; A first determining unit, configured to determine whether the requirement description information belongs to a fuzzy requirement class based on the analysis result; a second determining unit, configured to determine, in response to the requirement description information belonging to a fuzzy requirement class, a requirement identification result of the requirement description information based on the number of requirements determined from the requirement description information; a third determining unit, configured to, in response to the requirement description information not belonging to the fuzzy requirement class, determine a requirement recognition result of the requirement description information based on a predicted output type of the requirement description information; the predicted output type is obtained by performing output type prediction processing on the requirement description information; The module selection unit is used to select a target module from at least one candidate processing module according to the requirement identification result of the requirement description information; the target module is used to determine the feedback result of the requirement description information.

12. An electronic device, characterized in that: The device includes: a processor and a memory; The memory is used to store instructions or computer programs; The processor is configured to execute the instructions or computer program in the memory, so that the electronic device executes the method according to any one of claims 1 to 10.

13. A computer-readable medium, characterized in that The computer-readable medium stores instructions or a computer program, and when the instructions or the computer program are executed on a device, the device is caused to execute the method according to any one of claims 1 to 10.

14. A computer program product, characterized in that The method comprises a computer program carried on a non-transitory computer-readable medium, the computer program comprising a program code for executing the method according to any one of claims 1 to 10.

Images