CN118093851A

CN118093851A - Task processing method, automatic question answering method and legal task processing method

Info

Publication number: CN118093851A
Application number: CN202410153944.6A
Authority: CN
Inventors: 卿立之; 康杨杨; 孙常龙
Original assignee: Alibaba China Co Ltd
Current assignee: Alibaba China Co Ltd
Priority date: 2024-02-02
Filing date: 2024-02-02
Publication date: 2024-05-28

Abstract

The embodiment of the specification provides a task processing method, an automatic question-answering method and a legal task processing method, wherein the task processing method comprises the following steps: acquiring task description information of a target analysis task; carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes candidate data screening intention corresponding to the target analysis task; according to the task key information, retrieving at least one task reference data from a plurality of candidate data, wherein the task reference data is data related to the task key information in the plurality of candidate data; and inputting at least one task reference data into a task processing model to obtain a task processing result of the target analysis task. The task key information is obtained through analysis, so that candidate data screening conditions are accurately determined, data retrieval is automatically carried out by utilizing the task key information, task reference data obtained through retrieval is processed by utilizing a task processing model, and task processing efficiency and accuracy are improved.

Description

Task processing method, automatic question answering method and legal task processing method

Technical Field

The embodiment of the specification relates to the technical field of computers, in particular to a task processing method, an automatic question-answering method and a legal task processing method.

Background

With the development of computer technology, automated data analysis is becoming an important research point. The data analysis is widely applied to various scenes such as academic research, legal consultation and the like. The data analysis refers to the process of collecting, sorting, processing and interpreting data to find valuable information hidden in the data, to build models, and to perform the processes of prediction, decision support, project optimization and the like based on the findings.

Currently, related analysis documents are manually screened by a user, and analysis results are manually extracted. However, the quality of manually screening related analysis documents by users may be poor, the number is usually small, and the whole analysis process needs to consume a lot of time, so that the analysis efficiency is extremely poor, the analysis result is incomplete, and the accuracy is poor, and therefore, an efficient and accurate task processing scheme is needed.

Disclosure of Invention

In view of this, the present embodiment provides a task processing method. One or more embodiments of the present specification relate to an automatic question-answering method, a legal task processing method, a task processing device, an automatic question-answering device, a legal case studying and judging device, a computing device, a computer readable storage medium and a computer program product, which solve the technical defects existing in the prior art.

According to a first aspect of embodiments of the present specification, there is provided a task processing method, including:

acquiring task description information of a target analysis task;

Carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes candidate data screening intention corresponding to the target analysis task;

according to the task key information, retrieving at least one task reference data from a plurality of candidate data, wherein the task reference data is data related to the task key information in the plurality of candidate data;

and inputting at least one task reference data into a task processing model to obtain a task processing result of the target analysis task.

According to a second aspect of embodiments of the present specification, there is provided an automatic question-answering method, including:

Acquiring task description information of a target question-answering task;

carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes candidate data screening intention corresponding to the target question-answering task;

And inputting at least one task reference data into a task processing model to obtain a reply result of the target question-and-answer task.

According to a third aspect of embodiments of the present specification, there is provided a legal task processing method, including:

acquiring task description information of legal tasks;

carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes legal candidate data screening intention corresponding to legal tasks;

retrieving at least one legal task reference data from a plurality of legal candidate data according to the task key information, wherein the legal task reference data is data related to the task key information in the plurality of legal candidate data;

and inputting at least one legal task reference data into a task processing model to obtain a task processing result of the legal task.

According to a fourth aspect of embodiments of the present specification, there is provided a task processing device including:

The first acquisition module is configured to acquire task description information of a target analysis task;

the first analysis module is configured to perform structural analysis on the task description information to obtain task key information, wherein the task key information characterizes candidate data screening intention corresponding to the target analysis task;

The first retrieval module is configured to retrieve at least one task reference data from the plurality of candidate data according to the task key information, wherein the task reference data is data related to the task key information in the plurality of candidate data;

The first input module is configured to input at least one task reference data into the task processing model to obtain a task processing result of the target analysis task.

According to a fifth aspect of embodiments of the present specification, there is provided an automatic question-answering apparatus, including:

the second acquisition module is configured to acquire task description information of the target question-answering task;

the second analysis module is configured to perform structural analysis on the task description information to obtain task key information, wherein the task key information characterizes candidate data screening intents corresponding to the target question-answering task;

The second retrieval module is configured to retrieve at least one task reference data from the plurality of candidate data according to the task key information, wherein the task reference data is data related to the task key information in the plurality of candidate data;

and the second input module is configured to input at least one task reference data into the task processing model to obtain a reply result of the target question-and-answer task.

According to a sixth aspect of embodiments of the present specification, there is provided a legal case study judgement device comprising:

The third acquisition module is configured to acquire task description information of legal tasks;

The third analysis module is configured to perform structural analysis on the task description information to obtain task key information, wherein the task key information characterizes legal candidate data screening intention corresponding to legal tasks;

The third retrieval module is configured to retrieve at least one legal task reference data from the plurality of legal candidate data according to the task key information, wherein the legal task reference data is data related to the task key information in the plurality of legal candidate data;

the third input module is configured to input at least one legal task reference data into the task processing model to obtain a task processing result of the legal task.

According to a seventh aspect of embodiments of the present specification, there is provided a computing device comprising:

A memory and a processor;

The memory is adapted to store a computer program/instruction which, when executed by the processor, implements the steps of the method provided in the first or second or third aspect described above.

According to an eighth aspect of embodiments of the present specification, there is provided a computer readable storage medium storing a computer program/instruction which when executed by a processor implements the steps of the method provided in the first or second or third aspect described above.

According to a ninth aspect of embodiments of the present description, there is provided a computer program product comprising computer programs/instructions which when executed by a processor implement the steps of the method provided in the first or second or third aspect described above.

The task processing method provided by one embodiment of the present specification includes: acquiring task description information of a target analysis task; carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes candidate data screening intention corresponding to the target analysis task; according to the task key information, retrieving at least one task reference data from a plurality of candidate data, wherein the task reference data is data related to the task key information in the plurality of candidate data; and inputting at least one task reference data into a task processing model to obtain a task processing result of the target analysis task. The task key information is obtained through analysis from the task description information, so that candidate data screening conditions are accurately determined, data retrieval is further automatically carried out by utilizing the task key information, a plurality of candidate data are truly retrieved, accuracy of task reference data is guaranteed, meanwhile, the retrieved task reference data are processed by utilizing a task processing model, a complete automatic task processing flow is realized, and task processing efficiency and accuracy of task processing results are improved.

Drawings

FIG. 1 is an architecture diagram of a task processing system provided in one embodiment of the present description;

FIG. 2 is an architecture diagram of another task processing system provided by one embodiment of the present description;

FIG. 3 is a flow chart of a method of task processing provided in one embodiment of the present disclosure;

FIG. 4 is a flow chart of an automatic question-answering method provided by one embodiment of the present disclosure;

FIG. 5 is a flow chart of a legal task processing method provided by one embodiment of the present disclosure;

FIG. 6a is a process flow diagram of a legal task processing method provided by one embodiment of the present disclosure;

FIG. 6b is a schematic diagram of a research and judgment graph report in a legal task processing method according to one embodiment of the present disclosure;

FIG. 7 is an interface schematic of a task processing interface provided by one embodiment of the present disclosure;

FIG. 8 is a schematic diagram of a task processing device according to an embodiment of the present disclosure;

fig. 9 is a schematic structural diagram of an automatic question answering device according to one embodiment of the present disclosure;

FIG. 10 is a schematic diagram of a legal case study decision device provided in accordance with one embodiment of the present disclosure;

FIG. 11 is a block diagram of a computing device provided in one embodiment of the present description.

Detailed Description

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present description. This description may be embodied in many other forms than described herein and similarly generalized by those skilled in the art to whom this disclosure pertains without departing from the spirit of the disclosure and, therefore, this disclosure is not limited by the specific implementations disclosed below.

