Document information integration method, device, equipment, medium and program product

Abstract

The embodiments of the present invention relate to the field of computer technology, and disclose a document information integration method, device, system, equipment, medium and program product, wherein the method comprises: obtaining a list of objects to be evaluated, obtaining corresponding unstructured information source documents according to the list of objects to be evaluated, matching content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, extracting target key information from the unstructured information source document based on the content extraction rule, and then obtaining a target structured document according to a preset information integration template and the target key information integration, thereby realizing information extraction and information integration of multi-source unstructured documents, without the need to separately configure an integration template for each type of source document, solving the technical problem of relying on complex template configuration for information extraction in the prior art, and improving the efficiency of information extraction and integration of multi-source unstructured documents.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of computer technology, and in particular to a document information integration method, device, equipment, medium and program product. Background Art

[0002] In some project evaluation or credit assessment scenarios in the financial field, a large amount of relevant information needs to be analyzed and evaluated. However, the document sources of the information to be evaluated are large, and the number of documents is also large, so it is necessary to extract and integrate the effective information in each document in advance.

[0003] Currently, templates are mostly written based on template engines (Freemarker) for each type of document, and then the document content is extracted according to the document template structure. However, the sources of various documents are diverse and the formats are diverse. Template configuration is difficult, the exported content takes up a lot of space, and the exported content is a non-standard document, which is difficult to dynamically reprocess, resulting in low efficiency of information integration. Summary of the invention

[0004] The embodiments of the present invention provide a document information integration method, apparatus, device, medium and program product to improve the efficiency of multi-source non-structured document information extraction and integration.

[0005] In a first aspect, an embodiment of the present invention provides a document information integration method, the method comprising:

[0006] Obtaining a list of objects to be evaluated, and obtaining corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents;

[0007] For each object to be evaluated in the list of objects to be evaluated, matching the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, and extracting the target key information according to the content extraction rules;

[0008] According to the preset information integration template, the target key information is integrated to generate a target structured document.

[0009] Optionally, matching the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, and extracting the target key information according to the content extraction rules, includes:

[0010] Determine the document type corresponding to the information source document according to the document number, and match the corresponding content extraction rule according to the document type;

[0011] Target key information is extracted from information source documents of corresponding types based on preset keyword information in the content extraction rule.

[0012] Optionally, extracting target key information from information source documents of corresponding types based on preset keyword information in the content extraction rule includes:

[0013] Acquire a first preset keyword phrase in the preset keyword information;

[0014] According to the first start keyword and the first end keyword in the first preset keyword phrase, the first paragraph starting at the first start keyword and ending at the first end keyword is matched and extracted.

[0015] Optionally, the extracting target key information from the corresponding type of information source document based on the preset keyword information in the content extraction rule further includes:

[0016] Acquire a second preset keyword phrase in the preset keyword information;

[0017] According to the second start keyword and the second end keyword in the second preset keyword phrase, matching a second paragraph starting with the second start keyword and ending with the second end keyword;

[0018] In the second paragraph, sentences containing first preset keywords are extracted.

[0019] Optionally, the extracting target key information from the corresponding type of information source document based on the preset keyword information in the content extraction rule further includes:

[0020] Extract the table or cell content in the table containing the second preset keyword.

[0021] Optionally, obtaining a list of objects to be evaluated includes:

[0022] Establish a connection with the target risk early warning system, request and obtain the business risk alarm list;

[0023] The business risk warning list is used as the list of objects to be evaluated.

[0024] Optionally, obtaining corresponding information source documents according to the list of objects to be evaluated includes:

[0025] Request and obtain the credit documents of the subject to be assessed from the preset credit business system;

[0026] The credit granting document includes at least one information source document selected from the group consisting of an investigation report, a preset project evaluation and rating report, a preset business application form, a request document, and a loan amount application form.

[0027] Optionally, integrating the target key information according to a preset information integration template to generate a target structured document includes:

[0028] Matching the target key information content with the document content hierarchy of the preset information integration template;

[0029] The target key information is integrated according to the matching results of the document content hierarchy to generate a target structured document.

[0030] Optionally, the method further includes:

[0031] The modification made by the user to the target key information extraction item in the content extraction rule is obtained, and the target key information is extracted according to the modified content extraction rule.

[0032] Optionally, the method further includes:

[0033] The modification of the document content hierarchy structure in the preset information integration template by the user is obtained, and the target key information is integrated according to the modified document content hierarchy structure.

[0034] Optionally, the method further includes:

[0035] When the user downloads the target structured document, the user is presented with information source document download prompt information and a download link;

[0036] In response to a user triggering operation on the download link, the information source document associated with the target structured document is downloaded.

[0037] In a second aspect, an embodiment of the present invention provides a document information integration device, the device comprising:

[0038] A source document acquisition module, used to acquire a list of objects to be evaluated, and acquire corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents;

[0039] An information extraction module is used to match the content extraction rules corresponding to each document in the list of objects to be evaluated according to the document number of each document in the information source document corresponding to the object to be evaluated, and extract the target key information according to the content extraction rules;

[0040] The information integration module is used to integrate the target key information according to a preset information integration template to generate a target structured document.

[0041] In a third aspect, an embodiment of the present invention further provides a computer device, the computer device comprising:

[0042] one or more processors;

[0043] A memory for storing one or more programs;

[0044] When the one or more programs are executed by the one or more processors, the one or more processors implement the document information integration method provided by any embodiment of the present invention.

[0045] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a document information integration method as provided in any embodiment of the present invention.

