Information extraction method and system, electronic equipment and computer readable storage medium

Abstract

The invention relates to the technical field of natural language processing, and particularly discloses an information extraction method and system, electronic equipment and a computer readable storage medium, the method comprises the following steps: determining a topic type of to-be-extracted information, and preprocessing the to-be-extracted information to obtain the preprocessed to-be-extracted information; based on the topic type and the preprocessed to-be-extracted information, extracting a plurality of key information fragments, and integrating all the key information fragments to obtain integrated information; and when the integrated information is compliant, determining the integrated information as target information. The accuracy and efficiency of information extraction are improved, and the method can be applied to the field needing efficient and accurate information processing.

Description

Background Art

[0002] In recent years, with the rapid development of NLP (natural language processing) technology, NLP technology has more practical application scenarios, such as insurance sales, after-sales service, etc. In these application scenarios, the interaction between agents and customers is diverse and full of variables. The extraction of key information often requires going beyond the direct extraction of a single sentence, but rather requires inference and confirmation through in-depth understanding of the context, such as license plate number, ID number, name, address and other sensitive information. Due to the complex correlation between the various fragments of key information, it does not conform to the assumption of independent distribution of labels in traditional machine learning models, which makes it difficult for classic technologies such as NER (named entity recognition) to meet business needs. At the same time, the limitations of traditional regular expressions in intelligence and flexibility have limited their effectiveness in extracting key information, especially in terms of recall.

[0003] Therefore, it is urgent to provide a technical solution to solve the above problems. Summary of the invention

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

[0005] In a first aspect, the present invention provides an information extraction method, the technical solution of the method is as follows:

[0006] S1, determining the topic type of the information to be extracted, and preprocessing the information to be extracted to obtain the preprocessed information to be extracted;

[0007] S2, extracting multiple key information fragments based on the topic type and the preprocessed information to be extracted, and integrating all the key information fragments to obtain integrated information;

[0008] S31. When the integrated information complies with the regulations, the integrated information is determined as target information.

[0009] The beneficial effects of an information extraction method of the present invention are as follows:

[0010] The method of the present invention improves the accuracy and efficiency of information extraction and can be applied to fields requiring efficient and accurate information processing.

[0011] Based on the above solution, an information extraction method of the present invention can also be improved as follows.

[0012] In an optional manner, the step of preprocessing the information to be extracted to obtain the preprocessed information to be extracted includes:

[0013] When the information to be extracted is voice information, the voice transcription information of the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted;

[0014] When the information to be extracted is text information, the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted.

[0015] In an optional manner, the step of extracting a plurality of key information fragments based on the topic type and the preprocessed information to be extracted includes:

[0016] By utilizing the large model and combining the topic type, multiple key information fragments are extracted from the preprocessed information to be extracted.

[0017] In an optional manner, the step of integrating all key information fragments to obtain integrated information includes:

[0018] Determine the relationship between each two key pieces of information;

[0019] The integrated information is generated by utilizing the large model and combining all the key information fragments and the relationship between every two key information fragments.

[0020] In an optional manner, the method further includes:

[0021] S32: When the integrated information is not compliant, return to execute S1 or S2 until the integrated information is compliant, and then determine the integrated information as the target information.

[0022] In a second aspect, the present invention provides an information extraction system, the technical solution of which is as follows:

[0023] It includes: a processing module, an integration module and a first determination module;

[0024] The processing module is used to: determine the topic type of the information to be extracted, and preprocess the information to be extracted to obtain the preprocessed information to be extracted;

[0025] The integration module is used to: extract multiple key information fragments based on the topic type and the pre-processed information to be extracted, and integrate all the key information fragments to obtain integrated information;

[0026] The first determination module is used to: when the integrated information is compliant, determine the integrated information as target information.

[0027] The beneficial effects of an information extraction system of the present invention are as follows:

[0028] The system of the present invention improves the accuracy and efficiency of information extraction and can be applied to fields requiring efficient and accurate information processing.

[0029] Based on the above solution, an information extraction system of the present invention can also be improved as follows.

[0030] In an optional manner, the step of preprocessing the information to be extracted to obtain the preprocessed information to be extracted in the processing module includes:

[0031] When the information to be extracted is voice information, the voice transcription information of the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted;

[0032] When the information to be extracted is text information, the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted.

[0033] In an optional manner, the step of extracting a plurality of key information fragments based on the topic type and the preprocessed information to be extracted in the integration module includes:

[0034] By utilizing the large model and combining the topic type, multiple key information fragments are extracted from the preprocessed information to be extracted.

[0035] In a third aspect, a technical solution of an electronic device of the present invention is as follows:

[0036] The invention comprises a memory, a processor and a program stored in the memory and running on the processor. When the processor executes the program, the steps of the information extraction method of the present invention are realized.

[0037] In a fourth aspect, the present invention provides a computer-readable storage medium having the following technical solution:

[0038] The computer-readable storage medium stores instructions, and when the computer-readable storage medium reads the instructions, the computer-readable storage medium executes the steps of the information extraction method of the present invention.

