A method, device and medium for viewing PDF reference documents in intelligent knowledge base

Abstract

The present invention discloses a method for viewing a PDF reference document in an intelligent knowledge base, comprising: an H5 page receives a question input by a user and processes it; the H5 page receives feedback information transmitted back from an AI knowledge base; the H5 page renders according to the received feedback information; when the PDF document content rendering is completed, the H5 page receives a notification sent by an iframe tag through postMessage; when the H5 page receives a notification sent by an iframe tag, the relevant information is transmitted back to the iframe tag; the iframe tag matches and highlights keywords according to the content of the PDF document of a specified page number. The beneficial effect is that through the multi-keyword highlighting function, the user can quickly find all keywords related to the AI ​​answer in the PDF document of the reference source, reducing the time of manual search.

Description

Technical Field

[0001] The present invention belongs to the technical field of software management technology and software management system, and in particular relates to a method, device and medium for viewing PDF reference documents in an intelligent knowledge base. Background Art

[0002] With the continuous advancement of artificial intelligence technology, the application of AI in the field of knowledge processing and information services is becoming increasingly in-depth. In the process of building an intelligent knowledge base, AI systems are widely used to read, analyze and interpret documents in various formats. Among them, PDF documents have become an important carrier for knowledge storage and transmission due to their stable format and detailed content. However, when AI provides users with knowledge question and answer services, although detailed original text links can be attached, users still face a significant challenge when consulting these PDF documents: how to quickly and accurately find the original text fragments corresponding to the AI ​​answers in the huge text content. Summary of the invention

[0003] To solve the problems raised in the above background technology, the present invention provides a method for viewing PDF reference documents in an intelligent knowledge base.

[0004] To achieve the above object, the present invention provides the following technical solution: a method for viewing PDF reference documents in an intelligent knowledge base, comprising:

[0005] The H5 page receives questions entered by users and sends them to the AI ​​knowledge base;

[0006] The H5 page receives feedback information sent back from the AI ​​knowledge base; wherein the feedback information includes: PDF information and keyword information;

[0007] Sending PDF information and a rendering signal to an iframe tag, wherein the rendering signal is used to trigger the iframe tag to render the content of the PDF document of a specified page number;

[0008] The H5 page receives the rendering completion signal of the iframe tag, and sends the keyword information and the highlight signal to the iframe tag; wherein the highlight signal is used to trigger the iframe tag to match and highlight the keywords in the content of the PDF document of the specified page number;

[0009] The H5 page receives the iframe tag to complete the highlight signal.

[0010] The complete process from user inputting questions to displaying the processed PDF document content on the H5 page is realized. Users can easily input questions in a unified interface and obtain PDF reference documents with highlighted keywords, which improves the efficiency and convenience of information acquisition.

[0011] Specifically: the method for generating feedback information includes:

[0012] The AI ​​knowledge base uses intelligent semantic matching algorithms to find relevant answers in the knowledge base and generate feedback information.

[0013] By generating feedback information through intelligent semantic matching algorithms, it is possible to find answers related to user questions more accurately, improving the query accuracy and response speed of the knowledge base.

[0014] Specifically: PDF information includes: the address of the answer reference PDF document and the page number of the answer reference PDF document source;

[0015] The keyword information includes: answer, answer keyword and answer keyword absolute position.

[0016] By determining the specific content of PDF information and keyword information, the system can be more accurate and targeted when processing and displaying documents, making it easier for users to quickly locate key information.

[0017] Specifically: The iframe tag matches and highlights keywords in the content of the PDF document of the specified page number, including:

[0018] The iframe tag matches all keywords in the rendered PDF document content based on the PDF document content and keyword information of the specified page number;

[0019] The specific position of the keyword is located according to its absolute position, and all matched keywords are highlighted.

[0020] It achieves accurate matching and highlighting of keywords in PDF document content, helping users quickly identify key content when viewing documents, improving the efficiency and accuracy of document reading.

[0021] Specifically: After the H5 page receives the iframe tag to complete the highlight signal, it also includes:

[0022] Determine whether there are multiple reference sources. If yes, display the reference source switch button, otherwise do not display it.

[0023] If there are multiple reference sources, determine whether there is a specified reference source. If yes, set the current reference source according to the specified one. If no, set the first reference source as the current reference source by default.

[0024] It provides the function of judging multiple reference sources and switching between them, meeting the needs of users to compare and select between different reference documents, and increasing the flexibility and practicality of the system.

[0025] PDF highlights reference sources based on the current reference source.