[0046] In a fifth aspect, an embodiment of the present invention further provides a computer program product on which a computer program is stored, and when the computer program is executed by a processor, the document information integration method provided by any embodiment of the present invention is implemented.

[0047] The embodiments of the above invention have the following advantages or beneficial effects:

[0048] In the embodiment of the present invention, by obtaining a list of objects to be evaluated, obtaining corresponding unstructured information source documents according to the list of objects to be evaluated, for each object to be evaluated in the objects to be evaluated, according to the document number of each document in the information source document corresponding to the object to be evaluated, matching the content extraction rules corresponding to each document, extracting target key information from the unstructured information source document based on the content extraction rule, and then integrating the target key information extracted according to the preset information integration template and the extracted target structured document, realizing information extraction and information integration of multi-source unstructured documents, without the need to configure the integration template separately for each type of source document, solving the technical problem of relying on complex template configuration for information extraction in the prior art, and improving the efficiency of information extraction and integration of multi-source unstructured documents. In addition, the method can generate structured documents according to the extracted information, solving the problem that the unstructured documents generated after the information is extracted in the prior art are difficult to display flexibly, and the generated structured documents are helpful for analyzing the extracted information, and improving the efficiency of users in obtaining information. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a flow chart of a document information integration method provided by Embodiment 1 of the present invention;

[0050] Figure 2 is a flow chart of a document information integration method provided by Embodiment 2 of the present invention;

[0051] Figure 3A is a flow chart of a document information integration method provided by Embodiment 3 of the present invention;

[0052] Figure 3BIt is a schematic diagram of the interaction between a document information integration system, a target risk early warning system and a preset credit business system provided in the third embodiment of the present invention;

[0053] Figure 3C This is an information extraction process of a document information integration system provided by Embodiment 3 of the present invention;

[0054] Figure 4 is a flow chart of a document information integration method provided by real-time example 4 of the present invention;

[0055] Figure 5 It is a structural schematic diagram of a document information integration device provided by the fifth real-time example of the present invention;

[0056] Figure 6 It is a structural diagram of a computer device provided in Embodiment 6 of the present invention. DETAILED DESCRIPTION

[0057] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0058] Embodiment 1

[0059] Figure 1 The flowchart of a document information integration method provided in the first embodiment of the present invention is applicable to the case where target key information is extracted from multiple unstructured information source documents, and then a target structured document is generated according to the extracted target key information. For example, in the credit assessment scenario of the financial industry, target key information can be extracted from unstructured documents such as investigation reports, preset project assessment rating reports, preset business declarations, request documents, and loan amount declarations in customer credit files to generate a structured customer risk verification report; or, in the physical examination scenario of the medical industry, target key information can be extracted from unstructured documents such as basic information documents, imaging report documents, inspection project documents, and inspection conclusion documents in customer physical examination files to generate a structured physical examination report; or, in the scenario of semester grade evaluation in the education industry, target key information can be extracted from basic information documents, grade record documents, grade analysis documents, and teacher evaluation documents in student files to generate a structured grade evaluation report. The above scenarios are only examples, and the method provided in this embodiment is not limited to the above scenarios. The method can be executed by a document information integration device, which can be implemented by software and / or hardware and integrated in a computer device with application development functions.

[0060] like Figure 1As shown, the document information integration method includes the following modules:

[0061] S110: Obtain a list of objects to be evaluated, and obtain corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents.

[0062] The list of objects to be evaluated may include at least one object to be evaluated, and the object to be evaluated may be an object for which a target structured document needs to be generated according to its corresponding information source document. In this embodiment, the list of objects to be evaluated may be sent by a third-party system, or may be determined according to the current selection operation of the user.

[0063] For example, in the credit assessment scenario of the financial industry, the target risk warning system (third-party system) can automatically identify customers who meet the preset warning conditions as objects to be evaluated, and periodically send a list of objects to be evaluated, where the preset warning conditions can be business conditions that require special attention, such as capital shortage, project loss, lack of guarantee, or dishonest behavior in the industry; or, the credit business system can identify new customers added within a preset period as objects to be evaluated, and periodically send a list of objects to be evaluated. For another example, in the physical examination scenario of the medical industry, the hospital information system (third-party system) can identify a certain enterprise's customers as objects to be evaluated, or identify customers whose preset inspection items do not meet the standard reference values ​​as objects to be evaluated, or identify customers whose inspection time is within the preset time window as objects to be evaluated, etc.

[0064] In this embodiment, after obtaining the list of objects to be evaluated, the information source documents corresponding to the list of objects to be evaluated are further obtained, that is, the information source documents of each object in the list of objects to be evaluated. Among them, the information source document can specifically be a source document corresponding to the object to be evaluated that needs to be extracted. In this embodiment, the information source document is an unstructured document, such as a word document, a pdf document, a txt document, etc. An information source document can be composed of at least one of text information, table information and image information. Through the method of this embodiment, the target key information can be extracted from the text information, table information or image information of the information source document.

[0065] Exemplarily, taking the credit assessment scenario in the financial industry as an example, the information source document may include at least one of an investigation report, a preset project assessment and rating report, a preset business application, a request document, and a loan amount application. Among them, the investigation report may include at least one of a corporate new credit customer investigation report, a corporate existing customer investigation report, and a business new credit and existing investigation report. The preset project assessment and rating report may include at least one of a real estate development loan assessment and rating report and a fixed asset loan assessment and rating report. The request document may include the operating and management opinions of the reporting unit. The loan amount application may include at least one of a corporate and business new credit customer application and a corporate and business existing customer application.