[0039] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings are only used to illustrate the embodiments and are not to be considered as limiting the present invention. In addition, the same reference symbols are used to represent the same components throughout the accompanying drawings. In the accompanying drawings:

[0041] Figure 1 A schematic diagram of a flow chart of an embodiment of an information extraction method of the present invention;

[0042] Figure 2 A schematic diagram of the structure of an embodiment of an information extraction system of the present invention;

[0043] Figure 3 The figure is a schematic structural diagram of an embodiment of an electronic device of the present invention. DETAILED DESCRIPTION

[0044] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0045] Figure 1 FIG. 1 is a flow chart showing an embodiment of an information extraction method provided by the present invention, which is executed by an Agent. Figure 1 As shown, the following steps are included:

[0046] S1. Determine the topic type of the information to be extracted, and preprocess the information to be extracted to obtain the preprocessed information to be extracted.

[0047] The information to be extracted is: information that needs to be extracted for key information, which is voice information or text information provided by the user during the conversation. The topic type can be identified based on the large model (LLM model), for example, the topic type is: vehicle insurance declaration.

[0048] In S1, specifically:

[0049] The step of preprocessing the information to be extracted to obtain the preprocessed information to be extracted comprises:

[0050] When the information to be extracted is voice information, the voice transcription information of the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted.

[0051] The speech transcription information is the text information obtained by converting the speech.

[0052] When the information to be extracted is text information, the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted.

[0053] Among them, the correction method includes: removing noise and errors, etc. The simplification method includes: removing descriptive words and repeated redundancies, etc.

[0054] S2. Based on the topic type and the preprocessed information to be extracted, extract multiple key information fragments and integrate all the key information fragments to obtain integrated information.

[0055] In S2, specifically:

[0056] 1) The step of extracting multiple key information fragments based on the topic type and the preprocessed information to be extracted includes:

[0057] Using a large model and combining with the topic type, extract multiple key information fragments from the preprocessed information to be extracted.

[0058] Among them, the large model is: Large Language Model, specifically a deep learning model trained with a large amount of text data, capable of understanding and generating natural language. The key information fragments correspond to the topic type. For example, when the topic type is vehicle insurance declaration, the key information fragments include: license plate number, owner's name, owner's ID number, address, time, etc.

[0059] 2) The step of integrating all the key information fragments to obtain integrated information includes:

[0060] Determine the relationship between every two key information fragments.

[0061] Using the large model and combining with all the key information fragments and the relationship between every two key information fragments, generate the integrated information.

[0062] Among them, the large model can reason by deeply understanding the context to determine whether there is a relationship between every two key information fragments. For example, when the topic type is vehicle insurance declaration, there is a relationship between the key information fragment "owner's name" and the key information fragment "owner's ID number".

[0063] S31. When the integrated information is compliant, determine the integrated information as the target information.

[0064] Among them, it is determined whether the integrated information is compliant according to regulatory documents such as compliance documents and national standard documents. For example, when the integrated information is: license plate number "Zhe A*****" (each "*" represents a number or a letter), it is determined whether this license plate number is compliant according to the national standard document GA36-2018. When the integrated information is: license plate number "Zhe Z1234567", according to the national standard document GA36-2018, there are two errors in this license plate number, namely: ① There is no city code abbreviation in Zhejiang Province represented by the letter Z; ② A license plate number with 7 characters. At this time, it is determined that this integrated information is non-compliant.

[0065] In an optional manner, it further includes:

[0066] S32: When the integrated information is not compliant, return to execute S1 or S2 until the integrated information is compliant, and then determine the integrated information as the target information.

[0067] Specifically, when the integrated information is not compliant, the feedback audit information (determined to be non-compliant) is processed in combination with S1 or S2, and the cycle is repeated until the integrated information is compliant, and the integrated information is determined as the target information.

[0068] It should be noted that this embodiment can be processed by one or more Agents. This embodiment combines the large model LLM and the intelligent agent Agent, designs and constructs an information extraction method that is Agent-led, LLM-based, and integrates deep learning and machine learning. The system breaks through the difficulty of extracting key information in the customer service field (such as insurance sales, after-sales service, etc.), and improves the accuracy and efficiency of information extraction through in-depth understanding of the context to reason and confirm, and can be applied to fields that require efficient and accurate information processing.

[0069] Figure 2 FIG. 2 is a schematic diagram showing a structure of an embodiment of an information extraction system 200 provided by the present invention. Figure 2 As shown, the system 200 includes: a processing module 210, an integration module 220 and a first determination module 230;

[0070] The processing module 210 is used to: determine the topic type of the information to be extracted, and pre-process the information to be extracted to obtain the pre-processed information to be extracted;

[0071] The integration module 220 is used to: extract multiple key information fragments based on the topic type and the pre-processed information to be extracted, and integrate all the key information fragments to obtain integrated information;

[0072] The first determination module 230 is used to: when the integrated information is compliant, determine the integrated information as target information.

[0073] In an optional manner, the step of preprocessing the information to be extracted to obtain the preprocessed information to be extracted in the processing module 210 includes:

[0074] When the information to be extracted is voice information, the voice transcription information of the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted;

[0075] When the information to be extracted is text information, the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted.

