Information processing method and device, equipment, medium and program product

Abstract

The invention relates to the technical field of problem management, and discloses an information processing method and device, equipment, a medium and a program product, and the method comprises the steps: receiving problem description information which is input by a user and is related to a vehicle from a first terminal, and obtaining a first type of the problem description information according to the problem description information through a neural network model; sending the first type to the first terminal, so that the first terminal displays the first type; when a second type of question description information is received from the first terminal, the first type and the second type are sent to a second terminal, and the second type is different from the first type; receiving an audit result from the second terminal, and determining a target type according to the audit result; generating problem data, and storing the problem data, the problem data including the target type. The subjective influence of setting the problem type by the user is weakened, and the problem that the definition of the problem type related to the vehicle is inaccurate is solved.

Description

Technical Field

[0001] The present application relates to the field of problem management technology, and specifically to an information processing method, apparatus, device, medium, and program product. Background Art

[0002] Collecting information about vehicle-related issues, such as information about vehicle malfunctions, and feeding this information back to engineers responsible for resolving vehicle-related issues is a key step in resolving vehicle-related issues. The type of issue, such as a problem level indicating the severity of the issue, is a key item in the information about vehicle-related issues.

[0003] In the related art, personnel responsible for collecting information on vehicle-related issues select problem types based on prior experience and provide these types to engineers who resolve vehicle-related issues. However, due to the limitations of prior experience, personnel responsible for collecting information on vehicle-related issues often select incorrect problems, resulting in adverse effects. For example, a serious problem may be classified as a minor one, resulting in the inability to promptly resolve the serious problem. Improving the accuracy of problem types has become a pressing issue. Summary of the Invention

[0004] In view of this, the present application provides an information processing method, apparatus, device, medium and program product.

[0005] In a first aspect, the present application provides an information processing method, the method comprising: receiving vehicle-related problem description information input by a user from a first terminal, and obtaining a first type of problem description information based on the problem description information using a neural network model; sending the first type to the first terminal so that the first terminal displays the first type; when the second type of the problem description information is received from the first terminal, sending the first type and the second type to a second terminal, wherein the second type is different from the first type and is determined based on the user's operation for determining the second type; receiving an audit result from the second terminal, and determining a target type of the problem description information based on the audit result, wherein the audit result indicates that the first type or the second type is used as the target type; generating problem data, and storing the problem data, the problem data including: the target type.

[0006] In one possible implementation, the method further includes: when the audit result indicates that the second type is used as the target type, generating a new training sample based on the second type, the new training sample including: the problem description information and the second type; and using the new training sample to train the neural network model.

[0007] In one possible implementation, the method further includes: when the audit result indicates that the first type is to be used as the target type, pushing the first type to the user.

[0008] In a possible implementation, the second type is obtained according to preset rule information and information used to determine the second type, and is used to determine that an information item in the second type of information is selected by a user.

[0009] In one possible implementation, before receiving the problem description information related to the vehicle input by the user from the first terminal, the method further includes: receiving candidate input information input by the user in the problem description information input area from the first terminal; generating problem description suggestion information based on the candidate input information using the neural network model, wherein the problem description suggestion information indicates information items recommended to the user, and the information items are used to constitute the problem description information related to the vehicle; and sending the problem description suggestion information to the first terminal so that the first terminal displays the problem description suggestion information in the problem description information input area.

[0010] In one possible implementation, generating problem data includes: using the neural network model to generate a quality level of the vehicle-related problem description information; when the quality level meets a quality level threshold condition, generating problem data including the vehicle-related problem description information; when the quality level does not meet the quality level threshold condition, using the neural network model to generate improvement information corresponding to the vehicle-related problem description information, the improvement information being used to indicate errors involved in the vehicle-related problem description information; and sending the improvement information to the first terminal.

[0011] In a second aspect, the present application provides an information processing device, comprising: a problem description acquisition module for receiving vehicle-related problem description information input by a user from a first terminal, and using a neural network model to obtain a first type of problem description information based on the problem description information; a type suggestion module for sending a first type to the first terminal so that the first terminal displays the first type; a type review module for sending the first type and the second type to the second terminal when receiving the second type of the problem description information from the first terminal, wherein the second type is different from the first type and is determined based on the user's operation for determining the second type; a type calibration module for receiving an audit result from the second terminal, and determining a target type of the problem description information based on the audit result, wherein the audit result indicates that the first type or the second type is used as the target type; a problem storage module for generating problem data and storing the problem data, the problem data including: the target type.

[0012] In a third aspect, the present application provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0013] In a fourth aspect, the present application provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method of the above-mentioned first aspect or any corresponding embodiment thereof.

[0014] In a fifth aspect, the present application provides a computer program product, comprising computer instructions, which are used to enable a computer to execute the method of the above-mentioned first aspect or any corresponding embodiment thereof.