The terminology used in the one or more embodiments of the specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of the specification. As used in this specification, one or more embodiments and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present specification refers to and encompasses any or all possible combinations of one or more of the associated listed items.

It should be understood that, although the terms first, second, etc. may be used in one or more embodiments of this specification to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first may also be referred to as a second, and similarly, a second may also be referred to as a first, without departing from the scope of one or more embodiments of the present description. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "in response to a determination" depending on the context.

Furthermore, it should be noted that, user information (including, but not limited to, user equipment information, user personal information, etc.) and data (including, but not limited to, data for analysis, stored data, presented data, etc.) according to one or more embodiments of the present disclosure are information and data authorized by a user or sufficiently authorized by each party, and the collection, use, and processing of relevant data is required to comply with relevant laws and regulations and standards of relevant countries and regions, and is provided with corresponding operation entries for the user to select authorization or denial.

In one or more embodiments of the present description, a large model refers to a deep learning model with large scale model parameters, typically including hundreds of millions, billions, trillions, and even more than one billion model parameters. The large Model can be called as a Foundation Model, a pre-training Model of the large Model is carried out through a large-scale unlabeled corpus, a pre-training Model with more than one hundred million parameters is produced, the Model can adapt to a wide downstream task, and the Model has better generalization capability, stronger natural language understanding, generating and zero-sample-less capability, such as a large-scale language Model (LLM, large Language Model), a multi-mode pre-training Model (multi-modal pre-training Model) and the like.

When the large model is actually applied, the pretrained model can be applied to different tasks by fine tuning with a small amount of samples, the large model can be widely applied to the fields of natural language processing (NLP, natural Language Processing), computer vision and the like, and particularly can be applied to the tasks of the computer vision fields such as vision question and answer (VQA, visual Question Answering), image description (IC), image generation and the like, and the tasks of the natural language processing fields such as emotion classification based on texts, text abstract generation, machine translation and the like, and main application scenes of the large model comprise digital assistants, intelligent robots, searching, online education, office software, electronic commerce, intelligent design and the like.

First, terms related to one or more embodiments of the present specification will be explained.

Legal case study and judgment: the legal case study and judgment refers to multi-dimensional statistical analysis of a batch of legal cases, and the statistical analysis is presented by a chart to form a final research report about commonalities and trends of judicial cases.

An intelligent agent: agent refers to an Agent that is capable of autonomously handling a unique task scenario, including but not limited to memory, planning, and tool-use capabilities, based on large model capabilities.

With the development of computer technology, automated data analysis is becoming an important research point. The data analysis is widely applied to various scenes such as academic research, legal consultation and the like. Taking legal consultation scenario as an example, a user may need to perform comprehensive research and judgment analysis on legal cases corresponding to different requirements, where multi-dimensional data aggregation analysis, visual chart presentation, natural language analysis text generation and the like may be performed in the legal case research and judgment process.

Currently, cases are manually screened by users, and a research chart and text are generated through a predefined template. However, the number of cases manually screened by the user is usually small, which results in incomplete legal case task processing results and poor accuracy, and the predefined templates are too monotonous, which results in single legal case task processing results and lack of diversity.

In order to solve the above problems, the embodiments of the present disclosure provide a task processing agent, which can intelligently arrange the whole task processing flow, and implement end-to-end automatic task processing. Specifically, task description information of a target analysis task is obtained; carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes candidate data screening intention corresponding to the target analysis task; according to the task key information, retrieving at least one task reference data from a plurality of candidate data, wherein the task reference data is data related to the task key information in the plurality of candidate data; and inputting at least one task reference data into a task processing model to obtain a task processing result of the target analysis task. The task key information is obtained through analysis from the task description information, so that candidate data screening conditions are accurately determined, data retrieval is further automatically carried out by utilizing the task key information, a plurality of candidate data are truly retrieved, accuracy of task reference data is guaranteed, meanwhile, the retrieved task reference data are processed by utilizing a task processing model, a complete automatic task processing flow is realized, and task processing efficiency and accuracy of task processing results are improved.

In the present specification, a task processing method, an automatic question-answering method, a legal task processing method, a task processing device, an automatic question-answering device, a legal case studying and judging device, a computing device, a computer readable storage medium and a computer program product are provided, and are described in detail in the following embodiments one by one.

Referring to fig. 1, fig. 1 illustrates an architecture diagram of a task processing system provided in one embodiment of the present disclosure, where the task processing system may include a client 100 and a server 200;

the client 100 is configured to send task description information of a target analysis task to the server 200;

The server 200 is configured to perform structural analysis on the task description information to obtain task key information, where the task key information characterizes candidate data screening intent corresponding to the target analysis task; according to the task key information, retrieving at least one task reference data from a plurality of candidate data, wherein the task reference data is data related to the task key information in the plurality of candidate data; inputting at least one task reference data into a task processing model to obtain a task processing result of a target analysis task; sending a task processing result to the client 100;

the client 100 is further configured to receive a task processing result sent by the server 200.

By applying the scheme of the embodiment of the specification, the task key information is obtained through analysis from the task description information, so that the candidate data screening condition is accurately determined, the task key information is further utilized to automatically search data, a plurality of candidate data are truly searched, the accuracy of task reference data is ensured, meanwhile, the task reference data obtained through searching is processed by utilizing the task processing model, the complete automatic task processing flow is realized, and the task processing efficiency and the accuracy of task processing results are improved.

Referring to fig. 2, fig. 2 illustrates an architecture diagram of another task processing system provided in one embodiment of the present disclosure, where the task processing system may include a plurality of clients 100 and a server 200, where the clients 100 may include an end-side device and the server 200 may include a cloud-side device. Communication connection can be established between the plurality of clients 100 through the server 200, in a task processing scenario, the server 200 is used to provide task processing services between the plurality of clients 100, and the plurality of clients 100 can respectively serve as a transmitting end or a receiving end, so that communication is realized through the server 200.

The user may interact with the server 200 through the client 100 to receive data transmitted from other clients 100, or transmit data to other clients 100, etc. In the task processing scenario, it may be that the user issues a data stream to the server 200 through the client 100, and the server 200 generates a task processing result according to the data stream and pushes the task processing result to other clients that establish communications.

Wherein, the client 100 and the server 200 establish a connection through a network. The network provides a medium for a communication link between client 100 and server 200. The network may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others. The data transmitted by the client 100 may need to be encoded, transcoded, compressed, etc. before being distributed to the server 200.

The client 100 may be a browser, APP (Application), or a web Application such as H5 (HyperText Markup Language, hypertext markup language (htv) 5 th edition) Application, or a light Application (also called applet, a lightweight Application) or cloud Application, etc., and the client 100 may be based on a software development kit (SDK, software Development Kit) of a corresponding service provided by the server 200, such as a real-time communication (RTC, real Time Communication) based SDK development acquisition, etc. The client 100 may be deployed in an electronic device, need to run depending on the device or some APP in the device, etc. The electronic device may for example have a display screen and support information browsing etc. as may be a personal mobile terminal such as a mobile phone, tablet computer, personal computer etc. Various other types of applications are also commonly deployed in electronic devices, such as human-machine conversation type applications, model training type applications, text processing type applications, web browser applications, shopping type applications, search type applications, instant messaging tools, mailbox clients, social platform software, and the like.

The server 200 may include a server that provides various services, such as a server that provides communication services for multiple clients, a server for background training that provides support for a model used on a client, a server that processes data sent by a client, and so on. It should be noted that, the server 200 may be implemented as a distributed server cluster formed by a plurality of servers, or may be implemented as a single server. The server may also be a server of a distributed system or a server that incorporates a blockchain. The server may also be a cloud server for cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN, content Delivery Network), basic cloud computing services such as big data and artificial intelligence platforms, or an intelligent cloud computing server or an intelligent cloud host with artificial intelligence technology.

It should be noted that, the task processing method provided in the embodiments of the present disclosure is generally executed by the server, but in other embodiments of the present disclosure, the client may also have a similar function to the server, so as to execute the task processing method provided in the embodiments of the present disclosure. In other embodiments, the task processing method provided in the embodiments of the present disclosure may be performed by the client and the server together.

