Method, system and equipment for extracting key appeal elements and medium

Abstract

The invention provides a key appeal element extraction method, system and device and a medium, and belongs to the technical field of element extraction. According to the method, a training set, a verification set and a test set are constructed by utilizing obtained word segmentation results. Constructing an entity extraction model and an abstract extraction model, training the entity extraction model and the abstract extraction model based on the constructed training set, and evaluating the performance of the entity extraction model and the abstract extraction model based on the constructed verification set. And testing the training results of the entity extraction model and the abstract extraction model based on the constructed test set to obtain the trained entity extraction model and abstract extraction model. And inputting the work order text data and the telephone recording data of the key appeal elements to be extracted into the trained entity extraction model and the abstract extraction model to extract the key appeal elements to obtain an extraction result of the key appeal elements. According to the method, the problem of low accuracy of key appeal element extraction caused by limited generalization ability is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of factor extraction, and specifically relates to a method, system, device and medium for extracting key demand factors. Background Art

[0002] With the rapid development of big data and artificial intelligence technology, information processing technology has become an indispensable part of all walks of life. In the current context, all major companies and organizations are facing challenges in the management and analysis of massive data. How to quickly and accurately extract valuable information from massive data has become the key.

[0003] Although some people have proposed methods for extracting key appeal elements based on natural language processing and machine learning in the existing technology, these methods have the following problems when processing large-scale and diverse data sets: When faced with complex and ever-changing language expressions and contexts, the current methods for extracting key appeal elements have limited generalization capabilities, resulting in low accuracy in extracting key appeal elements. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system, device and medium for extracting key demand elements, so as to solve the problem in the prior art that the accuracy of extracting key demand elements is low due to limited generalization ability.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for extracting key appeal elements, comprising the following steps: Obtain work order text data and telephone recording data, pre-process the obtained work order text data and telephone recording data, and remove invalid data and noise; Perform word segmentation on the pre-processed work order text data and telephone recording data to obtain word segmentation results; Use the obtained word segmentation results to construct training sets, validation sets, and test sets; Constructing an entity extraction model and a summary extraction model, training the entity extraction model and the summary extraction model based on the constructed training set, evaluating the performance of the entity extraction model and the summary extraction model based on the constructed validation set, and testing the training results of the entity extraction model and the summary extraction model based on the constructed test set to obtain the trained entity extraction model and the summary extraction model; The work order text data and telephone recording data of the key demand elements to be extracted are input into the trained entity extraction model and summary extraction model to extract the key demand elements and obtain the extraction results of the key demand elements.

[0006] A further improvement of the present invention is that an adaptive filtering algorithm is used to remove noise from the obtained work order text data and telephone recording data.

[0007] A further improvement of the present invention is that the jieba word segmentation tool is used to perform word segmentation processing on the pre-processed work order text data and telephone recording data to obtain a word segmentation result.

[0008] A further improvement of the present invention is that the indicator for evaluating the performance of the entity extraction model and the summary extraction model is accuracy.

[0009] A further improvement of the present invention is that the key elements of the claim include personnel, time, address and the name of the organization involved.

[0010] A further improvement of the present invention is that after obtaining the extraction results of the key demand elements, the key demand elements are classified to obtain a number of classification results.

[0011] A further improvement of the present invention is that after obtaining a number of classification results, the number of classification results can be displayed and queried.

[0012] In a second aspect, the present invention provides a key appeal element extraction system, including a data acquisition module, a data preprocessing module, a word segmentation processing module, a data set construction module, a model training module and a key appeal element extraction module; The data acquisition module is used to acquire work order text data and telephone recording data; The data preprocessing module is used to preprocess the obtained work order text data and telephone recording data to remove invalid data and noise; The word segmentation processing module is used to perform word segmentation processing on the pre-processed work order text data and telephone recording data to obtain a word segmentation result; The data set construction module is used to construct a training set, a verification set and a test set using the obtained word segmentation results; The model training module is used to construct an entity extraction model and a summary extraction model, train the entity extraction model and the summary extraction model based on the constructed training set, evaluate the performance of the entity extraction model and the summary extraction model based on the constructed validation set, and test the training results of the entity extraction model and the summary extraction model based on the constructed test set to obtain the trained entity extraction model and the summary extraction model; The key demand element extraction module is used to input the work order text data and telephone recording data of the key demand elements to be extracted into the trained entity extraction model and summary extraction model to extract the key demand elements and obtain the extraction results of the key demand elements.

