Question answering method and device, storage medium and electronic equipment

Abstract

The present disclosure relates to a question answering method, device, storage medium and electronic equipment. The question answering method comprises: performing feature extraction on a question to be answered to obtain first question feature data; matching the first question feature data with a preset question feature template; when no first question feature template corresponding to the first question feature data is matched, performing data dimension reduction processing on the first question feature data to obtain second question feature data; matching the second question feature data with a dimension-reduced question feature template to obtain a target question feature template matched with the second question feature data; determining a target answering step corresponding to the target question feature template; and generating a target answer corresponding to the question to be answered according to the target answering step. Through the present disclosure, the coverage and generalization of question answering can be improved.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more specifically, to a problem-solving method, apparatus, storage medium, and electronic device. Background Technology

[0002] With the rapid development of internet technology, the use of intelligent devices to replace customer service personnel in specific question-and-answer or dialogue scenarios to communicate with customers and provide them with the information they need has become very common. For example, a machine receives a human question, performs semantic understanding, and obtains the answer by searching a semantic template; then, the machine organizes the answer into a language form suitable for human communication and provides the answer to the human question.

[0003] However, how to improve the coverage and generalization of responses to user questions in human-computer dialogue scenarios has become an urgent problem to be solved. Summary of the Invention

[0004] This summary section is provided to briefly introduce the concepts, which will be described in detail in the detailed description section below. This summary section is not intended to identify key or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0005] Firstly, this disclosure provides a method for answering questions, including:

[0006] Feature extraction is performed on the question to be answered to obtain the feature data of the first question;

[0007] The first problem feature data is matched with the preset problem feature template;

[0008] When no first problem feature template corresponding to the first problem feature data is matched, the first problem feature data is subjected to data dimensionality reduction processing to obtain the second problem feature data.

[0009] The second problem feature data and the dimensionality-reduced problem feature template are matched to obtain a target problem feature template that matches the second problem feature data, wherein the dimensionality-reduced problem feature template is obtained by data dimensionality reduction processing of the problem feature template;

[0010] Determine the target solution steps corresponding to the target problem feature template;

[0011] Based on the target solution steps, the target answer corresponding to the question to be answered is generated.

[0012] Secondly, this disclosure provides a problem-solving apparatus, comprising:

[0013] The feature extraction module is used to extract features from the problem to be solved in order to obtain the first problem feature data;

[0014] The first matching module is used to match the first problem feature data with a preset problem feature template;

[0015] The processing module is used to perform data dimensionality reduction processing on the first problem feature data to obtain the second problem feature data when no first problem feature template corresponding to the first problem feature data is matched.

[0016] The second matching module is used to match the second problem feature data and the dimensionality-reduced problem feature template to obtain a target problem feature template that matches the second problem feature data, wherein the dimensionality-reduced problem feature template is obtained by data dimensionality reduction processing of the problem feature template;

[0017] The determination module is used to determine the target solution steps corresponding to the target problem feature template;

[0018] The solution module is used to generate the target answer corresponding to the question to be solved based on the target solution steps.

[0019] Thirdly, this disclosure provides a computer-readable medium having a computer program stored thereon, which, when executed by a processing device, implements the steps of the problem-solving method provided in the first aspect of this disclosure.

[0020] Fourthly, this disclosure provides an electronic device, comprising:

[0021] A storage device on which computer programs are stored;

[0022] A processing device is configured to execute the computer program in the storage device to implement the steps of the problem-solving method provided in the first aspect of this disclosure.

[0023] The above technical solution matches the problem feature data of the question to be answered with a preset problem feature template. When no first problem feature template corresponding to the first problem feature data is found, the first problem feature data is dimensionality reduced to obtain the second problem feature data. Matching is then performed based on the dimensionality-reduced problem feature data and the dimensionality-reduced problem feature template. This eliminates the need to completely match all feature data included in the first problem feature data with the feature data included in the problem feature template, thus improving the generalization and coverage of feature matching, which in turn improves the generalization and coverage of question answering. Furthermore, when the problem to be answered includes a large amount of problem feature data, matching based on the dimensionality-reduced problem feature data and the dimensionality-reduced problem feature template can improve the efficiency of feature matching while ensuring the accuracy of feature data matching, thereby improving the efficiency of question answering and reducing the complexity of question answering.