[0026] Specifically: The iframe tag matches and highlights keywords in the content of the PDF document of the specified page number, including:

[0027] When the PDF document is located at the specified page number and rendered, obtain all the text information on the page number and the absolute position information corresponding to the text, and record it as allTextData;

[0028] Receive keyword information and record it as keywordsData;

[0029] Compare each text segment in allTextData with each keyword in keywordsData through variable declaration and conditional judgment;

[0030] Determine whether the text contains the keyword and whether the position matches based on the comparison result;

[0031] If there is a match, the keyword is added to the matches array;

[0032] If it does not match, skip it;

[0033] Until allTextData and keywordsData are traversed;

[0034] According to the traversal results, all matching matches data are highlighted and rendered on the PDF document page;

[0035] Matches usually refers to matching items in a programming context. In this application, it refers to a set of text data that matches a keyword.

[0036] keywordsData is keyword data, which in this application refers to a data set that records keywords and their absolute location information.

[0037] By comparing and matching text information and keywords and processing matches data, more accurate keyword highlighting is achieved. At the same time, the previous and next reference source switching function is provided to facilitate users to fully understand relevant information, improving user experience and the integrity of information acquisition.

[0038] allTextData is all text data. In this application, it refers to the collection of all text information in a specified page number of the PDF and the absolute position information corresponding to the text.

[0039] The H5 page provides the previous and next buttons. When the user clicks the previous or next reference source button, it is determined whether there is a previous or next reference source. If so, the current reference source is set accordingly. If not, the last or first reference source is set as the current reference source. The determination process ends.

[0040] When the process is judged to be finished, a PDF document address is generated according to the current reference source, and multiple keywords are highlighted;

[0041] Specifically: the iframe tag matches and highlights keywords in the PDF document content of the specified page number;

[0042] include:

[0043] When the specific location of the keyword is determined, the data in matches is sorted according to the location information;

[0044] When traversing the sorted matches data, determine whether two adjacent characters are in the same line or adjacent lines;

[0045] If yes, merge the two text data and repeat the process until there are no adjacent texts.

[0046] If not, skip;

[0047] For the merged text and position, draw a div element highlight prompt on the PDF document preview page.

[0048] Sorting and merging the data in matches makes the highlight prompts clearer and more coherent, avoiding the confusion caused by scattered highlights and further improving users' recognition and understanding of key information.

[0049] In HTML, the div element It is a block-level element that can be used for layout and style settings. In this application, it is used to draw highlight prompts on the PDF document preview page.

[0050] A device for viewing PDF reference documents in an intelligent knowledge base: the device comprises: a receiving module for receiving user input questions sent by an iframe tag and feedback information sent back by an AI knowledge base;

[0051] The sending module is used to send rendering signals to the iframe tag;

[0052] Display module, used to display the rendering completion interface;

[0053] A computer-readable storage medium stores program instructions therein, and the program instructions are used to implement a method for viewing PDF reference documents in an intelligent knowledge base when executed by a processor.

[0054] The present invention solves the defects existing in the background technology and has the following beneficial effects:

[0055] 1. The method of viewing PDF reference documents in the intelligent knowledge base. Through the multi-keyword highlighting function, users can quickly find all keywords related to AI answers in the PDF documents of the reference source, reducing the time of manual search and improving the accuracy and efficiency of information retrieval compared with the existing technology.

[0056] 2. The method for viewing PDF reference documents in the intelligent knowledge base combines the search capability of the intelligent knowledge base, the PDF preview capability of the iframe tag and the positioning capability of the browser H5 page button, and can realize the search and highlighting of multi-keyword blocks or reference paragraphs for the specified number of pages, and quickly identify and locate the next / previous reference paragraph. Compared with the prior art, the work efficiency is improved.

[0057] 3. The method for viewing PDF reference documents in the intelligent knowledge base is closely integrated with the AI ​​knowledge base question and answer system, making the AI ​​answers more specific and traceable, enhancing users' trust in AI answers; compared with the existing technology, it enhances the practicality of AI knowledge base question and answer. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0059] Figure 1 It is a schematic diagram of the process of the present invention;

[0060] Figure 2 It is a schematic diagram of the process in the present invention;

[0061] Figure 3 It is a schematic diagram of the process in the present invention;

[0062] Figure 4 It is a schematic diagram of the process in the present invention; DETAILED DESCRIPTION

[0063] The present invention will now be further described in detail with reference to the accompanying drawings and embodiments. These drawings are simplified schematic diagrams that only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0064] Application Overview:

[0065] A method for viewing a PDF reference document in a browser H5 page includes using a nested iframe tag to implement the method, using a pdfjs-dist library in the iframe tag to implement PDF document preview, and the browser H5 page and the iframe tag communicate through postMessage, thereby realizing viewing the PDF reference document in the browser H5 page using the iframe tag.