[0015] The technical solution provided by the information processing method provided in the embodiments of the present application has the following advantages:

[0016] On the one hand, when setting the type of the problem description information, the neural network model is used to obtain the first type of the problem description information based on the problem description information. The first type can be equivalent to the recommended type. Sending the first type to the first terminal and having the first terminal display the first type can be equivalent to providing the user with a suggestion for setting the type of the problem description information, that is, suggesting that the type of the problem description information be set to the first type. This allows the user to refer to the suggestion for setting the type of the problem description information and set the type of the problem description information, thereby improving the accuracy of setting the problem type. On the other hand, a second terminal is used, such as a terminal used by a user with rich prior experience or a model terminal including a reasoning performance higher than the neural network model, to determine whether the first type or the second type is used as the final target type. Creating problem data including the target type based on the review results of the second terminal to the problem system further improves the accuracy of the problem type and avoids relying solely on the user's prior experience to set the problem type, which may cause adverse effects due to inaccurate problem types of vehicle-related problem description information. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the specific implementation methods or the description of the prior art. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1 is a flowchart of an information processing method according to an embodiment of the present application;

[0019] Figure 2 is another flowchart of an information processing method according to an embodiment of the present application;

[0020] Figure 3 is another flowchart of an information processing method according to an embodiment of the present application;

[0021] Figure 4 is another flowchart of an information processing method according to an embodiment of the present application;

[0022] Figure 5 is a structural diagram of an information processing device according to an embodiment of the present application;

[0023] Figure 6 It is a schematic diagram of the hardware structure of the computer device of an embodiment of the present application. DETAILED DESCRIPTION

[0024] To make the purpose, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this application.

[0025] According to an embodiment of the present application, an information processing method embodiment is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0026] refer to Figure 1 , which shows a flow chart of an information processing method provided in an embodiment of the present application, which can be executed by a computer device such as a server.

[0027] In step S101, problem description information related to a vehicle input by a user is received from a first terminal, and a first type of problem description information is obtained based on the problem description information using a neural network model.

[0028] As an example, the first terminal includes but is not limited to a mobile phone, a vehicle-mounted terminal with a vehicle computer, a tablet computer, etc.

[0029] As an example, the user of the first terminal may be a person who uses a vehicle, such as a vehicle owner or a vehicle passenger, or a person who tests the performance of the vehicle.

[0030] The vehicle-related problem description information entered by the user can indicate that there are some usage problems with the vehicle. Feedback of the problem description information to the problem management system can be used by developers to perform backend repairs and resolve user-reported problems during subsequent vehicle upgrades.

[0031] For ease of description, the computer device or server that executes the information processing method provided in the embodiments of the present application is hereinafter referred to as the "local machine." The local machine may include a problem management system and may also be used to communicate and interact with a server deployed with the problem management system. When the local machine includes a problem management system, the software program that executes steps S101 to S105 may be integrated with the problem management system so that after generating reliable problem data, the problem data can be directly saved to the problem management system. When the local machine does not include a problem management system, the local machine may independently receive vehicle-related problem description information input by the user, define the problem type based on the problem description information, generate problem data, and then send the generated problem data to the server of the problem management system for problem storage.

[0032] As an example, when using the car computer, the user finds that the control of the rising and falling of the window has a time delay and the control effect is not sensitive enough, which affects normal use. The user can describe the control problem of the rising and falling of the window through the first terminal and then feedback the problem to the problem management system.

[0033] For example, the problem description information entered by the user includes not only fixed-format information such as the problem title, number, and vehicle location, but also textual explanations that are not in a fixed format but are intended to explain the specific problem. For example, the textual explanation might be: "The vehicle's windows will not move until approximately 1 to 2 seconds after clicking the vehicle control button."

[0034] After the user inputs the vehicle-related problem description information from the first terminal, the problem is sent to this machine. At this time, the machine calls the neural network model to read and analyze the problem description information, and through natural language processing technology, intelligently identifies the first type of problem description information.

[0035] In the embodiment of the present application, the neural network model is a natural language processing model, and the neural network model is pre-trained, that is, the neural network model is trained before step S101. As an example, before step S101, the neural network model is trained using a corpus text set and labels of the corpus text. It should be noted that in the embodiment of the present application, the first type includes but is not limited to a problem level used to describe the severity of the problem, a problem domain used to describe whether it belongs to the hardware domain or the software domain, etc.

[0036] As an example, the input and output of a neural network model can be as follows:

[0037] The first input of the neural network model: extract the title and description content from the entire question information and accurately call it for the main judge;

[0038] The second input of the neural network model: the domain to which the problem belongs, such as the vehicle body domain, infotainment domain, intelligent driving domain, powertrain domain, etc.

[0039] The third input of the neural network model: knowledge base (determination rules of problem level);

[0040] An example of the first type of format output by a neural network model: Type B50 is recommended.

[0041] In step S102 , a first type is sent to a first terminal, so that the first terminal displays the first type.