[0066] S120: For each object to be evaluated in the list of objects to be evaluated, match the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, and extract the target key information according to the content extraction rules.

[0067] Specifically, after obtaining the information source documents corresponding to each object to be evaluated in the list of objects to be evaluated, a content extraction rule corresponding to the information source document of each object to be evaluated may be determined for each object to be evaluated, so as to extract target key information from each information source document according to the content extraction rule.

[0068] In this embodiment, the content extraction rules corresponding to each document can be matched according to the document number of each document in the information source document, wherein the document number can be a number representing the document type or document subtype, that is, documents of the same type can have the same document number.

[0069] Specifically, a corresponding document number can be prepared in advance for each document type or document subtype, and then the content extraction rules corresponding to each document type or document subtype can be associated and stored with the document number corresponding to the document type or document subtype to generate a directory so that the corresponding content extraction rules can be queried in the directory based on the document numbers of each document in each information source document obtained.

[0070] In this embodiment, taking the credit assessment scenario in the financial industry as an example, since the information source documents can be divided into types such as investigation reports, preset project evaluation and rating reports, preset business applications, request documents and loan amount applications, the investigation reports can be divided into sub-types such as corporate new credit customer investigation reports, corporate existing customer investigation reports and business new credit and existing customer investigation reports. The content extraction rules corresponding to each sub-type may be different. Therefore, preferably, the document number can use the number that represents the document sub-type to accurately obtain the content extraction rules that conform to the document sub-type.

[0071] Furthermore, after obtaining the content extraction rules corresponding to each document, the target key information can be extracted from the document according to the content extraction rules. The content extraction rules can be keyword extraction rules, that is, the target key information is selected by keywords. The content extraction rules can also be focus content extraction rules, that is, determining the document focus content in the document, and extracting the document focus content as the target key information; for example, the document is divided into paragraphs, the frequency of occurrence of each phrase in each paragraph is counted, and the paragraph where the phrase with the highest frequency of occurrence is located is determined as the document focus content.

[0072] Following the example of the credit assessment scenario in the financial industry mentioned above, for the investigation report, the extracted target key information may include at least one of the main product situation, regional advantages, production capacity, the top five accounts with the largest amount of funds, the details of the receivables and payables of the top five customers, the key information table of banks and enterprises, the summary table of various business income, and the related guarantee situation. For the preset project evaluation and rating report, the extracted target key information may include at least one of the project's geographical location, construction content, construction scale, investment and funding source, development period, construction progress, development cost, project income estimation, break-even analysis, project sensitivity analysis, supporting conditions around the project, real estate sales in the project area, project net present value, and project payment. For the preset business declaration, the extracted target key information may include at least one of the purpose of the credit business, the operating and management opinions of the reporting unit, the guarantee situation (including the guarantee method, risk mitigation measures, collateral assessment value and mortgage rate), and the main repayment source. For the request document, the extracted target key information may include the operating and management opinions of the reporting unit. For the loan quota application form, the extracted target key information may include the purpose of the credit business, the type and allocation of the credit business quota, the changes in the credit of existing customers compared with the previous time, the operating and management opinions of the reporting unit, the guarantee situation (including the guarantee method, risk mitigation measures, the assessed value of the collateral, the mortgage rate and the changes in the guarantee situation of existing customers compared with the previous time) and at least one of the main repayment sources.

[0073] In this embodiment, the extracted target key information is structured data, which may include business number, title (can be divided into multiple levels), document block logical order (extract record item sorting code, sorted in ascending order), structured data type (text: paragraph, table: table), structured data body (JSON format), table size (row / column), structured tag, supplementary description and data date. Among them, a table item can be a JSON array, and each JSON object contains the following attributes:

[0074] Type: object type (t_head header comment / tex text / table table); Data: specific content, a) for table header: comment content, b) for text: text content, c) for table: json array, each array object is a cell, containing properties, including, Col: current column, Colspan: number of column merges, Row current row, number of row merges, Text cell content; Hit: hit keyword list; Doc: source document.

[0075] S130: Integrate the target key information according to a preset information integration template to generate a target structured document.

[0076] Specifically, after obtaining the target key information, the target key information can be integrated into a target structured document according to a preset information integration template. The preset information integration template may include a document content hierarchy, that is, a hierarchy of each target key information in the target structured document, which may include arrangement order information or preset position information.

[0077] Of course, the preset information integration template can also include text attribute information and cell format information, wherein the text attribute information includes but is not limited to text size, text color, font type, whether bold and text background, and the cell format information includes but is not limited to cell size, the number of text rows contained in the cell, the combination relationship between cells and cell alignment.

[0078] In this embodiment, the target key information extracted from various types of information source documents can be integrated through a unified preset information integration template, without the need to formulate a corresponding integration template for each type of information source document. Specifically, the target key information can be arranged in sequence according to the document content hierarchy in the preset information integration template, or the target key information can be placed in a preset position to obtain a target structured document.

[0079] In an optional implementation, the target key information is integrated according to a preset information integration template to generate a target structured document, which may be: matching the target key information content with the document content hierarchy of the preset information integration template; integrating the target key information according to the matching result of the document content hierarchy to generate a target structured document.