[0066] However, the pdfjs-dist library does not support multiple keywords at the same time, or specifying the number of pages to match keywords. In addition, the PDF preview function of the iframe tag and the highlight function are usually limited to searching and marking a single word or phrase. This method may be sufficient for simple queries, but it is not enough for complex query requirements.

[0067] For example, when users need to find an entire article segment containing multiple keywords or phrases, the search highlight function is often not competent. When highlighting, only a single word can be marked, and it is impossible to achieve continuous highlighting of the entire article segment, which further increases the difficulty of users' search and understanding. The PDF preview function is not competent for quickly locating the next matching paragraph in the document.

[0068] The present invention combines the retrieval capability of the intelligent knowledge base, the PDF preview capability of the iframe tag and the positioning capability of the browser H5 page button, and can realize the retrieval and highlighting of multi-keyword blocks or reference paragraphs for a specified number of pages, and quickly identify and locate the next / previous reference paragraph.

[0069] Exemplary methods:

[0070] like Figure 1 FIG. 1 is a flowchart of a method for viewing PDF reference documents in an intelligent knowledge base in the present invention.

[0071] The method specifically comprises the following steps:

[0072] A method for viewing a PDF reference document in an intelligent knowledge base, comprising:

[0073] The H5 page receives questions entered by users and sends them to the AI ​​knowledge base;

[0074] The H5 page receives feedback information sent back from the AI ​​knowledge base; wherein the feedback information includes: PDF information and keyword information;

[0075] Sending PDF information and a rendering signal to an iframe tag, wherein the rendering signal is used to trigger the iframe tag to render the content of the PDF document of a specified page number;

[0076] The H5 page receives the rendering completion signal of the iframe tag, and sends the keyword information and the highlight signal to the iframe tag; wherein the highlight signal is used to trigger the iframe tag to match and highlight the keywords in the content of the PDF document of the specified page number;

[0077] The H5 page receives the iframe tag to complete the highlight signal.

[0078] Specifically: the method for generating feedback information includes:

[0079] The AI ​​knowledge base uses intelligent semantic matching algorithms to find relevant answers in the knowledge base and generate feedback information.

[0080] Specifically: PDF information includes: the address of the answer reference PDF document and the page number of the answer reference PDF document source;

[0081] The keyword information includes: answer, answer keyword and answer keyword absolute position.

[0082] Specifically: The iframe tag matches and highlights keywords in the content of the PDF document of the specified page number, including:

[0083] The iframe tag matches all keywords in the rendered PDF document content based on the PDF document content and keyword information of the specified page number;

[0084] The specific position of the keyword is located according to its absolute position, and all matched keywords are highlighted.

[0085] Specifically: After the H5 page receives the iframe tag to complete the highlight signal, it also includes:

[0086] Determine whether there are multiple reference sources. If yes, display the reference source switch button, otherwise do not display it.

[0087] If there are multiple reference sources, determine whether there is a specified reference source. If yes, set the current reference source according to the specified one. If no, set the first reference source as the current reference source by default.

[0088] PDF highlights reference sources based on the current reference source.

[0089] Specifically: The iframe tag matches and highlights keywords in the content of the PDF document of the specified page number, including:

[0090] When the PDF document is located at the specified page number and rendered, obtain all the text information on the page number and the absolute position information corresponding to the text, and record it as allTextData;

[0091] Receive keyword information and record it as keywordsData;

[0092] Compare each text segment in allTextData with each keyword in keywordsData through variable declaration and conditional judgment;

[0093] Determine whether the text contains the keyword and whether the position matches based on the comparison result;

[0094] If there is a match, the keyword is added to the matches array;

[0095] If it does not match, skip it;

[0096] Until allTextData and keywordsData are traversed;

[0097] According to the traversal results, all matching matches data are highlighted and rendered on the PDF document page;

[0098] The H5 page provides the previous and next buttons. When the user clicks the previous or next reference source button, it is determined whether there is a previous or next reference source. If so, the current reference source is set accordingly. If not, the last or first reference source is set as the current reference source. The determination process ends.