[0013] In a third aspect, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method for extracting key appeal elements introduced above when executing the computer program.

[0014] In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for extracting key appeal elements introduced above.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention belongs to an improved invention. Compared with the existing method for extracting key demand elements, the method for extracting key demand elements proposed by the present invention inputs the work order text data and telephone recording data of the key demand elements to be extracted into the trained entity extraction model and summary extraction model to extract the key demand elements, and obtains the extraction results of the key demand elements. It can be seen that after the entity extraction model and the summary extraction model are trained, the present invention can input the work order text data and telephone recording data of the key demand elements to be extracted into the trained entity extraction model and summary extraction model to extract the key demand elements, and judge whether the accuracy of each key demand element reaches the set threshold, and then extract the key demand elements that meet the requirements. In addition, the present invention pre-processes the obtained work order text data and telephone recording data, eliminates some invalid information, and improves the accuracy of the later extraction of key demand elements, thereby effectively solving the problem of low accuracy of key demand element extraction due to limited generalization ability in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of a method for extracting key appeal elements of the present invention; Figure 2 A schematic diagram of a system for extracting key elements of the present invention; Figure 3 It is a schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0017] In order to further understand the content of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the embodiments are only for explaining the present invention and are not intended to limit it.

[0018] The method for extracting key demand elements proposed in the present invention inputs the work order text data and telephone recording data of the key demand elements to be extracted into the trained entity extraction model and summary extraction model to extract the key demand elements, and obtains the extraction results of the key demand elements. Compared with the prior art, the present invention effectively solves the problem of low accuracy in extracting key demand elements due to limited generalization ability in the prior art.

[0019] Embodiment 1: The flowchart of the method for extracting key appeal elements of the present invention is as follows: Figure 1 As shown, the method for extracting key appeal elements of the present invention includes the following steps: S1. Obtain work order text data and telephone recording data, pre-process the obtained work order text data and telephone recording data, and remove invalid data and noise.

[0020] S2. Perform word segmentation on the preprocessed work order text data and telephone recording data to obtain word segmentation results.

[0021] S3. Use the obtained word segmentation results to construct training sets, validation sets, and test sets.

[0022] S4. Construct an entity extraction model and a summary extraction model, train the entity extraction model and the summary extraction model based on the constructed training set, evaluate the performance of the entity extraction model and the summary extraction model based on the constructed validation set, and test the training results of the entity extraction model and the summary extraction model based on the constructed test set to obtain the trained entity extraction model and the summary extraction model.

[0023] S5. Input the work order text data and telephone recording data of the key demand elements to be extracted into the trained entity extraction model and summary extraction model to extract the key demand elements and obtain the extraction results of the key demand elements.

[0024] Embodiment 2: The schematic diagram of the extraction system of the key appeal elements of the present invention is as follows Figure 2 As shown, the key appeal element extraction system of the present invention includes a data acquisition module, a data preprocessing module, a word segmentation processing module, a data set construction module, a model training module and a key appeal element extraction module.

[0025] The data acquisition module is used to obtain work order text data and telephone recording data.

[0026] The data preprocessing module is used to preprocess the obtained work order text data and telephone recording data to remove invalid data and noise.

[0027] The word segmentation processing module is used to perform word segmentation on the pre-processed work order text data and telephone recording data to obtain the word segmentation results.

[0028] The dataset construction module is used to construct training sets, validation sets, and test sets using the obtained word segmentation results.

[0029] The model training module is used to construct an entity extraction model and a summary extraction model, train the entity extraction model and the summary extraction model based on the constructed training set, evaluate the performance of the entity extraction model and the summary extraction model based on the constructed validation set, and test the training results of the entity extraction model and the summary extraction model based on the constructed test set to obtain the trained entity extraction model and the summary extraction model.