[0080] Among them, the document content hierarchical structure may include hierarchical structure tags of titles at various levels, which are used to organize the corresponding structured target key information with the object to be evaluated as the dimension. In this optional implementation, the titles at various levels in the target key information can be matched with the document content hierarchical structure to determine the positions of the titles at various levels in the target key information content, and then according to the positions of the titles at various levels, the information corresponding to the titles at various levels in the target key information content is filled into the preset information integration template to generate the target structured document. Through this optional implementation, the integrated target structured document can have a document content hierarchical structure, which is convenient for users to quickly obtain information.

[0081] After generating the target structured document, the method provided in this embodiment may further include: when the user downloads the target structured document, displaying information source document download prompt information and a download link to the user; in response to the user's triggering operation on the download link, downloading the information source document associated with the target structured document.

[0082] That is, this embodiment can also download the information source document used to generate the target structured document while downloading the target structured document, so that the user can view and verify the content of the target structured document based on the downloaded target structured document and information source document.

[0083] The technical solution of this embodiment is to obtain a list of objects to be evaluated, obtain the corresponding unstructured information source documents according to the list of objects to be evaluated, and for each object to be evaluated, match the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, extract the target key information from the unstructured information source document based on the content extraction rule, and then integrate the target key information extracted according to the preset information integration template and the extracted target structured document, so as to realize the information extraction and information integration of multi-source unstructured documents, without the need to configure the integration template separately for each type of source document, solve the technical problem of relying on complex template configuration for information extraction in the prior art, and improve the efficiency of information extraction and integration of multi-source unstructured documents. In addition, the method can generate structured documents according to the extracted information, solve the problem that the unstructured documents generated after the information is extracted in the prior art are difficult to display flexibly, and the generated structured documents are helpful for analyzing the extracted information, and improve the efficiency of users in obtaining information.

[0084] Embodiment 2

[0085] Figure 2 This is a flowchart of a document information integration method provided by Embodiment 2 of the present invention. This embodiment is optimized based on the above embodiments and further describes the specific process of extracting target key information. Figure 2The document information integration method provided in this embodiment includes the following steps:

[0086] S210: Obtain a list of objects to be evaluated, and obtain corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents.

[0087] S220. For each object to be evaluated in the list of objects to be evaluated, determine the document type corresponding to the information source document according to the document number, match the corresponding content extraction rule according to the document type, and extract target key information from the information source document of the corresponding type based on preset keyword information in the content extraction rule.

[0088] In this embodiment, each document type and the document number corresponding to each document type can be stored in association in advance, and each document type and the content extraction rule corresponding to each document type can be stored in association, and then the corresponding document type can be queried by the document number of the information source document, and then the corresponding content extraction rule can be queried by the document type. It should be noted that the number of content extraction rules corresponding to a document type can be one or more.

[0089] The content extraction rule in this embodiment includes preset keyword information, that is, the content extraction rule is keyword extraction. Specifically, the preset keyword information can be matched with the text content in the information source document to determine the target key information. Optionally, before matching the preset keyword information with the text content in the information source document, the text content in the information source document can also be identified by a text recognition method, or the information source document can be formatted to obtain the text content in the information source document. For example, a word document is converted into an HTML document.

[0090] Exemplarily, the content extraction rules may include structured marks, titles at various levels, serial numbers corresponding to the titles at various levels, preset keyword information, document subtypes, extraction methods, and positioning rules. For example, using the credit assessment scenario in the financial industry, if there are four levels of titles, the content extraction rules may be: 0, 1, basic customer information, 7, main product information, 1,, 1,, main products | main products | business scope, 3910001, 4, w, w.

[0091] Among them, the structured flag is used to identify whether the extraction object targeted by the content extraction rule is structured data or unstructured data. For example, the value 1 represents the extraction of structured data. Each level of title corresponds to the corresponding level of title in the generated target structured document. Following the example of the credit assessment scenario in the financial industry mentioned above, the first-level title can include four types of titles. From the order of titles in the generated target structured document (customer risk verification report), the first-level titles can be customer basic information, current credit business information, project evaluation and rating status (real estate) and project evaluation and rating status (fixed loan projects). For the level of customer basic information, the second-level titles under its category may include customer name, registered capital, the group, establishment date, main business, main product status, regional advantages, production capacity, the top five accounts of capital transactions, and details of accounts receivable and payable of the top five customers.

[0092] The title number is used to indicate the order between titles of the same level. Each level of title has its own corresponding title number, and the title number of each level starts from 1. The division of titles at each level can be determined according to the actual application scenario. The preset keyword information can be a single keyword or a group of keywords, which is used to match the corresponding text content in unstructured information source documents. The combined keywords can be separated by "|" to indicate an "or" relationship. For example, "Main products|Main products|Business scope" means that if the text matches "Main products" or "Main products" or "Business scope", it means that the text is the text that needs to be extracted by the content extraction rule.

[0093] In this embodiment, different unstructured information source documents require different contents to be extracted. Even if they have the same title, the content extraction rules applied to different information source documents may be different. Based on this, document subtypes are used in content extraction rules to associate rules with documents, so that when extracting unstructured information source documents, all content extraction rules related to the document can be found according to the document subtype corresponding to the document.

[0094] The extraction method defines the type of target key information extracted by the content extraction rule, including but not limited to extracting tables without headers, extracting paragraphs, extracting paragraphs with keywords, extracting sentences with keywords, extracting tables with headers, headers on titles, extracting tables with headers, headers on the previous row of the table, mixed paragraph extraction, and extracting cell data in the table. The positioning rule describes the location of the target key information that the content extraction rule needs to extract in the entire information source document. Different extraction methods have different positioning rules.