[0042] Specifically, the device sends the first type predicted by the neural network model to the first terminal and displays it on the first terminal. This gives the user some suggestions for defining the question type based on the first type predicted by the neural network model, eliminating the need for the user to rely solely on their own understanding of the type of question description information. The user can then enter the second type they believe is the type of question, using the first type predicted by the neural network model as a prompt or by selecting from a drop-down box.

[0043] In this embodiment, the second type input by the user may be exactly the same as the first type, or may be another type selected by the user near the first type based on the first type.

[0044] If the second type is identical to the first type, the target type can be determined based on the second or first type. Problem data is then created using the target type and the previously entered problem description information and stored in the problem management system. This solution prevents users from relying solely on their own unprofessional experience to define problem types, reduces the subjective influence of the first end user in setting problem types, and improves the accuracy of problem type determination.

[0045] In step S103, when the second type of problem description information is received from the first terminal, the first type and the second type are sent to the second terminal, wherein the second type is different from the first type and is determined according to the user's operation for determining the second type.

[0046] Specifically, when the second type input by the user is different from the first type predicted by the neural network model (for example, the question creator selects a second type that is similar to but different from the first type based on the prompt of the first type), the embodiment of the present application also sends the first type and the second type to the second terminal. The second terminal serves as an audit terminal for auditing the reliability of the first type and the second type.

[0047] As an example, the second terminal can include an automated review algorithm model trained based on expert experience and a rule library, storing a large amount of expert experience. Alternatively, the second terminal can be used by a highly experienced vehicle development expert. After the first and second types are sent to the second terminal, the expert model or the vehicle development expert reviews the accuracy of the first and second types.

[0048] In step S104, an audit result is received from the second terminal, and a target type of the problem description information is determined based on the audit result, wherein the audit result indicates that the first type or the second type is used as the target type;

[0049] Specifically, after the second terminal completes the review of the first type and the second type, it sends the review result to the local machine. The review result received by the local machine is used to indicate that the first type or the second type is used as the target type representing the final question type.

[0050] In step S105, question data is generated and stored, and the question data includes: target type.

[0051] Specifically, after determining the target type, this machine generates complete problem data by summarizing the problem description information and target type related to the vehicle, and saves the problem data to the problem management system.

[0052] The technical solution provided by the embodiments of this application also uses a more experienced and reliable second terminal to review the second type input by the user and the first type determined by the neural network model, indicating whether the first type or the second type should be used as the final target type. Based on the review results of the second terminal, question data including the target type is created and sent to the question system, further improving the accuracy of question types, reducing the subjective influence of question type settings by the first terminal user, and resolving the issue of inaccurate definition of vehicle-related question types.

[0053] refer to Figure 2 , which shows an example flow chart of another data acquisition method provided in an embodiment of the present application, which can be executed by a computer device such as a server.

[0054] In step S201, problem description information related to the vehicle input by a user is received from a first terminal, and a first type of problem description information is obtained based on the problem description information using a neural network model.

[0055] Specifically, in the embodiments of the present application, the first terminal refers to a terminal device used by a user, including but not limited to a mobile phone, a vehicle computer, a tablet computer, etc. The vehicle-related problem description information input by the user refers to the user discovering some usage problems with the vehicle during the use of the vehicle, thereby describing the problems found, and then feeding the described problems into the problem management system for developers to perform background repairs and resolve the user-reported problems during subsequent vehicle upgrades.

[0056] For ease of description, the computer device or server that executes the information processing method provided in the embodiments of the present application is hereinafter referred to as the "local machine". The local machine may include a problem management system and may also be used to communicate and interact with a server deployed with the problem management system. When the local machine includes a problem management system, the software program that executes steps S201 to S207 may be integrated with the problem management system so that after generating reliable problem data, the problem data can be directly saved to the problem management system. When the local machine does not include a problem management system, the local machine may independently receive vehicle-related problem description information input by the user, define the problem type based on the problem description information, generate problem data, and then send the generated problem data to the server of the problem management system for problem storage.

[0057] As an example, when using the car computer, the user finds that the control of the rising and falling of the window has a time delay and the control effect is not sensitive enough, which affects normal use. The user can describe the control problem of the rising and falling of the window through the first terminal and then feedback the problem to the problem management system.

[0058] For example, the problem description information entered by the user includes not only fixed-format information such as the problem title, number, and vehicle location, but also textual explanations that are not in a fixed format but are intended to explain the specific problem. For example, the textual explanation might be: "The vehicle's windows will not move until approximately 1 to 2 seconds after clicking the vehicle control button."

[0059] After the user inputs the vehicle-related problem description information from the first terminal, the problem is sent to this machine. At this time, the machine calls the neural network model to read and analyze the problem description information, and through natural language processing technology, intelligently identifies the first type of problem description information.

[0060] In an embodiment of the present application, the neural network model is a pre-trained natural language processing model, for example, trained using a large amount of text corpus and pre-labeled type tags to read and understand the problem description information entered by the user and provide a first type of recommendation. The first type includes, but is not limited to, a problem level used to describe the severity of the problem, a problem domain used to describe whether the problem belongs to the hardware or software domain, and the like.