[0099] When the process is judged to be finished, a PDF document address is generated according to the current reference source, and multiple keywords are highlighted;

[0100] Specifically: the iframe tag matches and highlights keywords in the PDF document content of the specified page number;

[0101] include:

[0102] When the specific location of the keyword is determined, the data in matches is sorted according to the location information;

[0103] When traversing the sorted matches data, determine whether two adjacent characters are in the same line or adjacent lines;

[0104] If yes, merge the two text data and repeat the process until there are no adjacent texts.

[0105] If not, skip;

[0106] For the merged text and position, draw a div element highlight prompt on the PDF document preview page.

[0107] like Figure 4 A device for viewing PDF reference documents in an intelligent knowledge base is shown: the device includes: a receiving module, which is used to receive user input questions sent by an iframe tag and feedback information sent back by the AI ​​knowledge base;

[0108] The sending module is used to send rendering signals to the iframe tag;

[0109] Display module, used to display the rendering completion interface;

[0110] A computer-readable storage medium stores program instructions therein: the program instructions are executed by a processor to provide a method for viewing PDF reference documents in an intelligent knowledge base.

[0111] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of viewing PDF reference documents in the intelligent knowledge base according to various embodiments of the present application described in the above-mentioned method section of this specification.

[0112] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user computing device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0113] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of viewing PDF reference documents in the intelligent knowledge base according to various embodiments of the present application described in the above method section of this specification.

[0114] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable 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.

[0115] Specifically, Figure 2 The steps of the flowchart shown are as follows:

[0116] Step S1, determining the number of reference sources in the PDF document for the answer returned by the AI ​​knowledge base;

[0117] If the number of reference sources is greater than 1, the H5 page displays a switch reference source button and proceeds to step S2;

[0118] If the number of reference sources is equal to 1, the H5 page does not display the reference source button, and proceeds to step S5;

[0119] Step S2: for the reference source of the answer returned by the AI ​​knowledge base in the PDF document, determine whether there is a designated reference source;

[0120] If there is a specified reference source, proceed to step S3;

[0121] If no reference source is specified, proceed to step S5;

[0122] Step S3, setting the current reference source according to the specified reference source;

[0123] Step S4: Generate a PDF document address based on the current reference source, and highlight multiple keywords, and the current process ends;

[0124] Step S5, setting the first reference source as the current reference source by default, and then proceeding to step S3;

[0125] Step S6: In case of multiple reference sources;

[0126] If the user clicks the previous reference source button, the process goes to step S7;

[0127] The user clicks the next reference source button, and then proceeds to step S8;

[0128] Step S7, determining whether there is a previous reference source;

[0129] If there is a previous reference source, the previous reference source is set as the current reference source, and the process goes to step S3;

[0130] If there is no previous reference source, the last reference source is set as the current reference source, and the process goes to step S3;

[0131] Determine whether there is a next reference source;

[0132] Step S8, determining whether there is a next reference source;

[0133] If there is a next reference source, set the next reference source as the current reference source and go to step 3;

[0134] If there is no next reference source, set the first reference source as the current reference source and go to step 3;

[0135] Specifically, Figure 3 The steps of the flowchart shown are as follows:

[0136] Step S1, the PDF document is located at a specified page number, then rendered, and step S2 is executed;

[0137] S2, obtain all text information and the absolute position information corresponding to the text in the specified page number, record the variable as allTextData, and execute step S3;

[0138] S3, receiving multiple keywords and their absolute position information, recording the variable as keywordsData, and executing step S4;

[0139] S4, declare variable i=0, declare variable matches=[], and execute step S5;

[0140] S5, declare variable tData = allTextData[i], declare variable tPos = tData.position, and execute step S6;

[0141] S6, declare variable j=0, and execute step S7;

[0142] S7, declare variable wData = keywordsData[j], declare variable wPos = wData.position, and execute step S8;

[0143] S8. If tData.text contains wData.text and tPos.startX<=wPos.startX and tPos.endX>=wPos.endX and tPos.startY<=wPos.startY and tPos.endY>=wPos.endY, execute step S9; otherwise, execute step S10;

[0144] S9. Add wData to matches, matched.push(wData);

[0145] S10, add 1 to variable j, j++;

[0146] S11, if j is less than the length of keywordsData, execute step S7, otherwise execute step S12;

[0147] S12, add 1 to variable i, i++;

[0148] S13, if i is less than the length of allTextData, execute step S5, otherwise execute step S14;

[0149] S14. All matching matches data are highlighted and rendered on the PDF document page, and the process ends.