[0095] In an optional embodiment, the method of extracting target key information from information source documents of corresponding types based on preset keyword information in the content extraction rule may be: obtaining a first preset keyword phrase in the preset keyword information; and matching and extracting a first paragraph starting with the first start keyword and a first end keyword according to a first start keyword and a first end keyword in the first preset keyword phrase. The first paragraph starts at the first start keyword and ends at the first end keyword.

[0096] The first preset keyword phrase may include a first start keyword and a first end keyword. Specifically, the start position and the end position may be determined in the information source document by the first start keyword and the first end keyword, and the paragraph between the start position and the end position may be extracted.

[0097] For example, for the information source document of the request document, since the information extracted from the request document is the operating and management opinions of the reporting unit, the operating and management opinions of the reporting unit are usually extracted in the form of paragraphs. Therefore, the two titles before and after the paragraph to be extracted can be set as the first starting keyword and the first ending keyword respectively, and the above method can be used to locate the paragraph where the operating and management opinions of the reporting unit are located, and the operating and management opinions of the reporting unit are extracted.

[0098] Through this optional implementation, the extraction of paragraph content based on the first start keyword and the first end keyword can be realized, which is convenient for extracting text content in the form of paragraphs in the information source document. Optionally, when the first start keyword and the first end keyword are preset tags, it can be expressed as a full-text search, that is, extracting the full-text content; when the first start keyword and the first end keyword are titles, all paragraph content between the two titles can be extracted.

[0099] In another optional embodiment, the method of extracting target key information from information source documents of corresponding types based on preset keyword information in the content extraction rules also includes: obtaining a second preset keyword phrase in the preset keyword information; matching a second paragraph starting with the second start keyword and a second end keyword according to the second preset keyword phrase; and extracting a sentence containing the first preset keyword in the second paragraph.

[0100] For example, the titles in customer investigation and evaluation reports, credit limit applications, and business applications are relatively fixed. You can set the title as the second starting keyword and the second ending keyword to locate the paragraph where the content to be extracted is located, and set the fixed phrase in the sentence where the information to be extracted is located as the first preset keyword to filter out the sentences to be extracted from the paragraph.

[0101] In this optional embodiment, the preset keyword information includes a first preset keyword and a second preset keyword phrase, the second preset keyword phrase includes a second start keyword and a second end keyword, the second preset keyword phrase is used to locate the second paragraph in the information source document, and the first preset keyword is used to extract a sentence containing the first preset keyword in the located second paragraph.

[0102] In this way, the target key information can be accurately extracted from the information source document, avoiding the extraction of redundant information by directly using the first preset keyword for full-text matching when there are multiple sentences containing the first preset keyword in the information source document, thereby improving the extraction efficiency and accuracy of the target key information. Of course, in this optional implementation, the target sub-paragraph containing the first preset keyword can also be extracted from the second paragraph according to the first preset keyword.

[0103] In another optional implementation, the extracting target key information from the corresponding type of information source document based on the preset keyword information in the content extraction rule may also be: extracting a table containing a second preset keyword or cell content in the table.

[0104] Among them, the second preset keyword can be used to match the corresponding title in the information source document. Specifically, the title of the table to be extracted can be matched in the information source document according to the second preset keyword, and then the table can be extracted, or all the cell contents in the table can be extracted. Through this optional implementation, the extraction of the table content can be achieved. Of course, the header of the table can also be extracted while extracting the table, and the header is in the previous line of the table.

[0105] It is understandable that, in another embodiment, a table containing a second preset keyword can be determined in the information source document based on the first preset keyword, and then the cell content can be extracted from the table based on the second preset keyword. In this way, accurate extraction of some cell content in the table can be achieved.

[0106] S230: Integrate the target key information according to a preset information integration template to generate a target structured document.

[0107] The technical solution of this embodiment, after obtaining the information source documents corresponding to the list of objects to be evaluated, determines the document type corresponding to the information source document according to the document number, and matches the corresponding content extraction rules according to the document type, and then extracts the target key information from the information source documents of the corresponding type based on the preset keyword information in the content extraction rules, thereby realizing keyword-based information extraction and improving the efficiency of multi-source unstructured document information extraction.

[0108] Embodiment 3

[0109] Figure 3A This is a flowchart of a document information integration method provided by Embodiment 3 of the present invention. This embodiment can be combined with the solutions of the above embodiments to further describe the process of document information integration from the perspective of interaction between various systems. Figure 3A The document information integration method provided in this embodiment includes the following steps:

[0110] S310: Establish a connection with the target risk warning system, request and obtain a business risk warning list, and use the business risk warning list as the list of objects to be evaluated.

[0111] Among them, the target risk warning system can be a system that screens various objects according to risk warning rules in the credit assessment scenario of the financial industry. The risk warning rules can be that the project loss ratio exceeds the set ratio, or the capital shortage amount exceeds the set amount, or there is a lack of guarantee information, etc.

[0112] Specifically, the target risk warning system can evaluate each object based on the information source documents of each object obtained by the credit business system to determine the business risk alarm list. Among them, the credit business system can be a system that obtains the information source documents of each object (authorized credit file related documents).

[0113] S320. Obtain corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents.

[0114] Optionally, obtaining corresponding information source documents based on the list of objects to be evaluated includes: requesting and obtaining credit documents of the objects to be evaluated from a preset credit business system; wherein, the credit documents include at least one information source document among an investigation report, a preset project evaluation and rating report, a preset business declaration, a request document and a loan amount declaration.