[0030] The key demand element extraction module is used to input the work order text data and telephone recording data of the key demand elements to be extracted into the trained entity extraction model and summary extraction model to extract the key demand elements and obtain the extraction results of the key demand elements.

[0031] Embodiment 3: S1. Obtain work order text data and telephone recording data, pre-process the obtained work order text data and telephone recording data, and remove invalid data and noise.

[0032] First, the work order text data and telephone recording data are obtained, and the obtained work order text data and telephone recording data are preprocessed to remove invalid data and noise.

[0033] Specifically, in this step, after clearing the telephone recordings with a duration of less than 1 minute and the data with no response content and no demand content in the consultation replies in the work order text data, an adaptive filtering algorithm is used to clear the noise in the obtained work order text data and telephone recording data.

[0034] S2. Perform word segmentation on the preprocessed work order text data and telephone recording data to obtain word segmentation results.

[0035] The jieba word segmentation tool is used to perform word segmentation on the preprocessed work order text data and telephone recording data to obtain the word segmentation results.

[0036] S3. Use the obtained word segmentation results to construct training sets, validation sets, and test sets.

[0037] The obtained word segmentation results are used to construct a training set, a validation set, and a test set. 60% of the obtained word segmentation results are used as a training set, 20% as a validation set, and 20% as a test set. As other implementation methods, 60%, 20%, and 20% in this step can be adjusted according to actual needs.

[0038] S4. Construct an entity extraction model and a summary extraction model, train the entity extraction model and the summary extraction model based on the constructed training set, evaluate the performance of the entity extraction model and the summary extraction model based on the constructed validation set, and test the training results of the entity extraction model and the summary extraction model based on the constructed test set to obtain the trained entity extraction model and the summary extraction model.

[0039] Construct an entity extraction model and a summary extraction model, train the entity extraction model and the summary extraction model based on the constructed training set, evaluate the performance of the entity extraction model and the summary extraction model based on the constructed validation set (the indicator for evaluating the performance of the entity extraction model and the summary extraction model is accuracy. When the accuracy reaches 100%, the entity extraction model and the summary extraction model have the best performance), and test the training results of the entity extraction model and the summary extraction model based on the constructed test set to obtain the trained entity extraction model and the summary extraction model.

[0040] S5. Input the work order text data and telephone recording data of the key demand elements to be extracted into the trained entity extraction model and summary extraction model to extract the key demand elements and obtain the extraction results of the key demand elements.

[0041] The work order text data and telephone recording data of the key demand elements to be extracted are input into the trained entity extraction model and summary extraction model to extract the key demand elements and obtain the extraction results of the key demand elements. The key demand elements include personnel, time, address and the name of the organization involved.

[0042] After obtaining the extraction results of the key demand elements, the method for extracting the key demand elements proposed by the present invention classifies the key demand elements to obtain several classification results.

[0043] Specifically, after obtaining the extraction results of the key demand elements, this embodiment uses data classification technology to classify the key demand elements to obtain several classification results.

[0044] The method for extracting key appeal elements proposed in the present invention can display and query several classification results after obtaining the several classification results.

[0045] Specifically, this embodiment displays and queries several classification results through a data display interface.

[0046] In order to verify the effectiveness of the method for extracting key elements of appeal proposed in the present invention, this embodiment inputs 350,000 historical work order text data and telephone recording data that have been closed in the work order system of a hotline center into the entity extraction model and summary extraction model for training, and obtains the trained entity extraction model and summary extraction model. The work order system application personnel of a hotline center assist the seat staff to automatically register the key elements of appeal using the trained entity extraction model and summary extraction model, and automatically extract the key information of personnel, time, address and the name of the organization involved in the work order content of the key elements to be extracted and fill them into the corresponding text box, with an accuracy rate of 82.69%, reducing the time for the seat staff to enter information when registering the work order.

[0047] Embodiment 4: See also Figure 3 As shown, the present invention also provides an electronic device 100 for a method of extracting key appeal elements; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0048] The memory 101 can be used to store the computer program 103, and the processor 102 implements the steps of the method for extracting key appeal elements described in Example 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 101 may include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0049] The at least one processor 102 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and uses various interfaces and lines to connect various parts of the entire electronic device 100.