[0061] In step S202, a first type is sent to a first terminal, so that the first terminal displays the first type.

[0062] Specifically, the device sends the first type predicted by the neural network model to the first terminal and displays it on the first terminal. This provides the user with some suggestions for defining the question type based on the first type predicted by the neural network model, eliminating the need for the user to rely solely on their own understanding of the question description information. The user can then enter a second type based on the first type predicted by the neural network model or select from a drop-down box to indicate the type of question.

[0063] In this embodiment, the second type input by the user may be exactly the same as the first type, or may be another type selected by the user near the first type based on the first type.

[0064] If the second type is identical to the first type, the target type can be determined based on the second or first type. Problem data is then created using the target type and the previously entered problem description information and stored in the problem management system. This solution prevents users from relying solely on their own unprofessional experience to define problem types, reduces the subjective influence of the first end user in setting problem types, and improves the accuracy of problem type determination.

[0065] In step S203, when the second type of problem description information is received from the first terminal, the first type and the second type are sent to the second terminal, wherein the second type is different from the first type and is determined according to the user's operation for determining the second type.

[0066] Specifically, when the second type input by the user is different from the first type predicted by the neural network model, the embodiment of the present application also sends the first type and the second type to the second terminal. The second terminal serves as an audit terminal for auditing the reliability of the first type and the second type.

[0067] As an example, the second terminal may be a device that stores a large amount of expert experience and includes an automatic audit algorithm model trained based on the expert experience and a rule base.

[0068] As an example, the second terminal may also be a terminal device used by a vehicle development expert with extensive development experience. After the first type and the second type are sent to the second terminal, the vehicle development expert manually reviews the accuracy of the first type and the second type.

[0069] In step S204, an audit result is received from the second terminal, and a target type of the problem description information is determined based on the audit result, wherein the audit result indicates that the first type or the second type is used as the target type.

[0070] In step S205, question data is generated and stored, where the question data includes: target type.

[0071] Specifically, after the second terminal completes its review of the first and second types, it sends the review results to the local device. The received review results indicate that either the first or second type will be used as the target type representing the final problem type. After determining the target type, the local device generates complete problem data by summarizing the vehicle-related problem description information and the target type, and saves the problem data to the problem management system.

[0072] In step S206, when the audit result indicates that the first type is used as the target type, the first type is pushed to the user.

[0073] In step S207, when the audit result indicates that the second type is used as the target type, a new training sample is generated based on the second type, and the neural network model is trained using the new training sample. The new training sample includes: problem description information and the second type.

[0074] Specifically, in the embodiment of the present application, the machine also performs an optimization operation on the question type function based on the target type determined by the audit result. The optimization operation includes two parts: model optimization and user knowledge optimization.

[0075] In this embodiment, when the audit result indicates that the first type is used as the target type, it means that the second terminal determines that it is more accurate to use the first type predicted by the neural network model to describe the vehicle-related problem input by the user, so the local machine directly determines the target type as the first type and saves the problem data.

[0076] At the same time, the machine also sends the first type predicted by the neural network model to the first terminal, which displays the first type to the user. This allows the user to understand that the problem description information entered is closer to the first type, providing the user with prior knowledge of the problem description information, helping the user learn background knowledge about vehicle-related problems, and thus achieving the user's learning goal. The next time the user creates a similar problem description, they can enter the correct problem type based on their own experience, significantly improving the accuracy of the problem type entered by subsequent users.

[0077] When the audit result indicates that the second type is used as the target type, it means that the second terminal determines that it is more accurate to use the second type input by the user to describe the problem related to the vehicle, so the local machine directly determines the target type as the second type and saves the problem data.

[0078] At the same time, the machine also uses the question description information entered by the user as a new sample feature and the second type as a new sample label to generate new training samples. The neural network model is then incrementally trained using the new training samples to further improve the neural network model's ability to subsequently read and understand similar types of questions, thereby improving the accuracy of the neural network model in predicting question types for similar types of questions.

[0079] The second type in the new training sample is the type of annotation of the problem description information, that is, the type expected to be output by the neural network model after the problem description information is input into the neural network model.

[0080] In a possible implementation, the second type input by the user is obtained according to preset rule information and information used to determine the second type, and is used to determine that an information item in the second type of information is selected by the user.

[0081] Specifically, in the aforementioned embodiment, the first terminal can provide several candidate question types. During the process of the user inputting question description information, the second type corresponding to the question description information input by the user can be directly selected from the candidate question types based on personal experience.

[0082] In this embodiment of the present application, an input prompt method can also be used to further improve the accuracy of the second type of user input. Preset rule information refers to a rule for calculating rules based on certain question attribute information input by the user, and the question attribute information input by the user is converted to the second type using the preset rule information. The preset rule information can be a formula rule developed by experienced vehicle developers based on previous experience in type definition. The question attribute information required for conversion to the second type is input or selected by the user according to the rule.