Referring to fig. 3, fig. 3 shows a flowchart of a task processing method according to an embodiment of the present disclosure, which specifically includes the following steps:

step 302: and acquiring task description information of the target analysis task.

In one or more embodiments of the present disclosure, task description information of a target analysis task may be obtained, so that task processing is performed based on the task description information, and a task processing result of the target analysis task is obtained.

Specifically, the target analysis task may be an analysis task of different scenarios, such as a financial analysis task of a financial scenario, a legal task of a legal scenario (such as a legal case study task, a legal document analysis task, a legal document understanding task, etc.), a sales analysis task of an e-commerce scenario, etc. The task description information is used for describing task processing requirements of the target analysis task, if the target analysis task is a legal case study and judgment task of legal consultation scene, the task description information can be "help me to collect cases with the XXX amount reaching 20 ten thousand in 21 years in A city, and summarize a study and judgment report".

It should be noted that the task description information may be information of different modalities, such as task description information of a text modality, task description information of a voice modality, task description information of an image modality, task description information of a video modality, and the like. Under the condition that the task description information is of a non-text mode and the large model is utilized for task processing, the task description information can be subjected to mode conversion to obtain the task description information of a text mode, so that the large model is convenient for task processing. The mode of performing the mode conversion on the task description information is various, and is specifically selected according to the actual situation, which is not limited in any way in the embodiment of the present specification. Taking task description information of a voice mode as an example, the task description information of the voice mode can be converted into task description information of a text mode by utilizing a voice-to-text tool, and the task description information of the voice mode can be input into a voice recognition model to obtain the task description information of the text mode.

In practical applications, there are various ways to obtain task description information of the target analysis task, and the method is specifically selected according to practical situations, which is not limited in any way in the embodiments of the present specification. In one possible implementation manner of the present disclosure, task description information of a target analysis task sent by a user may be received. In another possible implementation manner of the present specification, task description information of the target analysis task may be read from other data acquisition devices or databases.

Step 304: and carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes candidate data screening intention corresponding to the target analysis task.

In one or more embodiments of the present disclosure, after task description information of a target analysis task is obtained, further, structural analysis may be performed on the task description information to obtain task key information, so that task reference data referred to in a target analysis task processing process is accurately screened out from multiple candidate data by using the task key information.

Specifically, the task key information characterizes candidate data screening intents corresponding to the target analysis task, and the task key information can be understood as candidate data screening conditions. Structural parsing refers to a process of extracting structural information from task description information, that is, task key information may be structural information. Structured information refers to data having a predefined format and organization architecture.

For example, assuming that the task description information may be "group me to collect cases in 21 years XXX amount of a city of a up to 20 ten thousand, and to collect a research report", the task description information is structured and parsed, and the obtained task key information may be a dictionary format structured information "region: market A; time: 21 years; case type: XXX; amount of money: up to 20 ten thousand).

In practical application, the task description information is structured and analyzed, and various manners of obtaining the task key information are available, and the task key information is specifically selected according to practical situations, which is not limited in any way in the embodiment of the present specification. In one possible implementation manner of the present disclosure, entity identification (NER, named Entity records) may be performed on task description information, so as to identify and classify named entities in the task description information, such as person names, place names, organization names, dates, times, and so on, to obtain task key information.

In another possible implementation manner of the present disclosure, the information analysis model may be used to extract the task key information from the task description information, that is, the above-mentioned performing structural analysis on the task description information to obtain the task key information, and the method may include the following steps:

Inputting the task description information into an information analysis model to obtain task key information, wherein the information analysis model is obtained based on a plurality of sample description information and sample key information corresponding to the sample description information respectively.

Specifically, the information analysis model may be a large model, or may be a model obtained by continuously training on data in a specified field with the large model as a base. For example, the objective analysis task is a legal case study and judgment task of legal consultation scene, the large model can be used as a base, and the large model is trained by using data in legal field (such as interpretation sample description information and understanding sample description information of legal common elements such as cases, laws, disputes and the like) to obtain an information analysis model for processing the legal case study and judgment task. By training the large model on the data in the appointed field, the large model can master the knowledge in the appointed field, and can carry out reasoning and judgment based on the knowledge in the appointed field, so that the accuracy of the task key information is improved. The information parsing model includes, but is not limited to BERT (Bidirectional Encoder Representations from Transformers) model, T5 (Text-to-Text Transfer Transformer) model.

It should be noted that, the process of training to obtain the information analysis model based on the plurality of sample description information and the sample key information corresponding to the plurality of sample description information respectively may include the following steps: acquiring a plurality of sample description information, wherein the sample description information carries sample key information; inputting the plurality of sample description information into an information analysis model to obtain prediction key information corresponding to the plurality of sample description information respectively; and adjusting model parameters of the information analysis model according to the sample key information and the prediction key information to obtain the information analysis model after training. The sample description information is a processing object of the model training process. The sample key information is a real analysis result corresponding to the sample description information, and the sample key information is an analysis target of the information analysis model. The method for obtaining the description information of the plurality of samples is various, and is specifically selected according to practical situations, and the embodiment of the present disclosure is not limited in any way. In a first possible implementation manner of the present disclosure, a plurality of sample description information may be read from other data acquisition devices or databases. In a second possible implementation manner of the present disclosure, a plurality of sample description information input by a user may be received.

Further, when the model parameters of the information analysis model are adjusted according to the sample key information and the prediction key information, the analysis loss value can be calculated according to the sample key information and the prediction key information, and the model parameters of the information analysis model are adjusted according to the analysis loss value until the preset stopping condition is reached, so that the information analysis model after training is obtained. There are many functions for calculating the analytic loss value, such as cross entropy loss function, L1 norm loss function, maximum loss function, mean square error loss function, logarithmic loss function, etc., which are specifically selected according to practical situations, and the embodiment of the present disclosure is not limited in any way. The preset stopping conditions include, but are not limited to, the analysis loss value being smaller than or equal to a preset threshold value, and the iteration number reaching a preset iteration number, wherein the preset threshold value and the preset iteration number are specifically selected according to actual conditions, and the embodiment of the present disclosure does not limit the above.

In one possible implementation manner of the present disclosure, after calculating the analysis loss value, the analysis loss value is compared with a preset threshold value. Specifically, if the analysis loss value is greater than a preset threshold value, the difference between the predicted key information and the sample key information is larger, the analysis capability of the information analysis model on the sample description information is poorer, at this time, model parameters of the information analysis model can be adjusted, the step of inputting the sample description information into the information analysis model to obtain the predicted key information corresponding to the sample description information is performed, training of the information analysis model is continued until the analysis loss value is less than or equal to the preset threshold value, the difference between the predicted key information and the sample key information is smaller, a preset stop condition is reached, and the information analysis model with complete training is obtained.

In another possible implementation manner of the present disclosure, in addition to comparing the magnitude relation between the analysis loss value and the preset threshold, it may also be determined whether the training of the current information analysis model is completed in combination with the iteration number. Specifically, if the analysis loss value is greater than a preset threshold value, the model parameters of the information analysis model are adjusted, the step of inputting the sample description information into the information analysis model to obtain the prediction key information corresponding to the sample description information is performed, training is continued on the information analysis model until the preset iteration times are reached, iteration is stopped, and the information analysis model with the training completed is obtained.

By applying the scheme of the embodiment of the specification, the task description information is input into the information analysis model to obtain the task key information, the task key information is efficiently and accurately determined, the analysis loss value is calculated according to the prediction key information and the sample key information, the analysis loss value is compared with the preset stop condition, the information analysis model is continuously trained under the condition that the preset stop condition is not met, the preset stop condition is reached, and the training is completed to obtain the information analysis model. The model parameters of the information analysis model are continuously adjusted, so that the finally obtained information analysis model is more accurate.