[0024] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0025] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale. In the drawings:

[0026] Figure 1 This is a flowchart illustrating a problem-solving method according to an exemplary embodiment of the present disclosure.

[0027] Figure 2 This is a block diagram illustrating a problem-solving apparatus according to an exemplary embodiment of the present disclosure.

[0028] Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure. Detailed Implementation

[0029] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0030] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0031] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0032] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0033] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0034] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0035] All actions involving the acquisition of signals, information, or data in this disclosure are carried out in accordance with the relevant data protection laws and policies of the country where the location is situated, and with the authorization granted by the owner of the relevant device.

[0036] Figure 1 This is a flowchart illustrating a problem-solving method according to an exemplary embodiment, such as... Figure 1 As shown, the solution to the problem includes the following steps.

[0037] In step S11, feature extraction is performed on the problem to be solved to obtain the first problem feature data.

[0038] In this disclosure, the questions to be answered can be text data, image data, or voice data input by the user, and this disclosure does not impose any restrictions.

[0039] In one embodiment, feature extraction of the question to be solved may include extracting intent features included in the question to be solved to obtain first question feature data including intent feature data.

[0040] Alternatively, for a problem to be solved that includes a formula, feature extraction can be performed on the problem to be solved. This can include extracting the intent features included in the problem to be solved and the parameter features included in the formula, to obtain first problem feature data that includes intent feature data and parameter feature data of the parameters included in the formula.

[0041] When the first problem feature data includes intent feature data, the intent feature data also includes multiple intent sub-feature data. The intent feature data is obtained by combining each intent sub-feature data in a chain structure.

[0042] When the problem to be solved includes a formula, and the first problem feature data also includes parameter feature data of the parameters contained in the formula, the parameter feature data can be obtained by normalizing the parameters in the formula according to the corresponding category rules.

[0043] For example, the first problem feature data can be extracted by a pre-trained problem feature extraction model, which is trained by using the problem feature annotation data corresponding to the sample problem as model constraints.

[0044] For example, the problem to be solved is: "Find the equation of the line passing through the point (2, -10) and perpendicular to the line 2x + y = 5". The problem is input into a pre-trained problem feature extraction model, which extracts the first problem feature data and outputs the following:

[0045] Intent: To determine the slope from the positional relationship between lines; to find the linear function expression based on the slope-intercept-point ratio; and to convert the linear expression format. Parameters: function_formula: 2x + y = 5; function_output_format: point-slope form; function_coordinate: (2, -10); function_position_relationship: perpendicular.

[0046] Next, based on the intent feature data and the parameter feature data, the problem feature data is obtained. Specifically, the parameter feature data can be obtained by normalizing the parameters in the formula according to their corresponding categories (or names): that is, the parameter feature data is "function_formula; function_output_format; function_coordinate; function_position_relationship". This parameter feature data can represent the function formula, the function output format, the function coordinates, and the function position relationship, respectively.

[0047] Furthermore, the intent feature data and parameter feature data can be concatenated to obtain the first problem feature data, namely: "Calculate the slope from the positional relationship between lines; know the slope-intercept-point; find the analytical expression of the linear function; convert the linear expression format; function_formula; function_output_format; function_coordinate; function_position_relationship".

[0048] In step S12, the first problem feature data is matched with the preset problem feature template.

[0049] After obtaining the problem feature data of the question to be answered, the problem feature data of the question to be answered is matched with the preset problem feature template. Specifically, all feature data included in the first problem feature data can be matched one-to-one with the feature data included in the problem feature template to obtain the solution steps corresponding to the question to be answered. Then, based on the solution steps corresponding to the question to be answered, the target answer corresponding to the question to be answered is generated.

[0050] The problem feature template can be a pre-built template that includes intent feature data, or a template that includes intent feature data and parameter feature data including the parameters contained in the formula. Different problem feature templates are associated with corresponding solution steps.

[0051] In step S13, if no first problem feature template corresponding to the first problem feature data is matched, the first problem feature data is subjected to data dimensionality reduction processing to obtain the second problem feature data.

[0052] Currently, when matching the problem feature data of the question to be answered with the preset problem feature template, it is necessary to completely match all feature data included in the first problem feature data with the feature data included in the problem feature template in order to obtain the solution steps corresponding to the question to be answered, and then generate the target answer corresponding to the question to be answered based on the solution steps corresponding to the question to be answered.