[0076] In an optional manner, the step of extracting a plurality of key information segments based on the topic type and the preprocessed information to be extracted in the integration module 220 includes:

[0077] By utilizing the large model and combining the topic type, multiple key information fragments are extracted from the preprocessed information to be extracted.

[0078] In an optional manner, the step of integrating all key information fragments to obtain integrated information in the integration module 220 includes:

[0079] Determine the relationship between each two key pieces of information;

[0080] The integrated information is generated by utilizing the large model and combining all the key information fragments and the relationship between every two key information fragments.

[0081] In an optional manner, the method further includes: a second determination module;

[0082] The second determination module is used for: when the integration information is not compliant, returning to call the processing module or the integration module, until when the integration information is compliant, determining the integration information as the target information.

[0083] The technical solution of this embodiment improves the accuracy and efficiency of information extraction and can be applied to fields that require efficient and accurate information processing.

[0084] The above parameters and steps for each module to implement corresponding functions in the information extraction system 200 of this embodiment can refer to the parameters and steps in the above embodiment of the information extraction method, and will not be repeated here.

[0085] like Figure 3 As shown, an electronic device 300 according to an embodiment of the present invention includes a processor 320, the processor 320 is coupled to a memory 310, and the memory 310 stores at least one computer program 330, and the at least one computer program 330 is loaded and executed by the processor 320, so that the electronic device 300 implements any of the above information extraction methods, specifically:

[0086] The electronic device 300 may have relatively large differences due to different configurations or performances, and may include one or more processors 320 (Central Processing Units, CPU) and one or more memories 310, wherein the one or more memories 310 store at least one computer program 330, and the at least one computer program 330 is loaded and executed by the one or more processors 320, so that the electronic device 300 implements any information extraction method provided in the above embodiments. Of course, the electronic device 300 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output, and the electronic device 300 may also include other components for implementing device functions, which will not be repeated here.

[0087] A computer-readable storage medium according to an embodiment of the present invention stores at least one computer program, and the at least one computer program is loaded and executed by a processor so that a computer implements any of the above-mentioned information extraction methods.

[0088] Optionally, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0089] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device performs any of the above-mentioned information extraction methods.

[0090] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects and represent the definition of a specific order or sequence. The order of use of similar objects can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than the order shown or described.

[0091] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0092] Any combination of one or more computer-readable media can be used. Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0093] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. An information extraction method, characterized in that: include: S1, determining the topic type of the information to be extracted, and preprocessing the information to be extracted to obtain the preprocessed information to be extracted; S2, extracting multiple key information fragments based on the topic type and the preprocessed information to be extracted, and integrating all the key information fragments to obtain integrated information; S31. When the integrated information complies with the regulations, the integrated information is determined as target information.

2. The information extraction method according to claim 1, characterized in that: The step of preprocessing the information to be extracted to obtain the preprocessed information to be extracted comprises: When the information to be extracted is voice information, the voice transcription information of the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted; When the information to be extracted is text information, the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted.

3. The information extraction method according to claim 1, characterized in that: The step of extracting a plurality of key information fragments based on the topic type and the preprocessed information to be extracted includes: By utilizing the large model and combining the topic type, multiple key information fragments are extracted from the preprocessed information to be extracted.

4. The information extraction method according to claim 3, characterized in that: The steps to integrate all the key information fragments and obtain the integrated information include: Determine the relationship between each two key pieces of information; The integrated information is generated by utilizing the large model and combining all the key information fragments and the relationship between every two key information fragments.

5. The information extraction method according to any one of claims 1 to 4, characterized in that: Also includes: S32: When the integrated information is not compliant, return to execute S1 or S2 until the integrated information is compliant, and then determine the integrated information as the target information.

6. An information extraction system, characterized in that: include: A processing module, an integration module and a first determination module; The processing module is used to: determine the topic type of the information to be extracted, and preprocess the information to be extracted to obtain the preprocessed information to be extracted; The integration module is used to: extract multiple key information fragments based on the topic type and the pre-processed information to be extracted, and integrate all the key information fragments to obtain integrated information; The first determination module is used to: when the integrated information is compliant, determine the integrated information as target information.

7. The information extraction system according to claim 6, characterized in that: The step of preprocessing the information to be extracted to obtain the preprocessed information to be extracted in the processing module includes: When the information to be extracted is voice information, the voice transcription information of the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted; When the information to be extracted is text information, the information to be extracted is corrected and simplified to obtain the pre-processed information to be extracted.

8. The information extraction system according to claim 6, characterized in that: The step of extracting a plurality of key information fragments based on the topic type and the preprocessed information to be extracted in the integration module includes: By utilizing the large model and combining the topic type, multiple key information fragments are extracted from the preprocessed information to be extracted.

9. An electronic device, characterized in that: The electronic device includes a processor, the processor is coupled to a memory, the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor so that the electronic device implements the information extraction method as described in any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one computer program, and the at least one computer program is loaded and executed by a processor so that the computer-readable storage medium implements the information extraction method according to any one of claims 1 to 5.

Images