[0115] That is, in the credit assessment scenario of the financial industry, the information source document may include at least one of an investigation report, a preset project assessment and rating report, a preset business application, a request document, and a loan amount application. Among them, the investigation report may include at least one of a corporate new credit customer investigation report, a corporate existing customer investigation report, and a business new credit and existing investigation report. The preset project assessment and rating report may include at least one of a real estate development loan assessment and rating report and a fixed asset loan assessment and rating report. The request document may include the operating and management opinions of the reporting unit. The loan amount application may include at least one of a corporate and business new credit customer application and a corporate and business existing customer application.

[0116] For example, Figure 3BAs shown, a schematic diagram of the interaction between a document information integration system, a target risk warning system and a preset credit business system is shown. Among them, the document information integration system is used to extract and integrate information from multi-source unstructured documents based on the document information integration method provided by each embodiment of the present invention to generate structured documents. The document information integration system can establish connections with the target risk warning system and the preset credit business system respectively. The document information integration system includes an outbound call component, which packages the credit file-related documents in the preset credit business system and transmits them end-to-end by configuring a scheduled task. At the same time, the document information integration system also supports batch outbound calls and self-increment outbound calls for customer files according to time intervals, so as to achieve the purpose of efficient, accurate and automatic configuration and collection of system data. The document information integration system can also obtain the business risk alarm list determined by the target risk warning system on a regular basis (e.g., by day), and obtain the full compressed package of the credit file-related documents of each customer in the business risk alarm list from the preset credit business system through the outbound call component.

[0117] For example, Figure 3C As shown, a document information integration system information extraction process is demonstrated. The system pre-configures document extraction rules for the chapter (paragraph) content and chart content that need to be extracted, starts a scheduled task in the job scheduling system, and regularly monitors the transmission results of the upstream outbound call component. The full compressed package of the system alarm and manually imported credit files is decompressed on a daily basis, and the decompressed documents are extracted and stored in the database. It supports the screening of information source documents of specified types including customer investigation reports, project evaluation and rating reports, business declarations, and quota declarations from the decompressed files, and through content extraction rules, the content that the business needs to focus on is parsed into structured data and stored in the database for subsequent data display and export of a unified customer risk verification report. Of course, the document information integration system can also calculate corresponding indicators that meet business needs based on the analysis of the extracted structured data.

[0118] S330: For each object to be evaluated in the list of objects to be evaluated, match the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, and extract the target key information according to the content extraction rules.

[0119] S340: Integrate the target key information according to a preset information integration template to generate a target structured document.

[0120] Specifically, after obtaining the target key information, the document information integration system can integrate the target key information through the APACHE POI package to generate a customer risk verification report, i.e., a target structured document. In this embodiment, the customer risk verification report can include basic customer information, financial information, single credit business information, project evaluation rating, branch feedback, and other contents.

[0121] The technical solution of this embodiment establishes a connection with the target risk warning system, requests and obtains the business risk alarm list, and then uses the business risk alarm list as the list of objects to be evaluated, so as to generate a structured document according to the information source document corresponding to the list of objects to be evaluated, thereby realizing the information extraction and integration of the bank authorization document, and realizing the extraction and integration of relevant information for the alarm objects.

[0122] Embodiment 4

[0123] Figure 4 This is a flowchart of a document information integration method provided by Embodiment 3 of the present invention. This embodiment is further optimized on the basis of the above embodiments, and illustrates the process of obtaining the user's operation instruction, and completing the information integration according to the user's customized settings of the content extraction rules and the preset information integration template, and according to the user's personalized settings. Figure 4 The document information integration method provided in this embodiment includes the following steps:

[0124] S410: Obtain a list of objects to be evaluated, and obtain corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents.

[0125] S420: For each object to be evaluated in the list of objects to be evaluated, match the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, and extract the target key information according to the content extraction rules.

[0126] S430: Integrate the target key information according to a preset information integration template to generate a target structured document.

[0127] S440: Obtain the modification made by the user to the target key information extraction item in the content extraction rule, and extract the target key information according to the modified content extraction rule.

[0128] In this embodiment, the user can modify the target key information extraction items in the content extraction rules (including adding extraction items, deleting extraction items or updating extraction items), extract information according to the content extraction rules modified by the user, and obtain a structured document that meets the user's personalized needs.

[0129] Specifically, this embodiment can display each extraction item in the content extraction rule on the first preset interface, and display modification, addition and deletion controls, and monitor the control trigger events on the first preset interface, and determine the user's modification of the target key information extraction item based on the monitored control trigger events.

[0130] Of course, the user can also adjust the preset information integration template. That is, optionally, the method can also include: obtaining the user's modification of the document content hierarchy in the preset information integration template, and integrating the target key information according to the modified document content hierarchy.

[0131] The user can modify the document content hierarchy in the preset information integration template, such as modifying the title level, deleting a certain level of title, etc. According to the preset information integration template modified by the user, the extracted target key information is integrated to obtain a structured document that meets the user's personalized needs.

[0132] Specifically, this embodiment can display the preset information integration template on the second preset interface, and display the adjustment controls for the document content hierarchy structure in the preset information integration template (such as moving, adding, deleting, etc.), and listen to the control trigger events on the second preset interface, and determine the user's modifications to the preset information integration template based on the monitored control trigger events.

[0133] The technical solution of this embodiment obtains the user's modifications to the target key information extraction items in the content extraction rules, and extracts the target key information according to the modified content extraction rules, so that the user can personalize the focus content in the structured document according to needs, thereby improving the user's usage experience.