In an optional embodiment of the present disclosure, in order to avoid a situation that a task processing fails due to a lack of task key information in task description information, before performing structural analysis on the task description information, whether the task description information can be analyzed currently or not may be determined by using a structural analysis condition, that is, before performing structural analysis on the task description information, the method may further include the following steps:

Under the condition that the task description information does not meet the structural analysis condition, the description guide information is sent to the user;

receiving updated task description information sent by a user based on the description guide information;

the method for carrying out structural analysis on the task description information to obtain the task key information can comprise the following steps:

And under the condition that the task description information meets the structural analysis condition, carrying out structural analysis on the task description information to obtain the task key information.

Specifically, the structural analysis condition is used for determining whether to perform structural analysis on the task description information. The structural analysis conditions include, but are not limited to, that the task description information includes a named entity, that the task description information is clear and smooth, and the like, and are specifically set according to actual situations, and the embodiment of the present disclosure does not limit the description. The description guiding information user guides the user to update the task description information and sends the updated task description information. Descriptive guidance information includes, but is not limited to, question back information (e.g., where the occurrence area of the case is), advice information (e.g., the occurrence area of the advice supplemental case).

It should be noted that after the task description information of the target analysis task is obtained, the task description information and the structural analysis condition may be matched, and whether the task description information meets the structural analysis condition may be determined. And under the condition that the task description information does not meet the structural analysis condition, acquiring description guide information corresponding to the task description information, and sending the description guide information to a user. When the description guide information is acquired, a description guide template can be acquired, and relevant contents of which the task description information does not meet the structural analysis condition are filled into the description guide template to acquire the description guide information. And the related content of which the task description information does not meet the structural analysis condition can be received and input into a large model, and the description guide information is generated by utilizing the understanding and reasoning capability of the large model. The method for acquiring the guiding information is described, and is specifically selected according to practical situations, and the embodiment of the present disclosure does not limit the method.

In practical application, after receiving updated task description information sent by a user based on description guide information, the updated task description information may meet structural analysis conditions and may still not meet structural analysis conditions, at this time, whether the updated task description information meets the structural analysis conditions may be continuously judged, and under the condition that the updated task description information does not meet the structural analysis conditions, the description guide information is obtained, the description guide information is sent to the user until the updated task description information sent by the user is received and meets the structural analysis conditions, and then structural analysis is performed on the task description information to obtain task key information.

By applying the scheme of the embodiment of the specification, description guide information is sent to a user under the condition that task description information does not meet structural analysis conditions; receiving updated task description information sent by a user based on the description guide information; and under the condition that the task description information meets the structural analysis condition, carrying out structural analysis on the task description information to obtain the task key information. The process of manually screening the conditions by the user is changed into a mode of automatically understanding and extracting the task key information aiming at the task description information of the user, so that the task processing efficiency is improved, and in the process of analyzing the task key information, the user is flexibly decided, asked back and suggested to perfect the task description information by describing the guide information, so that the accuracy and the comprehensiveness of the task processing can be ensured.

Step 306: and retrieving at least one task reference data from the plurality of candidate data according to the task key information, wherein the task reference data is data related to the task key information in the plurality of candidate data.

In one or more embodiments of the present disclosure, task description information of a target analysis task is obtained; and after the task description information is subjected to structural analysis to obtain the task key information, further, at least one task reference data can be retrieved from a plurality of candidate data according to the task key information so as to analyze the at least one task reference data and generate a task processing result of the target analysis task.

In particular, data retrieval refers to the process of finding and returning candidate data related to a large number of candidate data sets through task key information. The plurality of candidate data may be data in databases of different types, or may be data in a target database corresponding to a task type of a target task. Candidate data may be data of different modalities including, but not limited to, text, speech, video, images. The number of the task reference data may be one or more, for example, 10 ten thousand. The task reference data is data related to task key information in the plurality of candidate data, and if the task key information is a plurality of, the task reference data can simultaneously satisfy the plurality of task key information.

Illustratively, with the target task as legal task, the task key information is "area: market A; time: 21 years; case type: XXX; amount of money: up to 20 ten thousand "is an example, the task reference data may be a case of satisfying task key information at the same time, that is, the task reference data is a plurality of legal candidate data of which the XXX amount of a city reaches 20 ten thousand in 21 years.

In practical applications, there are various ways of retrieving at least one task reference data from multiple candidate data according to the task key information, and the embodiment of the present disclosure does not limit the manner in which the at least one task reference data is retrieved. In one possible implementation manner of the present disclosure, data key information of a plurality of candidate data may be extracted, and candidate data having the same data key information as task key information may be determined as task reference data. When the data key information of each candidate data is extracted, the named entity identification can be performed on each candidate data to obtain the data key information, and the keyword extraction can also be performed on each candidate data to obtain the data key information. In another possible implementation manner of the present disclosure, since the number of candidate data may be massive, and a lot of resources are consumed to extract the data key information of the plurality of candidate data, at least one task reference data may be retrieved from the plurality of candidate data by using a data retrieval interface of a database where the plurality of candidate data is located.

In an optional embodiment of the present disclosure, before retrieving at least one task reference data from the plurality of candidate data according to the task key information, the method may further include the following steps:

Performing type recognition on the task description information to obtain a task type of a target analysis task, and determining a target database corresponding to the task type, wherein the target database comprises a plurality of candidate data;

Retrieving at least one task reference data from the plurality of candidate data according to the task key information may include the steps of:

And calling a data retrieval interface of the target database, and retrieving at least one task reference data from the plurality of candidate data according to the task key information.

In particular, the task type of the target analysis task may be determined based on the context of the target analysis task, i.e., the task type includes, but is not limited to, financial analysis type tasks, legal case study type tasks. Each task type corresponds to at least one database, and the database corresponding to the legal case study type task is a legal case library. The data retrieval interface is a programmatic way to allow a software application to call, search, and retrieve specific candidate data stored in the database through a predefined interface.

It should be noted that, the task type identification is performed on the task description information, and various manners of obtaining the task type of the target analysis task are selected specifically according to the actual situation, which is not limited in any way in the embodiment of the present specification. In one possible implementation manner of the present disclosure, a task type of a target analysis task may be searched from a keyword-type relationship mapping table according to a keyword in task description information. In another possible implementation manner of the present disclosure, the task description information may be input into a type recognition model, to obtain a task type of the target analysis task, where the type recognition model is obtained by training based on a plurality of sample description information and sample types corresponding to the sample description information.

In practical application, when the data retrieval interface of the target database is called and at least one task reference data is retrieved from a plurality of candidate data according to the task key information, a retrieval query statement can be constructed according to the task key information, the retrieval query statement is packaged in a request, the data retrieval interface is called by using a programming language, and at least one task reference data is retrieved from the plurality of candidate data.

By applying the scheme of the embodiment of the specification, the task description information is subjected to type recognition to obtain the task type of the target analysis task, and a target database corresponding to the task type is determined, wherein the target database comprises a plurality of candidate data; and calling a data retrieval interface of the target database, retrieving at least one task reference data from the plurality of candidate data according to the task key information, and retrieving the plurality of candidate data through reality, thereby ensuring the accuracy of the task reference data.

Step 308: and inputting at least one task reference data into a task processing model to obtain a task processing result of the target analysis task.

In one or more embodiments of the present disclosure, task description information of a target analysis task is obtained; carrying out structural analysis on the task description information to obtain task key information; and according to the task key information, at least one task reference data is obtained from the plurality of candidate data in a searching mode, and further, the at least one task reference data can be input into a task processing model to obtain a task processing result of the target analysis task.

Specifically, the task processing model may be a large model, or may be a model obtained by continuously training on data in a specified field with the large model as a base. Specifically, the task processing model may be trained based on a plurality of sample data and sample processing results respectively corresponding to the plurality of sample data. Task processing models include, but are not limited to, BERT models, T5 models. Task processing results include, but are not limited to, data aggregation analysis results, visualization processing results, natural language analysis text.

In practical application, when the task processing model is utilized to generate a task processing result of the target analysis task based on at least one task reference data, the at least one task reference data can be directly input into the task processing model to obtain the task processing result of the target analysis task. Further, in order to ensure the comprehensiveness of the task processing results, model prompt information can be constructed based on preset data analysis dimensions, the processing process of the task processing model is guided, and the task processing results corresponding to the data analysis dimensions are generated.