[0050] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a method for extracting key appeal elements, and the processor 102 can execute the plurality of instructions to implement: Obtain work order text data and telephone recording data, pre-process the obtained work order text data and telephone recording data, and remove invalid data and noise; Perform word segmentation on the pre-processed work order text data and telephone recording data to obtain word segmentation results; Use the obtained word segmentation results to construct training sets, validation sets, and test sets; Constructing an entity extraction model and a summary extraction model, training the entity extraction model and the summary extraction model based on the constructed training set, evaluating the performance of the entity extraction model and the summary extraction model based on the constructed validation set, and testing the training results of the entity extraction model and the summary extraction model based on the constructed test set to obtain the trained entity extraction model and the summary extraction model; The work order text data and telephone recording data of the key demand elements to be extracted are input into the trained entity extraction model and summary extraction model to extract the key demand elements and obtain the extraction results of the key demand elements.

[0051] Embodiment 5: If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).

[0052] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0053] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0054] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0055] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for extracting key appeal elements, characterized in that: The following steps are involved: Obtain work order text data and telephone recording data, pre-process the obtained work order text data and telephone recording data, and remove invalid data and noise; Perform word segmentation on the pre-processed work order text data and telephone recording data to obtain word segmentation results; Use the obtained word segmentation results to construct training sets, validation sets, and test sets; Constructing an entity extraction model and a summary extraction model, training the entity extraction model and the summary extraction model based on the constructed training set, evaluating the performance of the entity extraction model and the summary extraction model based on the constructed validation set, and testing the training results of the entity extraction model and the summary extraction model based on the constructed test set to obtain the trained entity extraction model and the summary extraction model; The work order text data and telephone recording data of the key demand elements to be extracted are input into the trained entity extraction model and summary extraction model to extract the key demand elements and obtain the extraction results of the key demand elements.

2. The method for extracting key appeal elements according to claim 1, characterized in that: Adaptive filtering algorithm is used to remove noise from the work order text data and telephone recording data.

3. The method for extracting key appeal elements according to claim 1 is characterized in that: The jieba word segmentation tool is used to perform word segmentation on the preprocessed work order text data and telephone recording data to obtain the word segmentation results.

4. The method for extracting key appeal elements according to claim 1 is characterized in that: The indicator for evaluating the performance of entity extraction model and summary extraction model is accuracy.

5. The method for extracting key appeal elements according to claim 1, characterized in that: The key elements of the claim include personnel, time, address and the name of the organization involved.

6. The method for extracting key appeal elements according to claim 1, characterized in that: After obtaining the extraction results of the key demand elements, the key demand elements are classified to obtain several classification results.

7. The method for extracting key appeal elements according to claim 6, characterized in that: After obtaining several classification results, several classification results can also be displayed and queried.

8. A system for extracting key appeal elements, characterized in that: It includes data acquisition module, data preprocessing module, word segmentation processing module, data set construction module, model training module and key appeal element extraction module; The data acquisition module is used to acquire work order text data and telephone recording data; The data preprocessing module is used to preprocess the obtained work order text data and telephone recording data to remove invalid data and noise; The word segmentation processing module is used to perform word segmentation processing on the pre-processed work order text data and telephone recording data to obtain a word segmentation result; The data set construction module is used to construct a training set, a verification set and a test set using the obtained word segmentation results; The model training module is used to construct an entity extraction model and a summary extraction model, train the entity extraction model and the summary extraction model based on the constructed training set, evaluate the performance of the entity extraction model and the summary extraction model based on the constructed validation set, and test the training results of the entity extraction model and the summary extraction model based on the constructed test set to obtain the trained entity extraction model and the summary extraction model; The key demand element extraction module is used to input the work order text data and telephone recording data of the key demand elements to be extracted into the trained entity extraction model and summary extraction model to extract the key demand elements and obtain the extraction results of the key demand elements.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the processor implements the steps of the method for extracting key appeal elements described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for extracting key appeal elements described in any one of claims 1 to 7 are implemented.

Images