[0134] Embodiment 5

[0135] Figure 5 This is a structural diagram of a document information integration device provided in Example 5 of the present invention. This embodiment can be applied to extracting target key information from multiple unstructured information source documents, and then generating a target structured document based on the extracted target key information. The device can be implemented by software and / or hardware and integrated into a computer device with application development capabilities.

[0136] like Figure 5 As shown, the document information integration device includes: a source document acquisition module 510 , an information extraction module 520 and an information integration module 530 .

[0137] The source document acquisition module 510 is used to acquire a list of objects to be evaluated, and acquire corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents;

[0138] The information extraction module 520 is used to match the content extraction rules corresponding to each document in the list of objects to be evaluated according to the document number of each document in the information source document corresponding to the object to be evaluated, and extract the target key information according to the content extraction rules;

[0139] The information integration module 530 is used to integrate the target key information according to a preset information integration template to generate a target structured document.

[0140] The technical solution of this embodiment is to obtain a list of objects to be evaluated, obtain the corresponding unstructured information source documents according to the list of objects to be evaluated, and for each object to be evaluated, match the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, extract the target key information from the unstructured information source document based on the content extraction rule, and then integrate the target key information extracted according to the preset information integration template and the extracted target structured document, so as to realize the information extraction and information integration of multi-source unstructured documents, without the need to configure the integration template separately for each type of source document, solve the technical problem of relying on complex template configuration for information extraction in the prior art, and improve the efficiency of information extraction and integration of multi-source unstructured documents. In addition, the method can generate structured documents according to the extracted information, solve the problem that the unstructured documents generated after the information is extracted in the prior art are difficult to display flexibly, and the generated structured documents are helpful for analyzing the extracted information, and improve the efficiency of users in obtaining information.

[0141] Optionally, the information extraction module 520 includes a rule determination unit and an information extraction unit, wherein the rule determination unit is used to determine the document type corresponding to the information source document according to the document number, and match the corresponding content extraction rule according to the document type; and the information extraction unit is used to extract target key information from the information source document of the corresponding type based on the preset keyword information in the content extraction rule.

[0142] Optionally, the information extraction unit is specifically used to:

[0143] Obtain a first preset keyword phrase in the preset keyword information; match and extract a first paragraph starting with the first start keyword and a first end keyword according to the first preset keyword phrase, and ending with the first end keyword.

[0144] Optionally, the information extraction unit is also used to obtain a second preset keyword phrase in the preset keyword information; match a second paragraph with the second start keyword as the starting position and the second end keyword as the ending position according to the second start keyword and the second end keyword in the second preset keyword phrase; and extract a sentence containing the first preset keyword in the second paragraph.

[0145] Optionally, the information extraction unit is further used to extract a table or cell content in the table containing a second preset keyword.

[0146] Optionally, the source document acquisition module 510 includes an alarm list acquisition unit, and the alarm list acquisition unit is used to establish a connection with the target risk warning system, request and obtain a business risk alarm list; and use the business risk alarm list as the list of objects to be evaluated.

[0147] Optionally, the source document acquisition module 520 also includes a credit document acquisition unit, which is used to request and obtain the credit document of the object to be evaluated from the preset credit business system; wherein the credit document includes at least one information source document in the investigation report, the preset project evaluation and rating report, the preset business application, the request document and the loan amount application.

[0148] Optionally, the information integration module 530 is specifically used to:

[0149] The target key information content is matched with the document content hierarchy of the preset information integration template; the target key information is integrated according to the matching result of the document content hierarchy to generate a target structured document.

[0150] Optionally, the device further comprises a rule modification module, wherein the rule modification module is used to obtain a user's modification of a target key information extraction item in the content extraction rule, and extract the target key information according to the modified content extraction rule.

[0151] Optionally, the device further comprises a template modification module, wherein the template modification module is used to obtain the modification made by the user to the document content hierarchy structure in the preset information integration template, and integrate the target key information according to the modified document content hierarchy structure.

[0152] Optionally, the device also includes a source document download module, which is used to display information source document download prompt information and download link to the user when the user downloads the target structured document; in response to the user's triggering operation on the download link, download the information source document associated with the target structured document.

[0153] The document information integration device provided in the embodiment of the present invention can execute the document information integration method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0154] Embodiment 6

[0155] Figure 6 A schematic diagram of the structure of a computer device provided in Embodiment 6 of the present invention. Figure 6 A block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 6 The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capabilities, such as an intelligent controller and server, a mobile phone and other terminal devices.

[0156] like Figure 6 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0157] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0158] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0159] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 not shown, usually called a "hard drive"). Although Figure 6Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0160] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0161] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via an input / output (I / O) interface 22. Furthermore, the computer device 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with the other modules of the computer device 12 via the bus 18. It should be understood that although Figure 6 Not shown, other hardware and / or software modules may be used in conjunction with computer device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0162] The processing unit 16 executes various functional applications and data processing by running the programs stored in the system memory 28, for example, implementing the document information integration method provided in the embodiment of the present invention, which includes:

[0163] Obtaining a list of objects to be evaluated, and obtaining corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents;

[0164] For each object to be evaluated in the list of objects to be evaluated, matching the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, and extracting the target key information according to the content extraction rules;

[0165] According to the preset information integration template, the target key information is integrated to generate a target structured document.

[0166] Embodiment 7

[0167] The seventh embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the document information integration method provided in any embodiment of the present invention is implemented, including:

[0168] Obtaining a list of objects to be evaluated, and obtaining corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents;

[0169] For each object to be evaluated in the list of objects to be evaluated, matching the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, and extracting the target key information according to the content extraction rules;

[0170] According to the preset information integration template, the target key information is integrated to generate a target structured document.