In an alternative embodiment of the present disclosure, the task processing results include text processing results; the step of inputting at least one task reference data into the task processing model to obtain a task processing result of the target analysis task may include the following steps:

Acquiring a preset data analysis dimension;

And carrying out statistical analysis on at least one task reference data based on the data analysis dimension through the task processing model to obtain a text processing result corresponding to the data analysis dimension.

In particular, the data analysis dimension is used to organize, measure, and analyze the angle of the data. By different data analysis dimensions, data can be observed and analyzed in depth from multiple angles. For example, the data analysis dimensions include, but are not limited to, a time dimension, a geographic dimension, a customer dimension, wherein the time dimension is derived based on time information contained in the data. The data analysis dimension may be obtained in various manners, and is specifically selected according to practical situations, which is not limited in any way in the embodiment of the present disclosure. In one possible implementation manner of the present disclosure, a preset data analysis dimension sent by a user may be received. In another possible implementation manner of the present disclosure, the preset data analysis dimension may be read from other data acquisition devices or databases.

In practical application, through the task processing model, when at least one task reference data is statistically analyzed based on the data analysis dimension, model prompt information can be constructed based on the data analysis dimension, and the model prompt information and the at least one task reference data are spliced and input into the task processing model, so that a text processing result is obtained.

For example, assuming the data analysis dimension is the time dimension, the model hint information may be "you are a professional statistics analyst, please analyze { at least one task reference data } from the time dimension to generate a text processing result.

By applying the scheme of the embodiment of the specification, the preset data analysis dimension is obtained; and carrying out statistical analysis on at least one task reference data based on the data analysis dimension through the task processing model to obtain a text processing result corresponding to the data analysis dimension. The data analysis dimension is integrated in the process of generating the text processing result based on the task reference data, so that a result corresponding to the data analysis dimension can be accurately generated, and the accuracy of the text processing result is improved.

In the embodiment of the specification, after the text processing result of the target analysis task is obtained, a service tool can be called, and the target processing result of the target analysis task is generated based on the text processing result, so that processing results other than the text mode are generated, and the diversity of the processing results is improved. Among other things, service tools include, but are not limited to, visualization transformation tools, calculators, and form generation tools.

In an optional embodiment of the present disclosure, taking a service tool as an example of a visual conversion tool, the task processing result further includes a visual processing result; the task processing model performs statistical analysis on at least one task reference data based on the data analysis dimension, and after obtaining a text processing result corresponding to the data analysis dimension, the task processing model may further include the following steps:

And calling a visual conversion tool, and generating a visual processing result of the target analysis task according to the text processing result.

Specifically, the visual conversion tool is used for converting text processing results into visual processing results. Visualization processing results include, but are not limited to, image processing results, chart processing results, portable document format (PDF, portable Document Format) processing results. For example, the text processing result is 20 ten thousand profit in 21 years, 23 ten thousand profit in 22 years and 40 ten thousand profit in 23 years, and then the visual conversion tool can be called to generate a change broken line statistical graph of profit.

By applying the scheme of the embodiment of the specification, a visual conversion tool is called, and a visual processing result of the target analysis task is generated according to the text processing result, so that visual presentation of the task processing result is realized, and the diversity of the task processing result and the user experience are improved.

In an optional embodiment of the present disclosure, after invoking the visual transformation tool to generate the visual processing result of the target analysis task according to the text processing result, the method may further include the following steps:

Typesetting the text processing result and the visual processing result to generate a graphic processing report of the target analysis task.

In particular, the process of typesetting the text processing result and the visual processing result may be referred to as visual typesetting, such as graphic typesetting. The graphic typesetting is to combine text content with visual content, such as charts, and perform reasonable position arrangement, size adjustment, color matching, style unification and the like to create a report with good reading performance and visual attraction.

In practical application, typesetting is performed on the text processing result and the visual processing result, and various modes of generating the graphic processing report of the target analysis task are provided, and the method is specifically selected according to practical situations, and the embodiment of the specification does not limit the method. In one possible implementation manner of the present disclosure, an image-text report template may be obtained, and the text processing result and the visual processing result are filled into the image-text report template for typesetting, so as to obtain an image-text processing report. In another possible implementation manner of the present disclosure, a typesetting tool (such as office software and an online graphic editor) may be called to typeset the text processing result and the visual processing result, so as to generate a graphic processing report of the target analysis task.

By applying the scheme of the embodiment of the specification, typesetting is performed on the text processing result and the visual processing result, and the image-text processing report of the target analysis task is generated, so that visual presentation of the task processing result is realized, and the chart is explained through the text, so that the processing result is easier to understand, and the diversity of the task processing result and the user experience are improved.

In an alternative embodiment of the present disclosure, after inputting at least one task reference data into the task processing model to obtain a task processing result of the target analysis task, the method may further include the following steps:

and receiving updated task description information sent by a user based on the task processing result, and performing task processing based on the updated task description information to obtain the updated task processing result.

It should be noted that, after obtaining the task processing result of the target analysis task, the task processing result may be sent to the client, so that the client may display the task processing result to the user. The client side displays the task processing result to the user in various manners, and the method is specifically selected according to the actual situation, and the embodiment of the present disclosure is not limited in any way. In one possible implementation manner of the present disclosure, the task processing result may be directly displayed to the user. In another possible implementation manner of the present disclosure, the task processing result may be displayed to the user according to the display requirement information of the user. The display requirement information characterizes the requirement of a user for checking the task processing result, and includes but is not limited to downloading links for displaying the task processing result, and related information for displaying the task processing result and task reference data.

In practical applications, the user may not be satisfied with the task processing result, for example, the user wants to obtain a result corresponding to other data analysis dimensions, at this time, updated task description information sent by the user based on the task processing result may be received, and multiple rounds of task processing may be performed according to the updated task description information, so as to obtain an updated task processing result. The implementation manner of performing task processing based on the updated task description information and obtaining the updated task processing result is the same as that of the task processing method, and the embodiments of the present disclosure will not be described in detail.

By applying the scheme of the embodiment of the specification, updated task description information sent by a user based on the task processing result is received, task processing is performed based on the updated task description information, the updated task processing result is obtained, the accuracy of the task processing result is improved, and meanwhile, the user experience is improved.

In an optional embodiment of the present disclosure, after the inputting the at least one task reference data into the task processing model and obtaining the task processing result of the target analysis task, the method may further include the following steps:

And inputting the task processing result into a statistical reasoning model to obtain a task reasoning result, wherein the statistical reasoning model is obtained based on a plurality of sample data and sample reasoning results respectively corresponding to the plurality of sample data.

Specifically, the statistical reasoning model is used for carrying out statistical analysis on the input task processing result, and the task reasoning result is obtained through secondary summarization. The statistical reasoning model can be a generated large model, or can be a model which is obtained by taking the large model as a base and continuously training on data in the appointed field, such as a legal task processing model and an academic task processing model. Statistical inference models include, but are not limited to, BERT models, T5 models, LLaMA (Large Language Model Meta AI) models, GPT (GENERATIVE PRE-trained Transformers) models.

It should be noted that, the process of training to obtain the statistical inference model based on the plurality of sample data and the sample inference results corresponding to the plurality of sample data respectively may include the following steps: acquiring a plurality of sample data, wherein the sample data carries a sample reasoning result; inputting a plurality of sample data into a statistical reasoning model to obtain a prediction reasoning result corresponding to the plurality of sample data respectively; and adjusting model parameters of the statistical inference model according to the sample inference result and the prediction inference result to obtain the statistical inference model after training. The sample data are processing objects of the model training process. The sample reasoning result is a real reasoning result corresponding to the sample data, and the sample reasoning result is an analytic target of the information analytic model. The implementation manner of "adjusting the model parameters of the statistical inference model according to the sample inference result and the prediction inference result" is the same as the implementation manner of "adjusting the model parameters of the information analysis model according to the sample key information and the prediction key information" described above, and will not be described in detail in the embodiment of the present specification.