[0053] However, in practical applications, on the one hand, when the problem to be solved includes a large amount of problem feature data, matching the problem feature data to be solved one by one with the feature data included in the problem feature template, and the existing method of precisely matching the problem feature data to be solved with the preset problem feature template, will result in low matching efficiency and reduce the accuracy of feature data matching.

[0054] On the other hand, it often happens that when some of the question's feature data matches some of the features in the question feature template, the solution steps for that question can be obtained. Therefore, the current method of completely matching the question's feature data with a preset question feature template suffers from several drawbacks in human-computer dialogue scenarios: low coverage of responses to user questions, weak generalization of responses to user questions, low matching efficiency, and low accuracy of feature data matching.

[0055] Therefore, in order to improve the coverage and generalization of responses to user questions and reduce the difficulty of matching question feature data and question feature templates, this disclosure can perform data dimensionality reduction on the first question feature data to obtain the second question feature data when no first question feature template corresponding to the first question feature data is matched.

[0056] In step S14, the second problem feature data and the dimensionality reduction problem feature template are matched to obtain the target problem feature template that matches the second problem feature data.

[0057] In this disclosure, the preset problem feature template can be pre-processed to reduce the data dimensionality, thereby obtaining a dimensionality-reduced problem feature template.

[0058] The method for performing dimensionality reduction on the first problem feature data is the same as the method for performing dimensionality reduction on the problem feature template.

[0059] In one implementation, the first problem feature data can be dimensionality reduced to obtain the second problem feature data, for example, in the following manner:

[0060] The first problem feature data is subjected to local sensitive hashing to obtain the target hash data of the first problem feature data, and the target hash data is determined as the second problem feature data.

[0061] Locality-Sensitive Hashing (LSH) refers to a series of hashing processes used to map data from a high-dimensional data space to a low-dimensional data space. The basic idea of ​​LSH is that when two adjacent data points in a high-dimensional data space are mapped to a low-dimensional data space, the mapped data will have a high probability of remaining adjacent. Conversely, when two non-adjacent data points in a high-dimensional data space are mapped to a low-dimensional data space, the mapped data will have a high probability of remaining non-adjacent.

[0062] Specifically, by performing local sensitive hashing on the first problem feature data, which includes problem feature data, we can obtain the target hash data of the first problem feature data after dimensionality reduction, for example, represented as a string. The target hash data is then determined as the second problem feature data.

[0063] In this disclosure, due to the use of locality-sensitive hashing, the problem feature data that was originally highly similar to the dimensionality-reduced problem feature template remains highly similar, and the problem feature data that was originally low in similarity to the dimensionality-reduced problem feature template remains low in similarity.

[0064] In step S15, the target solution steps corresponding to the target problem feature template are determined.

[0065] In this disclosure, the first problem feature data is subjected to data dimensionality reduction processing to obtain the second problem feature data. The second problem feature data is matched with the dimensionality-reduced problem feature template to obtain the target problem feature template that matches the second problem feature data. Then, the target solution steps corresponding to the target problem feature template can be determined from the pre-stored solution steps corresponding to the problem feature template.

[0066] In step S16, the target answer corresponding to the question to be answered is generated according to the target solution steps.

[0067] After determining the target solution steps corresponding to the target problem feature template, the problem to be solved can be solved or calculated step by step according to the target solution steps to generate the target answer corresponding to the problem to be solved.

[0068] For example, the target solution steps and the question to be answered can be input into the question-answering engine. The question-answering engine will then execute the first, second, third, and so on steps of the target solution steps in sequence until all steps of the target solution steps are completed, generating the target answer corresponding to the question to be answered.

[0069] For example, the problem to be solved is: "Find the equation of the line passing through the point (2, -10) and perpendicular to the line 2x + y = 5". The steps to obtain the solution to this problem are: ① Find the slope of the line; ② Find the perpendicular slope of the line; ③ Obtain the point-slope form based on the point and the slope of the line; ④ Simplify the point-slope form to obtain the general form.

[0070] Input the target solution steps and the question to be solved into the problem-solving engine. The engine executes the first step of the target solution steps to obtain the slope of the line: slope = -2; the second step yields the vertical slope: slope = 1 / 2; the third step gives the point-slope form: y - (-10) = 1 / 2(x - 2); and the fourth step derives the simplified general form: y = -1 / 2x - 11. Therefore, the target answer is y = -1 / 2x - 11.