[0083] As an example, a question input page displayed on the first terminal may include multiple different information input areas and information selection areas. The information input area allows the user to enter question information by typing, and the question selection area may provide several optional question attribute information via a drop-down box for the user to select. After the user completes entering the question attribute information, the local terminal uses preset rule information to calculate the question attribute information to obtain a second type, which is used as the second type entered by the user.

[0084] As an example, when the user inputs a functional question, the question attribute information may include the following:

[0085] Recurrence probability (enter via input box): 0%-100%.

[0086] Whether it is a user scenario (selected through the drop-down box): Yes (1), No (0).

[0087] Impact on the vehicle (selected through the drop-down box): unable to drive / safety risk / black screen (1), limp / unavailable function (0.75), partially available function (0.5), no missing function, only poor experience (0.25), subjective optimization item (0.1).

[0088] Increase coefficient: 0-1, the default is 1.

[0089] In the question attribute information input by the user above, the numbers after the brackets indicate that each piece of information is used to substitute into the preset rule information to calculate the second type of parameters.

[0090] For example, the calculation formula of the preset rule information defined in this embodiment is: second type parameter = ((recurrence probability + whether it is a user scenario) / 2) * impact on the vehicle * coefficient.

[0091] In this embodiment, the type determination rule defined in the preset rule information may be:

[0092] If: parameter of the second type = 1, then output the second type = S300;

[0093] elif: 1> parameter of the second type ≧ 0.5, then output the second type = A100;

[0094] elif:0.5>parameter of the second type ≧0.25, then output the second type = B50;

[0095] elif: 0.25>parameter of the second type ≧ 0.1, then output the second type = B30;

[0096] else: output second type = C10.

[0097] As an example, when the user inputs a vehicle mode management question, the question attribute information may include the following:

[0098] Recurrence probability (enter via input box): 0%-100%.

[0099] Whether it is a user scenario (selected through the drop-down box): Yes (1), No (0).

[0100] Impact on the vehicle (selected via the drop-down box): average current 2.5A or above (1), average current 0.5A-2.5A (0.5), average current below 0.5A (0.25).

[0101] Increase coefficient: 0-1, the default is 1.

[0102] For example, the calculation formula of the preset rule information defined in this embodiment is: second type parameter = ((recurrence probability + whether it is a user scenario) / 2) * impact on the vehicle * coefficient.

[0103] In this embodiment, the type determination rule defined in the preset rule information may be:

[0104] If: the parameter of the second type is > 0.5, then output the second type = S300;

[0105] elif: 0.5≧parameter of the second type>0.25, then output the second type=A100;

[0106] else: output second type = B50.

[0107] According to the technical solution provided in the embodiment of the present application, in the process of the user inputting the second type, the second type of the user input can be assisted in adjusting by the preset rule information that integrates the developer's experience, so that the second type of the user input can further reduce the influence of the user's subjective factors, and the question type is assigned a question type with a higher error rate by less professional users, thereby further improving the accuracy of setting the question type.

[0108] refer to Figure 3 , which shows an example flow chart of another data acquisition method provided in an embodiment of the present application, which can be executed by a computer device such as a server.

[0109] In step S301, candidate input information input by a user in an input area of ​​question description information is received from a first terminal.

[0110] Specifically, in embodiments of the present application, when a user enters a problem description into a first terminal, the information currently in the input area is referred to as candidate input information. Each time the user enters a new field into the input area, the first terminal sends the candidate input information within the input area to the first terminal. For example, suppose the user currently enters the following: "Title: Window Lift Failure, Component: Window, Frequency: Always Occurs, Problem Description: Each Window Raises and Lowers." This is an incomplete entry and does not yet constitute a problem description, but rather serves as candidate input information.

[0111] In step S302, a neural network model is used to generate problem description suggestion information based on candidate input information, wherein the problem description suggestion information indicates information items recommended to the user, and the information items are used to form problem description information related to the vehicle.

[0112] In step S303, the problem description suggestion information is sent to the first terminal, so that the first terminal displays the problem description suggestion information in the problem description information input area.

[0113] Specifically, the neural network model provided in the embodiment of the present application also provides the ability to search for similar questions based on a large amount of corpus training samples, thereby analyzing similar questions entered by other users in history based on the candidate input information entered by the user, and giving question description suggestions to prompt the next step to enter the question.

[0114] As an example, the neural network model has been trained with a large number of corpus training samples and has handled similar car window problems. When the user enters a candidate input information such as "Title: Car window lifting failure, component: car window, frequency of occurrence: must occur, problem description: each car window when rising and falling", the neural network model understands the candidate input information, searches for similar information entered by other users, and infers the meaning the user wants to express, thereby giving more standardized and professional description suggestions.