It should be noted that, after the task processing result is input into the statistical reasoning model and the task reasoning result is obtained, the task reasoning result may be sent to the client, so that the client may display the task reasoning result to the user.

By applying the scheme of the embodiment of the specification, the task processing result is input into the statistical reasoning model, the task reasoning result is obtained, the statistical reasoning model is utilized to understand and secondarily generate summary reasoning for the task processing result, and the result richness of the target analysis task is enriched and expanded.

The following describes the task processing method provided in the present specification by taking an application of the task processing method in an intelligent question-answering scenario as an example with reference to fig. 4. Fig. 4 shows a flowchart of an automatic question-answering method according to an embodiment of the present disclosure, which specifically includes the following steps:

step 402: and acquiring task description information of the target question-answering task.

Step 404: and carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes candidate data screening intents corresponding to the target question-answering task.

Step 406: and retrieving at least one task reference data from the plurality of candidate data according to the task key information, wherein the task reference data is data related to the task key information in the plurality of candidate data.

Step 408: and inputting at least one task reference data into a task processing model to obtain a reply result of the target question-and-answer task.

It should be noted that, the implementation manners of step 402 to step 408 may refer to the implementation manners of step 302 to step 308, and the description of the embodiment of the present disclosure is omitted.

By applying the scheme of the embodiment of the specification, the task key information is obtained through analysis from the task description information, so that the candidate data screening condition is accurately determined, the task key information is further utilized to automatically search data, a plurality of candidate data are truly searched, the accuracy of task reference data is ensured, meanwhile, the task processing model is utilized to process the screened task reference data, the complete automatic question-answering flow is realized, and the automatic question-answering efficiency and the accuracy of answer results are improved.

The task processing method provided in the present specification will be further described with reference to fig. 5 by taking an application of the task processing method in a legal research scenario as an example. Fig. 5 shows a flowchart of a legal task processing method according to an embodiment of the present disclosure, which specifically includes the following steps:

Step 502: task description information of legal tasks is obtained.

Step 504: and carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes legal candidate data screening intention corresponding to legal tasks.

Step 506: and retrieving at least one legal task reference data from the plurality of legal candidate data according to the task key information, wherein the legal task reference data is data related to the task key information in the plurality of legal candidate data.

Step 508: and inputting at least one legal task reference data into a task processing model to obtain a task processing result of the legal task.

It should be noted that, the implementation manners of the steps 502 to 508 may refer to the implementation manners of the steps 302 to 308, and the description of the embodiment of the present disclosure is omitted.

By applying the scheme of the embodiment of the specification, the task key information is obtained through analysis from the task description information, so that the case screening condition is accurately determined, the case retrieval is further automatically carried out by utilizing the task key information, and a plurality of legal candidate data are truly retrieved, so that the accuracy of legal task reference data is ensured, and meanwhile, the legal task reference data obtained through screening is processed by utilizing a task processing model, so that a complete automatic legal case study and judgment flow is realized, and the legal case study and judgment efficiency and the accuracy of task processing results are improved.

In an optional embodiment of the present disclosure, after the at least one legal task reference data is input into the task processing model and the task processing result of the legal task is obtained, the method may further include the following steps:

extracting a key processing result from the task processing result, and sending the key processing result to the client;

receiving result adjustment information sent by a client, wherein the result adjustment information is obtained by a user of the client performing result adjustment based on a key processing result;

and adjusting the key processing result according to the result adjustment information to obtain an updated key processing result.

Specifically, the key processing result is key information in the task processing result, such as a keyword, abstract, subject matter, and the like. The result adjustment information characterizes the adjustment requirements for the key process results, including, but not limited to, result translation information, result editing information. For example, the result adjustment information may be "please mark out the key processing result", "please add the comment of the key processing result", "please generate the legend of the key processing result", or the like.

It should be noted that, there are various ways to extract the key processing result from the task processing result, and the method is specifically selected according to the actual situation, which is not limited in any way in the embodiment of the present disclosure. In one possible implementation of the present disclosure, key processing results may be extracted from task processing results using a keyword extraction algorithm, such as word Frequency-inverse document Frequency (TF-IDF, term Frequency-Inverse Document Frequency). In another possible implementation manner of the present disclosure, a task processing result may be input into a key information extraction model to obtain a key processing result.

By applying the scheme of the embodiment of the specification, the key processing result is extracted from the task processing result and is sent to the client; receiving result adjustment information sent by a client, wherein the result adjustment information is obtained by a user of the client performing result adjustment based on a key processing result; and adjusting the key processing result according to the result adjustment information to obtain an updated key processing result. Interaction with a user is increased, and task processing flexibility is improved.

And receiving processing feedback information sent by the client, and carrying out parameter adjustment on the task processing model according to the processing feedback, wherein the processing feedback information is information obtained by processing feedback of a user of the client based on a task processing result.

Specifically, the processing feedback information may be an adjusted task processing result, and at this time, a sample of the training task processing model may be updated based on the task description information and the adjusted task processing result, so as to further fine-tune the task processing model. The processing feedback information can also be the adjusted task description information and the adjusted task processing result, and at this time, the sample of the training task processing model can be updated based on the adjusted task description information and the adjusted task processing result, so as to further finely adjust the task processing model.

In practical application, the implementation manner of "performing parameter adjustment on the task processing model according to the processing feedback" is the same as the training manner of the task processing model, and the embodiments of the present disclosure will not be described in detail.

By applying the scheme of the embodiment of the specification, the processing feedback information sent by the client is received, and the parameter adjustment is carried out on the task processing model according to the processing feedback, so that the accuracy of the task processing model is improved, and meanwhile, the user experience is improved.

Referring to fig. 6a, fig. 6a shows a process flow chart of a legal task processing method according to an embodiment of the present disclosure, specifically including:

Information rewriting: receiving task description information of legal case research and judgment tasks sent by a user, namely 'group me to summarize cases with the XXX amount reaching 20 ten thousand in 21 years in A city, and summarize a research and judgment report'; judging whether the task description information meets the structural analysis condition or not; if not, the description guide information is sent to the user so that the user updates the description guide information; if yes, carrying out structural analysis on the task description information to obtain a task key information area: market A; time: 21 years; case type: XXX; amount of money: reaching 20 ten thousand ", and entering a class case retrieval stage;

And (5) class case retrieval: invoking a case retrieval interface of a case library, retrieving at least one task reference case from a plurality of cases according to the task key information, and entering a dimension analysis stage;

Dimension analysis: judging whether the case analysis dimension is included; if not, dimension guide information is sent to the user, so that the user sends a case analysis dimension according to actual requirements; if yes, carrying out statistical analysis on at least one task reference case based on the case analysis dimension through the task processing model to obtain a text processing result corresponding to the case analysis dimension, and entering a statistical description stage;

statistical description: inputting the text processing result into a statistical reasoning model to perform secondary statistical reasoning, obtaining a text reasoning result, and entering a report generation stage;

Report generation: judging whether a report format is included; if not, sending format guiding information to the user so that the user can send report formats, such as a portable document format and a double-column portable document format, according to actual requirements; if yes, calling a chart generation tool, and generating a chart processing result according to the text reasoning result; typesetting the text reasoning results and the chart processing results (such as generating reports in a certain order and modifying the titles of certain parts into new contents) to generate the research and judgment graphic reports of the legal case research and judgment tasks.

Referring to fig. 6b, fig. 6b shows a schematic diagram of a research and judgment graphic report in a legal task processing method according to an embodiment of the present disclosure, where the research and judgment graphic report generated by the model includes a report title, a subtitle, and a text corresponding to the subtitle. As shown in fig. 6b, the title of the research report is "research report about xxxxxxxxx", the text corresponding to the subtitle "first, basic analysis of case" includes a histogram and a line diagram, and the text corresponding to the histogram is "left: xxxxxxx ", the text corresponding to the line diagram is" right diagram: the text corresponding to the subheading "second, principal feature analysis" includes a table and a pie chart, and the text corresponding to the table is "left table: xxxxxxxx ", text corresponding to the pie chart is" right chart: xxxxxxx.