[0150] Through the above process, we can obtain all the texts and their location information of all the exact matches of the specified page number. If we directly highlight the text in the matches data, scattered highlighted texts will appear, and it is impossible to form a whole block of highlighted texts. If we want to highlight the whole block of text, some calculations are required. The logic of the whole block highlighting function is as follows:

[0151] Step S1, first sort the data in matches according to the location information;

[0152] Step S2, traverse the data in the sorted matches again, if two adjacent characters are in the same row or adjacent rows, merge the two character data, and repeat the traversal until there are no adjacent characters;

[0153] Step S3: Draw a div element highlight prompt for the text and position after the matches are merged on the PDF document preview page.

[0154] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0155] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example.

[0156] 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 alter, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for viewing PDF reference documents in an intelligent knowledge base, characterized in that include: The H5 page receives questions entered by users and sends them to the AI ​​knowledge base; The H5 page receives feedback information sent back from the AI ​​knowledge base; wherein the feedback information includes: PDF information and keyword information; Sending PDF information and a rendering signal to an iframe tag, wherein the rendering signal is used to trigger the iframe tag to render the content of the PDF document of a specified page number; The H5 page receives the rendering completion signal of the iframe tag, and sends the keyword information and the highlight signal to the iframe tag; wherein the highlight signal is used to trigger the iframe tag to match and highlight the keywords in the content of the PDF document of the specified page number; The H5 page receives the iframe tag to complete the highlight signal; The iframe tag matches and highlights keywords in the content of the PDF document of the specified page number, including: The iframe tag matches all keywords in the rendered PDF document content based on the PDF document content and keyword information of the specified page number; Locate the specific position of the keyword according to its absolute position, and highlight all the matched keywords; The iframe tag matches and highlights keywords in the content of the PDF document of the specified page number, including: When the PDF document is located at the specified page number and rendered, obtain all the text information on the page number and the absolute position information corresponding to the text, and record it as allTextData; Receive keyword information and record it as keywordsData; Compare each text segment in allTextData with each keyword in keywordsData through variable declaration and conditional judgment; Determine whether the text contains the keyword and whether the position matches based on the comparison result; If there is a match, the keyword is added to the matches array; If it does not match, skip it; Until allTextData and keywordsData are traversed; According to the traversal results, all matching matches data are highlighted and rendered on the PDF document page; The H5 page provides the previous and next buttons. When the user clicks the previous or next reference source button, it is determined whether there is a previous or next reference source. If so, the current reference source is set accordingly. If not, the last or first reference source is set as the current reference source. The determination process ends. When the process is judged to be finished, a PDF document address is generated according to the current reference source, and multiple keywords are highlighted; The iframe tag matches and highlights keywords in the PDF document content of the specified page number; include: When the specific location of the keyword is determined, the data in matches is sorted according to the location information; When traversing the sorted matches data, determine whether two adjacent characters are in the same line or adjacent lines; If yes, merge the two text data and repeat the process until there are no adjacent texts. If not, skip; For the merged text and position, draw a div element highlight prompt on the PDF document preview page.

2. The method for viewing PDF reference documents in an intelligent knowledge base according to claim 1, characterized in that: The method for generating feedback information includes: The AI ​​knowledge base uses intelligent semantic matching algorithms to find relevant answers in the knowledge base and generate feedback information.

3. The method for viewing PDF reference documents in an intelligent knowledge base according to claim 1, characterized in that: The PDF information includes: the address of the answer reference PDF document and the page number of the answer reference PDF document source; The keyword information includes: answer, answer keyword and answer keyword absolute position.

4. The method for viewing PDF reference documents in an intelligent knowledge base according to claim 1, characterized in that: After the H5 page receives the iframe tag to complete the highlight signal, it also includes: Determine whether there are multiple reference sources. If yes, display the reference source switch button, otherwise do not display it. If there are multiple reference sources, determine whether there is a specified reference source. If yes, set the current reference source according to the specified one. If no, set the first reference source as the current reference source by default. PDF highlights reference sources based on the current reference source.

5. A device for viewing PDF reference documents in an intelligent knowledge base, based on a method for viewing PDF reference documents in an intelligent knowledge base according to any one of claims 1 to 4, characterized in that: The device includes: The receiving module is used to receive the user input questions sent by the iframe tag and the feedback information sent back by the AI ​​knowledge base; The sending module is used to send rendering signals to the iframe tag; The display module is used to display the rendering completion interface.

6. A computer-readable storage medium having program instructions stored therein, characterized in that: When the program instructions are executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images