[0115] For example, the neural network model generates a problem description suggestion: "Sluggish movement occurs." This local device feeds this problem description suggestion back to the first terminal. The user of the first terminal can then assemble a complete problem description based on this problem description suggestion and the candidate input information entered by the user. For example, by clicking on the problem description suggestion and performing the combined operation, the complete problem description is obtained: "Each vehicle window experiences sluggish movement when raising and lowering."

[0116] Through the technical solution provided in the embodiment of the present application, based on the natural language prediction capability of the neural network model, description suggestions are also provided during the process of the user inputting problem description information, so as to increase the standardization, accuracy and speed of the user input problem description information, and further improve the quality of the user input of vehicle-related problem description information.

[0117] In step S304, a problem description information related to the vehicle input by a user is received from the first terminal, and a first type of problem description information is obtained based on the problem description information using a neural network model;

[0118] In step S305, the first type is sent to the first terminal, so that the first type is displayed by the first terminal;

[0119] In step S306, when the second type of the problem description information is received from the first terminal, the first type and the second type are sent to the second terminal, wherein the second type is different from the first type and is determined according to the user's operation for determining the second type;

[0120] In step S307, an audit result is received from the second terminal, and a target type of the problem description information is determined based on the audit result, wherein the audit result indicates that the first type or the second type is used as the target type;

[0121] In step S308, question data is generated and stored, where the question data includes: target type.

[0122] The principle explanation of steps S304 to S308 is the same as that of the aforementioned steps S101 to S105 , and reference may be made to the relevant description of the aforementioned embodiment, which will not be repeated here.

[0123] refer to Figure 4 , which shows an example flow chart of another data acquisition method provided in an embodiment of the present application, which can be executed by a computer device such as a server.

[0124] In step S401, a problem description information related to a vehicle input by a user is received from a first terminal, and a first type of problem description information is obtained based on the problem description information using a neural network model;

[0125] In step S402, a first type is sent to a first terminal, so that the first terminal displays the first type;

[0126] In step S403, when the second type of the problem description information is received from the first terminal, the first type and the second type are sent to the second terminal, wherein the second type is different from the first type and is determined according to the user's operation for determining the second type;

[0127] In step S404, an audit result is received from the second terminal, and a target type of the problem description information is determined based on the audit result, wherein the audit result indicates that the first type or the second type is used as the target type;

[0128] Specifically, the principle explanation of steps S401 to S404 is the same as that of the aforementioned steps S101 to S104, and reference may be made to the relevant description of the aforementioned embodiment, which will not be repeated here.

[0129] In step S405, a quality rating of the problem description information related to the vehicle is generated using the neural network model;

[0130] Specifically, the neural network model provided in the embodiments of the present application also undergoes quality training. Specifically, the neural network model is trained using training samples that contain defects such as unclear descriptions, incorrect text formatting, missing attachments, and missing titles, thereby enabling the neural network model to perform quality inspections. In this embodiment, the neural network model can not only check for fixed formatting issues such as word count, attachments, and titles entered by the user, but can also check for inappropriate descriptions of the problem information entered by the user.

[0131] For example, the neural network model can check whether there are ambiguous sentences in the problem description information, whether there are grammatical problems in the sentences, whether there are inappropriate punctuation marks, etc.

[0132] Based on this, the embodiment of the present application can predefine different quality levels according to the number and type of errors in the problem description information. For example: the greater the number of errors, the lower the quantitative coefficient of the calculated quality level, and thus the greater the impact on the final quality level. At the same time, some simple format errors that do not affect the subsequent problem processing can make the quality coefficient of the calculated quality level slightly higher, and some description errors that affect the understanding of the problem will make the quality coefficient of the calculated quality level lower. Finally, the neural network model checks the problem description information and outputs the corresponding quantitative coefficient and quality coefficient. This embodiment can determine the quality level by the product of the initial value, the quantitative coefficient, and the quality coefficient.

[0133] In step S406 , when the quality level meets the quality level threshold condition, problem data including problem description information related to the vehicle is generated and stored, the problem data including: target type.

[0134] In step S407, when the quality level does not meet the quality level threshold condition, the neural network model is used to generate improvement information corresponding to the problem description information related to the vehicle, where the improvement information is used to indicate errors involved in the problem description information related to the vehicle;

[0135] In step S408, the improvement information is sent to the first terminal, and then the process returns to step S401;

[0136] Specifically, the embodiment of the present application predefines a quality level threshold condition to measure the quality of the problem description information. When the quality level calculated in the above steps is greater than the quality level threshold condition, the embodiment can determine that the quality of the input problem description information is good and can be directly stored. When the quality level is less than or equal to the quality level threshold condition, the embodiment can determine that the problem description information entered by the user is of poor quality and needs to be modified before saving. Therefore, when the quality level does not meet the quality level threshold condition, the local machine uses a neural network model to identify the problem description information and then generates improvement information corresponding to the vehicle-related problem description information. The improvement information is used to indicate errors related to the vehicle-related problem description information. For example, if a sentence in a certain location is inappropriate, the improvement information indicates that the correct description should be: "XXX". The local machine then sends the improvement information to the first terminal, allowing the user to modify the problem description information and then resubmit it for type suggestion and quality inspection, etc., until the problem meets the quality level threshold condition. The problem data that meets the quality level threshold condition is saved to the problem management system.