It is worth to say that, in the report generation process, the embodiment of the specification supports the user to upload the graphic report template, to customize the existing graphic report template, and to generate the graphic report template according to the customized template elements.

By applying the scheme of the embodiment of the specification, in the legal case research and judgment process, a user can be guided to perfect task description information through dialogue drive, task key information is analyzed from the task description information, so that document screening conditions are accurately determined, document retrieval is further automatically carried out by utilizing the task key information, a plurality of documents are truly retrieved, accuracy of task reference documents is guaranteed, the illusion problem in the task processing process is avoided, and meanwhile, the task reference documents obtained through retrieval are subjected to secondary generation reasoning summary by utilizing a task processing model, so that research and judgment image-text reports are perfected.

Referring to fig. 7, fig. 7 illustrates an interface schematic diagram of a task processing interface provided in an embodiment of the present disclosure. The task processing interface is divided into a request input interface and a result display interface. The request input interface includes a request input box, a "determine" control, and a "cancel" control. The result display interface comprises a result display frame.

The method comprises the steps that a user inputs a task processing request through a request input box displayed by a client, wherein the task processing request carries task description information of a target analysis task, a 'determination' control is clicked, a server receives the task description information of the target analysis task sent by the client, the task description information is subjected to structural analysis, task key information is obtained, and the task key information characterizes candidate data screening intention corresponding to the target analysis task; according to the task key information, retrieving at least one task reference data from a plurality of candidate data, wherein the task reference data is data related to the task key information in the plurality of candidate data; inputting at least one task reference data into a task processing model to obtain a task processing result of a target analysis task; and sending the task processing result to the client. And the client displays the task processing result in a result display frame.

In practical applications, the manner in which the user operates the control includes any manner such as clicking, double clicking, touch control, mouse hovering, sliding, long pressing, voice control or shaking, and the like, and the selection is specifically performed according to the practical situation, which is not limited in any way in the embodiments of the present disclosure.

Corresponding to the task processing method embodiment, the present disclosure further provides a task processing device embodiment, and fig. 8 shows a schematic structural diagram of a task processing device provided in one embodiment of the present disclosure. As shown in fig. 8, the apparatus includes:

a first obtaining module 802 configured to obtain task description information of a target analysis task;

The first parsing module 804 is configured to perform structural parsing on the task description information to obtain task key information, where the task key information characterizes candidate data screening intents corresponding to the target analysis task;

A first retrieving module 806, configured to retrieve at least one task reference data from the plurality of candidate data according to the task key information, where the task reference data is data related to the task key information in the plurality of candidate data;

The first input module 808 is configured to input at least one task reference data into the task processing model to obtain a task processing result of the target analysis task.

Optionally, the apparatus further comprises: the sending module is configured to send the description guide information to the user under the condition that the task description information does not meet the structural analysis condition; receiving updated task description information sent by a user based on the description guide information; the first parsing module 804 is further configured to perform structural parsing on the task description information to obtain task key information when the task description information meets the structural parsing condition.

Optionally, the first parsing module 804 is further configured to input the task description information into an information parsing model to obtain the task key information, where the information parsing model is obtained by training based on the plurality of sample description information and sample key information corresponding to the plurality of sample description information.

Optionally, the apparatus further comprises: the identification module is configured to carry out type identification on the task description information, obtain the task type of the target analysis task and determine a target database corresponding to the task type, wherein the target database comprises a plurality of candidate data; the first retrieval module 806 is further configured to invoke a data retrieval interface of the target database to retrieve at least one task reference data from the plurality of candidate data according to the task key information.

Optionally, the task processing result includes a text processing result; the first input module 808 is further configured to obtain a preset data analysis dimension; and carrying out statistical analysis on at least one task reference data based on the data analysis dimension through the task processing model to obtain a text processing result corresponding to the data analysis dimension.

Optionally, the task processing result further includes a visual processing result; the apparatus further comprises: the first generation module is configured to call a visual conversion tool and generate a visual processing result of the target analysis task according to the text processing result.

Optionally, the apparatus further comprises: the receiving module is configured to receive updated task description information sent by a user based on the task processing result, and perform task processing based on the updated task description information to obtain the updated task processing result.

Optionally, the apparatus further comprises: and the fourth input module is configured to input the task processing result into the statistical reasoning model to obtain the task reasoning result, wherein the statistical reasoning model is obtained by training based on a plurality of sample data and sample reasoning results respectively corresponding to the plurality of sample data.

The above is a schematic solution of a task processing device of the present embodiment. It should be noted that, the technical solution of the task processing device and the technical solution of the task processing method belong to the same concept, and details of the technical solution of the task processing device, which are not described in detail, can be referred to the description of the technical solution of the task processing method.

Corresponding to the above-mentioned automatic question-answering method embodiment, the present disclosure further provides an automatic question-answering device embodiment, and fig. 9 shows a schematic structural diagram of an automatic question-answering device provided in one embodiment of the present disclosure. As shown in fig. 9, the apparatus includes:

A second obtaining module 902 configured to obtain task description information of a target question-answer task;

The second parsing module 904 is configured to perform structural parsing on the task description information to obtain task key information, where the task key information characterizes candidate data screening intents corresponding to the target question-answering task;

A second retrieving module 906 configured to retrieve at least one task reference data from the plurality of candidate data according to the task key information, where the task reference data is data related to the task key information in the plurality of candidate data;

A second input module 908 is configured to input at least one task reference data into the task processing model to obtain a reply result of the target question-and-answer task.

The above is a schematic scheme of an automatic question answering apparatus of this embodiment. It should be noted that, the technical solution of the automatic question-answering device and the technical solution of the automatic question-answering method belong to the same concept, and details of the technical solution of the automatic question-answering device, which are not described in detail, can be referred to the description of the technical solution of the automatic question-answering method.

Corresponding to the above legal task processing method embodiment, the present disclosure further provides an embodiment of a legal case study and judgment device, and fig. 10 shows a schematic structural diagram of a legal case study and judgment device provided in one embodiment of the present disclosure. As shown in fig. 10, the apparatus includes:

A third obtaining module 1002 configured to obtain task description information of legal tasks;

the third parsing module 1004 is configured to perform structural parsing on the task description information to obtain task key information, where the task key information characterizes legal candidate data screening intention corresponding to legal tasks;

a third retrieving module 1006 configured to retrieve at least one legal task reference data from the plurality of legal candidate data according to the task key information, where the legal task reference data is data related to the task key information in the plurality of legal candidate data;

and a third input module 1008 configured to input at least one legal task reference data into the task processing model to obtain a task processing result of the legal task.

Optionally, the apparatus further comprises: the extraction module is configured to extract a key processing result from the task processing result and send the key processing result to the client; receiving result adjustment information sent by a client, wherein the result adjustment information is obtained by a user of the client performing result adjustment based on a key processing result; and adjusting the key processing result according to the result adjustment information to obtain an updated key processing result.

Optionally, the apparatus further comprises: the adjustment module is configured to receive processing feedback information sent by the client and adjust parameters of the task processing model according to the processing feedback, wherein the processing feedback information is information obtained by processing feedback of a user of the client based on a task processing result.

The foregoing is a schematic illustration of a legal case study and judgment device of the present embodiment. It should be noted that, the technical solution of the legal case research and determination device and the technical solution of the legal task processing method belong to the same concept, and details of the technical solution of the legal case research and determination device which are not described in detail can be referred to the description of the technical solution of the legal task processing method.

FIG. 11 illustrates a block diagram of a computing device provided in one embodiment of the present description. The components of computing device 1100 include, but are not limited to, a memory 1110 and a processor 1120. Processor 1120 is coupled to memory 1110 via bus 1130, and database 1150 is used to hold data.