[0071] In an exemplary embodiment of this disclosure, the problem feature data of the question to be answered is matched with a preset problem feature template. When no first problem feature template corresponding to the first problem feature data is matched, the first problem feature data is subjected to data dimensionality reduction processing to obtain second problem feature data. Then, matching is performed based on the dimensionality-reduced problem feature data and the dimensionality-reduced problem feature template. It is not necessary to completely match all feature data included in the first problem feature data with the feature data included in the problem feature template, which improves the generalization and coverage of feature matching, that is, improves the generalization and coverage of question answering. Furthermore, when the problem to be answered includes a large scale of problem feature data, matching based on the dimensionality-reduced problem feature data and the dimensionality-reduced problem feature template can improve the efficiency of feature matching while ensuring the accuracy of feature data matching, thereby improving the efficiency of question answering and reducing the complexity of question answering.

[0072] Figure 2 This is a block diagram illustrating a problem-solving apparatus 200 according to an exemplary embodiment. (Refer to...) Figure 2 The problem-solving device includes a feature extraction module 201, a first matching module 202, a processing module 203, a second matching module 204, a determination module 205, and a solution module 206.

[0073] Among them, the feature extraction module 201 is used to extract features from the question to be answered in order to obtain the first question feature data;

[0074] The first matching module 202 is used to match the first problem feature data with a preset problem feature template;

[0075] Processing module 203 is used to perform data dimensionality reduction processing on the first problem feature data to obtain the second problem feature data when no first problem feature template corresponding to the first problem feature data is matched.

[0076] The second matching module 204 is used to match the second problem feature data and the dimensionality-reduced problem feature template to obtain a target problem feature template that matches the second problem feature data, wherein the dimensionality-reduced problem feature template is obtained after the problem feature template has undergone data dimensionality reduction processing;

[0077] The determining module 205 is used to determine the target solution steps corresponding to the target problem feature template;

[0078] The solution module 206 is used to generate the target answer corresponding to the question to be solved based on the target solution steps.

[0079] Optionally, the method for performing data dimensionality reduction on the first problem feature data is the same as the method for performing data dimensionality reduction on the problem feature template.

[0080] Optionally, the processing module 203 performs dimensionality reduction processing on the first problem feature data in the following manner to obtain the second problem feature data:

[0081] Locality-sensitive hashing is performed on the first problem feature data to obtain the target hash data of the first problem feature data;

[0082] The target hash data is determined as the feature data of the second problem.

[0083] Optionally, the first problem feature data includes intent feature data.

[0084] Optionally, the intent feature data includes multiple intent sub-feature data, which are obtained by combining each intent sub-feature data in a chain structure.

[0085] Optionally, the question to be answered includes a formula, and the first question feature data includes intent feature data and parameter feature data of the parameters contained in the formula.

[0086] Optionally, the first problem feature data is extracted by a pre-trained problem feature extraction model, wherein the problem feature extraction model is trained by using the problem feature annotation data corresponding to the sample problem as model constraints.

[0087] The following is for reference. Figure 3 It illustrates an electronic device suitable for implementing embodiments of the present disclosure (e.g., Figure 1The diagram below shows the structure of the terminal device 600. The terminal device in this embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0088] like Figure 3 As shown, electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processor, etc.) 601, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 602 or a program loaded from storage device 608 into random access memory (RAM) 603. RAM 603 also stores various programs and data required for the operation of electronic device 600. Processing device 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0089] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic device 600 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 600 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0090] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a storage device 608, or installed from a ROM 602. When the computer program is executed by the processing device 601, it performs the functions defined in the methods of embodiments of this disclosure.

[0091] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. 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, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0092] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0093] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0094] The aforementioned computer-readable medium carries one or more programs. When the electronic device executes the aforementioned one or more programs, the electronic device causes the following to occur: extract features from the question to be solved to obtain first question feature data; match the first question feature data with a preset question feature template; if no first question feature template corresponding to the first question feature data is matched, perform data dimensionality reduction processing on the first question feature data to obtain second question feature data; match the second question feature data with the dimensionality-reduced question feature template to obtain a target question feature template matching the second question feature data, wherein the dimensionality-reduced question feature template is obtained after data dimensionality reduction processing of the question feature template; determine a target solution step corresponding to the target question feature template; and generate a target answer corresponding to the question to be solved based on the target solution step.