[0137] As an example, the problem description information entered by the user can be:

[0138] Issue ID: SWIM-536245

[0139] Question title: VMM-AUD_23 speaker_When closing the door and locking the car, the static current of the whole vehicle remains around 65mA and does not sleep - occasional;

[0140] Problem description:

[0141] *Preconditions: User Mode*

[0142] {*}Procedure and result{*}{*}:{*}

[0143] |STEP|Procedure|Expectedresult|Result|

[0144] |1|Start the vehicle|||

[0145] |2|Open and close the main driving door, lock the vehicle and sleep|The whole vehicle is in sleep mode, the static current is within 20mA|The static current of the whole vehicle is about 65mA and cannot sleep normally|

[0146] |||||

[0147] |||||

[0148] *Violated Requirement ID:None*

[0149] Influence: Increased risk of vehicle battery failure

[0150] *Steps to recover to normal state after the issue has occurred: None*

[0151] Occurrence %: 90%

[0152] Problem level: ['B30(SWIM-3730965)']

[0153] Attachment to the question: []

[0154] Problem function: ['FUNCTION NOT KNOWN YET (SWIM-868)']

[0155] Problem sub-function: ['FUNCTION NOT KNOWN YET (SWIM-3730050)']

[0156] Problem version: 6608336798B".

[0157] After the neural network model quality test, the output of the quality test can be:

[0158] Issue ID: SWIM-536245

[0159] Question title: VMM-AUD_23 speaker_When closing the door and locking the car, the static current of the whole vehicle remains around 65mA and does not sleep - occasional;

[0160] Title quality: qualified;

[0161] Describe the quality: unsatisfactory;

[0162] Grade quality: unqualified;

[0163] Recommended grade: B50;

[0164] Accessory quality: unqualified;

[0165] Functional quality: qualified;

[0166] Version quality: qualified;

[0167] Items to be corrected: [The "Preconditions:" in the description is missing specific prerequisites; "Steps to recover to normal state after the issue has occurred:" is blank; the issue level does not match the recommended level; attachments are missing];

[0168] Total quality: unqualified.

[0169] In addition, in one possible implementation, the neural network model provided by this machine can also be rechecked after the problem is updated. For example, if some problem data that has been saved to the problem management system has been manually modified by external personnel, this machine will monitor the problem management system at regular intervals and obtain the updated problem data in the problem management system during each monitoring. It will then call the neural network model to recheck the updated problem data, further improving the reliability of the problem and ensuring the quality of problem creation. At the same time, when the re-detection passes, this machine will automatically transfer the status of the problem to the next node, for example, changing the status of the problem from "creating" to "processing". When the re-detection fails, the machine will still feed back the problem and improvement information to the first terminal to facilitate the user to modify the problem.

[0170] Through the technical solution provided in the embodiment of the present application, the quality of the input problem description information is further controlled, and the quality of the problem data saved in the problem management system is further improved through post-review.

[0171] In this embodiment, an information processing device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments. Details already described will not be repeated here. As used below, the term "module" may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0172] This embodiment provides an information processing device, such as Figure 5 Shown, including:

[0173] A problem description acquisition module 501 is configured to receive a problem description information related to a vehicle input by a user from a first terminal, and obtain a first type of problem description information based on the problem description information using a neural network model;

[0174] a type suggestion module 502, configured to send a first type to a first terminal, so that the first terminal displays the first type;

[0175] a type review module 503 configured to, upon receiving a second type of the problem description information from the first terminal, send the first type and the second type to the second terminal, wherein the second type is different from the first type and is determined based on an operation performed by a user to determine the second type;

[0176] a type calibration module 504 configured to receive an audit result from the second terminal and determine a target type based on the audit result, wherein the audit result indicates that the first type or the second type is used as the target type;

[0177] The question storage module 505 is used to generate question data and store the question data, where the question data includes: target type.

[0178] In one possible implementation, the information processing device provided by this embodiment further includes:

[0179] A new sample generating unit, configured to generate a new training sample based on the second type when the audit result indicates that the second type is used as the target type;

[0180] The incremental training unit is used to train the neural network model using new training samples.

[0181] In one possible implementation, the method further includes:

[0182] The type push unit is used to push the first type to the user when the review result indicates that the first type is used as the target type.

[0183] In one possible implementation, the second type is obtained by applying the rules indicated by the preset rule information to determine the second type of operation, and the operation for determining the second type includes: user input according to the rules, used to determine the second type of information, and / or user selection according to the rules, used to determine the second type of information.