[0171] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media 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 this document, 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.

[0172] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0173] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0174] Computer program code for performing the operation of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect through the Internet).

[0175] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0176] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A document information integration method, characterized in that: The method comprises: Obtain a list of objects to be evaluated, and obtain corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents, and the information source documents include at least one of an investigation report, a preset project evaluation and rating report, a preset business declaration, a request document, and a loan amount declaration; For each object to be evaluated in the list of objects to be evaluated, match the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, and extract the target key information according to the content extraction rules; wherein the content extraction rules include structured marks, titles at various levels, serial numbers corresponding to the titles at various levels, preset keyword information, document subtype, extraction method and positioning rules; Integrate the target key information according to the preset information integration template to generate a target structured document; The step of matching the content extraction rules corresponding to each document according to the document number of each document in the information source document corresponding to the object to be evaluated, and extracting the target key information according to the content extraction rules, includes: Determine the document type corresponding to the information source document according to the document number, and match the corresponding content extraction rule according to the document type; Extracting target key information from information source documents of corresponding types based on preset keyword information in the content extraction rule; The step of extracting target key information from information source documents of corresponding types based on preset keyword information in the content extraction rule includes: Acquire a first preset keyword phrase in the preset keyword information; According to the first start keyword and the first end keyword in the first preset keyword phrase, matching and extracting the first paragraph starting at the first start keyword and ending at the first end keyword; The step of extracting target key information from information source documents of corresponding types based on preset keyword information in the content extraction rule also includes: Acquire a second preset keyword phrase in the preset keyword information; According to the second start keyword and the second end keyword in the second preset keyword phrase, matching a second paragraph starting with the second start keyword and ending with the second end keyword; In the second paragraph, sentences containing first preset keywords are extracted.

2. The method according to claim 1, characterized in that: The step of extracting target key information from information source documents of corresponding types based on preset keyword information in the content extraction rule also includes: Extract the table or cell content in the table containing the second preset keyword.

3. The method according to claim 1, characterized in that The obtaining of the list of objects to be evaluated includes: Establish a connection with the target risk early warning system, request and obtain the business risk alarm list; The business risk warning list is used as the list of objects to be evaluated.

4. The method according to claim 3, characterized in that The step of obtaining corresponding information source documents according to the list of objects to be evaluated includes: Request and obtain the credit documents of the subject to be assessed from the preset credit business system; The credit granting document includes at least one information source document selected from the group consisting of an investigation report, a preset project evaluation and rating report, a preset business application form, a request document, and a loan amount application form.

5. The method according to any one of claims 1 to 4, characterized in that: The step of integrating the target key information according to the preset information integration template to generate a target structured document includes: Matching the target key information content with the document content hierarchy of the preset information integration template; The target key information is integrated according to the matching results of the document content hierarchy to generate a target structured document.

6. The method according to claim 1, characterized in that The method further comprises: The modification made by the user to the target key information extraction item in the content extraction rule is obtained, and the target key information is extracted according to the modified content extraction rule.

7. The method according to claim 6, characterized in that The method further comprises: The modification of the document content hierarchy structure in the preset information integration template by the user is obtained, and the target key information is integrated according to the modified document content hierarchy structure.

8. The method according to claim 1, characterized in that: The method further comprises: When the user downloads the target structured document, the user is presented with information source document download prompt information and a download link; In response to a user triggering operation on the download link, the information source document associated with the target structured document is downloaded.

9. A document information integration device, characterized in that: The device comprises: A source document acquisition module, used to acquire a list of objects to be evaluated, and acquire corresponding information source documents according to the list of objects to be evaluated, wherein the information source documents are unstructured documents, and the information source documents include at least one of an investigation report, a preset project evaluation and rating report, a preset business declaration, a request document, and a loan amount declaration; An information extraction module is used to match the content extraction rules corresponding to each document in the information source document corresponding to the object to be evaluated with respect to each object to be evaluated in the list of objects to be evaluated, according to the document number of each document in the information source document corresponding to the object to be evaluated, and extract the target key information according to the content extraction rules; wherein the content extraction rules include structured marks, titles at various levels, serial numbers corresponding to the titles at various levels, preset keyword information, document subtype, extraction method and positioning rules; An information integration module, used to integrate the target key information according to a preset information integration template to generate a target structured document; The information extraction module includes a rule determination unit and an information extraction unit. The rule determination unit is used to determine the document type corresponding to the information source document according to the document number, and match the corresponding content extraction rule according to the document type; The information extraction unit is used to extract target key information from the corresponding type of information source document based on the preset keyword information in the content extraction rule; The information extraction unit is specifically used for: Acquire a first preset keyword phrase in the preset keyword information; match and extract a first paragraph starting with the first start keyword and ending with the first end keyword according to a first start keyword and a first end keyword in the first preset keyword phrase; The information extraction unit is further used to obtain a second preset keyword phrase in the preset keyword information; match a second paragraph starting with the second starting keyword and a second ending keyword in the second preset keyword phrase, and ending with the second ending keyword; and extract a sentence containing the first preset keyword in the second paragraph.

10. A computer device, characterized in that: The computer device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the document information integration method as described in any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the document information integration method as described in any one of claims 1 to 8 is implemented.

12. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the document information integration method according to any one of claims 1 to 8 is implemented.

Images