The computing device 1100 also includes an access device 1140, the access device 1140 enabling the computing device 1100 to communicate via one or more networks 1160. Examples of such networks include public switched telephone networks (PSTN, public Switched Telephone Network), local area networks (LAN, local Area Network), wide area networks (WAN, wide Area Network), personal area networks (PAN, personal Area Network), or combinations of communication networks such as the internet. The access device 1140 may comprise one or more of any type of Network interface, wired or wireless, such as, for example, a Network interface card (NIC, network INTERFACE CARD), such as an IEEE802.11 wireless local area Network (WLAN, wireless Local Area Networks) wireless interface, a worldwide interoperability for microwave access (Wi-MAX, world Interoperability for Microwave Access) interface, an ethernet interface, a universal serial bus (USB, universal Serial Bus) interface, a cellular Network interface, a bluetooth interface, a Near Field Communication (NFC) interface, and so forth.

In one embodiment of the present description, the above components of computing device 1100, as well as other components not shown in FIG. 11, may also be connected to each other, such as by a bus. It should be understood that the block diagram of the computing device illustrated in FIG. 11 is for exemplary purposes only and is not intended to limit the scope of the present description. Those skilled in the art may add or replace other components as desired.

Computing device 1100 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., tablet, personal digital assistant, laptop, notebook, netbook, etc.), mobile phone (e.g., smart phone), wearable computing device (e.g., smart watch, smart glasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or personal computer (PC, personal Computer). Computing device 1100 may also be a mobile or stationary server.

The processor 1120 is configured to execute a computer program/instruction, where the computer program/instruction when executed by the processor implements the steps of the task processing method, the automatic question-answering method, or the legal task processing method.

The foregoing is a schematic illustration of a computing device of this embodiment. It should be noted that, the technical solution of the computing device belongs to the same concept as the technical solution of the task processing method, the automatic question-answering method and the legal task processing method, and details of the technical solution of the computing device which are not described in detail can be referred to the description of the technical solution of the task processing method, the automatic question-answering method or the legal task processing method.

An embodiment of the present specification also provides a computer-readable storage medium storing a computer program/instruction which, when executed by a processor, implements the steps of the task processing method or the automatic question-answering method or the legal task processing method described above.

The above is an exemplary version of a computer-readable storage medium of the present embodiment. It should be noted that, the technical solution of the storage medium belongs to the same concept as the technical solution of the task processing method, the automatic question-answering method and the legal task processing method, and details of the technical solution of the storage medium which are not described in detail can be referred to the description of the technical solution of the task processing method, the automatic question-answering method or the legal task processing method.

An embodiment of the present specification also provides a computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the task processing method or the automatic question-answering method or the legal task processing method described above.

The foregoing is a schematic version of a computer program product of this embodiment. It should be noted that, the technical solution of the computer program product and the technical solutions of the task processing method, the automatic question answering method and the legal task processing method belong to the same concept, and the details of the technical solution of the computer program product, which are not described in detail, can be referred to the description of the technical solutions of the task processing method, the automatic question answering method or the legal task processing method.

The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

The computer instructions include computer program code that may be in source code form, object code form, executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth. It should be noted that the content of the computer readable medium can be increased or decreased appropriately according to the requirements of the patent practice, for example, in some areas, according to the patent practice, the computer readable medium does not include an electric carrier signal and a telecommunication signal.

It should be noted that, for simplicity of description, the foregoing method embodiments are all expressed as a series of combinations of actions, but it should be understood by those skilled in the art that the embodiments are not limited by the order of actions described, as some steps may be performed in other order or simultaneously according to the embodiments of the present disclosure. Further, those skilled in the art will appreciate that the embodiments described in the specification are all preferred embodiments, and that the acts and modules referred to are not necessarily all required for the embodiments described in the specification.

In the foregoing embodiments, the descriptions of the embodiments are emphasized, and for parts of one embodiment that are not described in detail, reference may be made to the related descriptions of other embodiments.

The preferred embodiments of the present specification disclosed above are merely used to help clarify the present specification. Alternative embodiments are not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the teaching of the embodiments. The embodiments were chosen and described in order to best explain the principles of the embodiments and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. This specification is to be limited only by the claims and the full scope and equivalents thereof.

Claims

1. A task processing method, comprising:

acquiring task description information of a target analysis task;

Retrieving at least one task reference data from a plurality of candidate data according to the task key information, wherein the task reference data is data related to the task key information in the plurality of candidate data;

And inputting the at least one task reference data into a task processing model to obtain a task processing result of the target analysis task.

2. The method of claim 1, wherein the performing structural parsing on the task description information, before obtaining task key information, further comprises:

under the condition that the task description information does not meet the structural analysis condition, the description guide information is sent to a user;

receiving updated task description information sent by the user based on the description guide information;

the step of carrying out structural analysis on the task description information to obtain task key information comprises the following steps:

3. The method according to claim 1 or 2, wherein the performing structural parsing on the task description information to obtain task key information includes:

And inputting the task description information into an information analysis model to obtain task key information, wherein the information analysis model is obtained based on a plurality of sample description information and sample key information respectively corresponding to the plurality of sample description information in a training way.

4. The method of claim 1, further comprising, prior to retrieving at least one task reference data from a plurality of candidate data based on the task key information:

Performing type recognition on the task description information to obtain a task type of the target analysis task, and determining a target database corresponding to the task type, wherein the target database comprises a plurality of candidate data;

the retrieving, according to the task key information, at least one task reference data from a plurality of candidate data includes:

and calling a data retrieval interface of the target database, and retrieving at least one task reference data from a plurality of candidate data according to the task key information.

5. The method of claim 1, the task processing results comprising text processing results;

the step of inputting the at least one task reference data into a task processing model to obtain a task processing result of the target analysis task includes:

Acquiring a preset data analysis dimension;

And carrying out statistical analysis on the at least one task reference data based on the data analysis dimension through the task processing model to obtain a text processing result corresponding to the data analysis dimension.

6. The method of claim 5, the task processing results further comprising visualization processing results;

The performing statistical analysis on the at least one task reference data based on the data analysis dimension through the task processing model, after obtaining a text processing result corresponding to the data analysis dimension, further includes:

7. A method according to any one of claims 1-6, wherein the inputting the at least one task reference data into a task processing model, after obtaining the task processing result of the target analysis task, further comprises:

8. A method according to any one of claims 1-6, wherein the inputting the at least one task reference data into a task processing model, after obtaining the task processing result of the target analysis task, further comprises:

9. An automatic question-answering method, comprising:

Acquiring task description information of a target question-answering task;

And inputting the at least one task reference data into a task processing model to obtain a reply result of the target question-and-answer task.

10. A legal task processing method, comprising:

acquiring task description information of legal tasks;

carrying out structural analysis on the task description information to obtain task key information, wherein the task key information characterizes legal candidate data screening intention corresponding to the legal task;

And inputting the at least one legal task reference data into a task processing model to obtain a task processing result of the legal task.

11. The method of claim 10, wherein the inputting the at least one legal task reference data into a task processing model, after obtaining a task processing result of the legal task, further comprises:

Extracting a key processing result from the task processing result, and sending the key processing result to a client;

receiving result adjustment information sent by the client, wherein the result adjustment information is obtained by performing result adjustment on the basis of the key processing result by a user of the client;

12. The method of claim 10, wherein the inputting the at least one legal task reference data into a task processing model, after obtaining a task processing result of the legal task, further comprises:

And receiving processing feedback information sent by a client, and carrying out parameter adjustment on the task processing model according to the processing feedback, wherein the processing feedback information is information obtained by processing feedback of a user of the client based on the task processing result.

13. A computing device, comprising:

A memory and a processor;

The memory is adapted to store a computer program/instruction, the processor being adapted to execute the computer program/instruction, which when executed by the processor, implements the steps of the method of any one of claims 1 to 8 or 9 or any one of claims 10 to 12.

14. A computer readable storage medium storing a computer program/instruction which when executed by a processor performs the steps of the method of any one of claims 1 to 8 or claim 9 or any one of claims 10 to 12.

15. A computer program product comprising computer programs/instructions which when executed by a processor implement the steps of the method of any one of claims 1 to 8 or claim 9 or any one of claims 10 to 12.