[0095] Alternatively, the aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: extract features from the question to be answered in order to obtain first question feature data;

[0096] The first problem feature data is matched with the preset problem feature template;

[0097] When no first problem feature template corresponding to the first problem feature data is matched, the first problem feature data is subjected to data dimensionality reduction processing to obtain the second problem feature data.

[0098] The second problem feature data and the dimensionality-reduced problem feature template are matched to obtain a target problem feature template that matches the second problem feature data, wherein the dimensionality-reduced problem feature template is obtained by data dimensionality reduction processing of the problem feature template;

[0099] Determine the target solution steps corresponding to the target problem feature template;

[0100] Based on the target solution steps, the target answer corresponding to the question to be answered is generated.

[0101] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as 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, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via 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., via the Internet using an Internet service provider).

[0102] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in 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 using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0103] The modules described in the embodiments of this disclosure can be implemented in software or hardware. The names of the modules are not, in some cases, intended to limit the functionality of the module itself.

[0104] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0105] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction 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 be, 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 machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0106] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0107] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0108] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative forms of implementing the claims. Regarding the apparatus in the above embodiments, the specific manner in which the various modules perform their operations has been described in detail in the embodiments relating to the method, and will not be elaborated upon here.

Claims

1. A problem-solving method, characterized in that, include: Feature extraction is performed on the question to be answered to obtain the feature data of the first question; The first problem feature data is matched with the preset problem feature template; When no first problem feature template corresponding to the first problem feature data is matched, the first problem feature data is subjected to data dimensionality reduction processing to obtain the second problem feature data. The second problem feature data and the dimensionality-reduced problem feature template are matched to obtain a target problem feature template that matches the second problem feature data. The dimensionality-reduced problem feature template is obtained after the problem feature template has undergone data dimensionality reduction processing. Both the first problem feature data and the problem feature template are subjected to local sensitive hashing processing for data dimensionality reduction. Determine the target solution steps corresponding to the target problem feature template; Based on the target solution steps, the target answer corresponding to the question to be answered is generated.

2. The method according to claim 1, characterized in that, The step of performing dimensionality reduction processing on the first problem feature data to obtain the second problem feature data includes: Locality-sensitive hashing is performed on the first problem feature data to obtain the target hash data of the first problem feature data; The target hash data is determined as the feature data of the second problem.

3. The method according to claim 1, characterized in that, The first problem feature data includes intent feature data.

4. The method according to claim 3, characterized in that, The intent feature data includes multiple intent sub-feature data, which are obtained by combining each intent sub-feature data in a chain structure.

5. The method according to claim 1, characterized in that, The question to be answered includes a formula, and the first question feature data includes intent feature data and parameter feature data of the parameters contained in the formula.

6. The method according to any one of claims 1-5, characterized in that, The first problem feature data is obtained by extracting the problem feature extraction model through a pre-trained model, wherein the problem feature extraction model is trained by using the problem feature annotation data corresponding to the sample problem as the model constraint.

7. A question-answering device, characterized in that, include: The feature extraction module is used to extract features from the problem to be solved in order to obtain the first problem feature data; The first matching module is used to match the first problem feature data with a preset problem feature template; The processing module is used to perform data dimensionality reduction processing on the first problem feature data to obtain the second problem feature data when no first problem feature template corresponding to the first problem feature data is matched. The second matching module is used to match the second problem feature data and the dimensionality-reduced problem feature template to obtain a target problem feature template that matches the second problem feature data. The dimensionality-reduced problem feature template is obtained after the problem feature template has undergone data dimensionality reduction processing. Both the first problem feature data and the problem feature template are subjected to local sensitive hashing processing for data dimensionality reduction. The determination module is used to determine the target solution steps corresponding to the target problem feature template; The solution module is used to generate the target answer corresponding to the question to be solved based on the target solution steps.

8. A computer-readable medium having a computer program stored thereon, characterized in that, When executed by the processing device, the program implements the steps of the method according to any one of claims 1-6.

9. An electronic device, characterized in that, include: A storage device on which computer programs are stored; A processing device for executing the computer program in the storage device to implement the steps of the method according to any one of claims 1-6.

Images