[0184] In a possible implementation, before the problem description obtaining module 501, the following is further included:

[0185] A candidate information acquisition module, configured to receive, from the first terminal, candidate input information input by the user in the question description information input area;

[0186] a suggestion module, configured to generate problem description suggestion information based on candidate input information using a neural network model, wherein the problem description suggestion information indicates information items recommended to the user, the information items being used to form problem description information related to the vehicle;

[0187] The suggestion push module is used to send the problem description suggestion information to the first terminal, so that the first terminal displays the problem description suggestion information in the problem description information input area.

[0188] In one possible implementation, the question storage module 505 includes:

[0189] A quality detection module, for generating a quality grade of vehicle-related problem description information using a neural network model;

[0190] a problem generating module, configured to generate problem data including problem description information related to the vehicle when the quality level meets a quality level threshold condition;

[0191] an improvement generation module, configured to generate, by using a neural network model, improvement information corresponding to the vehicle-related problem description information when the quality level does not meet the quality level threshold condition, the improvement information being used to indicate errors involved in the vehicle-related problem description information;

[0192] The improvement push module is used to send improvement information to the first terminal.

[0193] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0194] The present application also provides a computer device having the above Figure 5 The information processing device shown.

[0195] See also Figure 6 , Figure 6 This is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present application. Figure 6 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In a possible implementation, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 6 A processor 10 is taken as an example.

[0196] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0197] The memory 20 stores instructions that can be executed by at least one processor 10, so as to enable at least one processor 10 to execute the method shown in the above embodiment.

[0198] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In one possible implementation, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0199] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0200] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0201] The embodiments of the present application also provide a computer-readable storage medium. The above-mentioned method according to the embodiment of the present application can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0202] Part of the present application may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present application through the operation of the computer. Those skilled in the art should understand that the form in which the computer program instruction exists in a computer-readable medium includes but is not limited to a source file, an executable file, an installation package file, etc. Accordingly, the way in which the computer program instruction is executed by the computer includes but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium that can be accessed by the computer.

[0203] Although the embodiments of the present application have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations shall fall within the scope defined by the appended claims.

Claims

1. An information processing method, characterized in that: The method comprises: Receiving, from a first terminal, problem description information related to a vehicle input by a user, and obtaining, using a neural network model, a first type of problem description information based on the problem description information; sending a first type to a first terminal, so that the first terminal displays the first type; When receiving the second type of the problem description information from the first terminal, sending the first type and the second type to the second terminal, wherein the second type is different from the first type and is determined according to an operation of the user for determining the second type; receiving an audit result from the second terminal, and determining a target type of the problem description information according to the audit result, wherein the audit result indicates that the first type or the second type is used as the target type; Generate question data and store the question data, wherein the question data includes: the target type.

2. The method according to claim 1, characterized in that The method further comprises: When the audit result indicates that the second type is used as the target type, generating a new training sample based on the second type, the new training sample including: the problem description information and the second type; The neural network model is trained using the new training samples.

3. The method according to claim 1 or 2, characterized in that The method further comprises: When the audit result indicates that the first type is used as the target type, the first type is pushed to the user.

4. The method according to claim 2, characterized in that The second type is obtained according to preset rule information and information used to determine the second type, and is used to determine that an information item in the second type of information is selected by the user.

5. The method according to claim 1, wherein Before receiving the vehicle-related problem description information input by the user from the first terminal, the method further includes: receiving, from the first terminal, candidate input information input by a user in a question description information input area; generating problem description suggestion information based on the candidate input information using the neural network model, wherein the problem description suggestion information indicates information items recommended to the user, the information items being used to form problem description information related to the vehicle; The problem description suggestion information is sent to the first terminal, so that the first terminal displays the problem description suggestion information in the problem description information input area.

6. The method according to claim 1, characterized in that Generating question data includes: generating a quality rating of the vehicle-related problem description information using the neural network model; When the quality level satisfies a quality level threshold condition, generating problem data including the problem description information related to the vehicle; When the quality level does not meet a quality level threshold condition, generating, using the neural network model, improvement information corresponding to the vehicle-related problem description information, the improvement information being used to indicate an error involved in the vehicle-related problem description information; The improvement information is sent to the first terminal.

7. An information processing device, characterized in that The device comprises: a problem description acquisition module, configured to receive problem description information related to the vehicle input by a user from the first terminal, and obtain a first type of problem description information based on the problem description information using a neural network model; a type suggestion module, configured to send a first type to a first terminal, so that the first terminal displays the first type; a type review module, configured to, when receiving the second type of the problem description information from the first terminal, send the first type and the second type to the second terminal, wherein the second type is different from the first type and is determined based on an operation of the user for determining the second type; a type calibration module, configured to receive an audit result from the second terminal, and determine a target type of the problem description information based on the audit result, wherein the audit result indicates that the first type or the second type is used as the target type; The question storage module is used to generate question data and store the question data, where the question data includes: the target type.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the method according to any one of claims 1 to 6.

Images