Scientific and technological information integrated management method and system

Abstract

The invention discloses a scientific and technological information comprehensive management method and system, and belongs to the technical field of information recommendation, and the method specifically comprises the following steps: obtaining browsed data and non-browsed data after a target object is retrieved; determining a target literature according to the browsing data, obtaining a related technical field n corresponding to the target literature, and sequentially calculating the interestingness of any related field; extracting an abstract text in the target literature and preprocessing the abstract text to obtain a comprehensive conclusion text; any to-be-determined literature is selected, text duplicate checking calculation is conducted on the preprocessed abstract text and the comprehensive conclusion text, and text duplicate checking values of all to-be-determined literatures are obtained; the text duplicate checking value of the to-be-determined literature is corrected according to the interestingness, all the to-be-determined literature is sorted from large to small, literature recommendation is conducted on the target object according to the sorting result, the retrieval efficiency is improved, the burden of a user for screening massive literature is reduced, and the scientific and technological information achievement conversion efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of information recommendation, and in particular to a comprehensive management method and system for scientific and technological information. Background Art

[0002] Room 506, No. G900, Rundong Industrial Park, Guangming Avenue, Xingcheng Street, High-tech Zone The transaction and transformation of scientific and technological achievements is one of the important links to promote scientific and technological innovation and economic development. It is usually a research result reflected in the form of papers, works, patents, etc. published through research. Scientific and technological achievements will form a large amount of integrated information data in the process of transaction and transformation. Only a small part of the cutting-edge scientific and technological achievements are applied to actual production, and there are even fewer scientific and technological achievements that can truly form industries. A big reason for the above situation is the lack of an effective supply and demand docking platform. Specifically, scientific and technological achievements are in the hands of various universities and scientific research institutions, but universities and scientific research institutions do not have the equipment and funds for the transformation of achievements; on the other hand, enterprises do not have the ability to identify technology, and are not clear about the development trends and industrial policy directions; therefore, promoting the transformation of achievements requires a professional third party to mediate and provide professional guidance on the transformation of scientific and technological achievements; In the related technology, the integrated information of scientific and technological achievements in transaction transformation is usually stored and managed in a unified and centralized manner. However, with the increasing number of scientific and technological achievements, the demander first searches and pays attention to some key information such as the title, keywords, abstract, etc. in the achievement documents based on the search results, and judges whether there is value for further reading / transaction based on these key information. For example, the demander clicks on the corresponding link, opens the corresponding document, and then reads the abstract to find the key information. If the document is valuable, it will be browsed. However, when a large number of achievement documents need to be viewed, multiple repeated operations will be performed, which is cumbersome and time-consuming, and the actual retrieval efficiency is low, resulting in low efficiency in managing the integrated information of scientific and technological achievements in transaction transformation. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for comprehensive management of scientific and technological information to solve the following technical problems: In related technologies, the integrated information of scientific and technological achievements in transaction transformation is usually stored and managed in a unified and centralized manner. However, with the increasing number of scientific and technological achievements, the demanders first search and pay attention to some key information such as titles, keywords, abstracts, etc. in the achievement documents based on the search results. The actual retrieval efficiency is not high, which leads to low efficiency in the management of the integrated information of scientific and technological achievements in transaction transformation.

[0004] The purpose of the present invention can be achieved through the following technical solutions: A method for comprehensive management of scientific and technological information comprises the following steps: S1, obtaining browsed data and unbrowsed data of the target object after retrieval, wherein the browsed data is the result documents browsed after retrieval, the document browsing time and the document download record; the unbrowsed data is the result document set that has not been browsed after retrieval; S2, determining the target document according to the browsing data, obtaining the relevant technical fields corresponding to the target document and marking them as y1, y2, ..., yn, calculating the interest of any of the relevant fields in turn and marking them as P1, P2, ..., Pn; extracting the abstract text in the target document and preprocessing it to obtain a comprehensive conclusion text; Select any pending document from the unbrowsed achievement document set, extract the abstract text in the pending document and perform preprocessing, perform text duplication calculation on the preprocessed abstract text and the comprehensive conclusion text, obtain the text duplication values ​​of all pending documents and mark them as C1, C2, ..., Cm; S3, obtaining the relevant technical field corresponding to any pending document, and determining the interest of the pending document according to the relevant technical field, correcting the text duplication value of the pending document according to the interest, sorting all pending documents from large to small according to the corrected text duplication value, and recommending pending documents to the target object according to the sorting result.

[0005] As a further solution of the present invention: S1 further includes: The literature database is screened according to the search terms entered by the target object to obtain the search data, which are then sorted according to the publication order of the literature, the relevance of the article title to the search terms, and the impact factor.

[0006] As a further solution of the present invention: In S2, the process of determining the target document is as follows: Obtain the document browsing time t and document download record S of any browsed document. If S is greater than 0, calculate the recommendation score W of all browsed documents according to the calculation formula W=b*t+d*S, where b and d are preset coefficients and b is less than d; if S is equal to 0, calculate the recommendation score of all browsed documents according to the calculation formula W=b*t; if there is any browsed document whose recommendation score W is greater than or equal to the preset threshold, mark the browsed document as the target text.

[0007] As a further solution of the present invention: in S2, the calculation process of the interest degree of the technical field is: S11, obtaining the recommendation scores of all browsed documents and summing them up to obtain a total recommendation score value W', calculating the ratio of any browsed document to the recommendation score, performing weighted summation based on the browsed document recommendation score and the ratio, and obtaining the interest level X of any browsed document; S12, obtaining all relevant technical fields corresponding to any browsed document and defining the interest degree of any relevant technical field corresponding to the browsed document as the interest degree of the browsed document; obtaining the relevant technical fields corresponding to the target document, and calculating the interest degree of each relevant technical field of the target document, the specific calculation is as follows: ; Pm represents the interest level of the target document corresponding to the relevant technical field m, i represents the browsed document corresponding to the relevant technical field m, n is the number of browsed documents corresponding to the relevant technical field m, and Xi represents the interest level corresponding to the browsed document i.

[0008] As a further solution of the present invention: in S2, the specific process of pretreatment is: Performing text analysis on the summary text to retrieve whether a preset signal word exists in the summary text; if the preset signal word exists in the summary text, extracting sentence information after the preset signal word, and using the sentence information as conclusion information in the summary text; if the preset signal word does not exist in the summary text, extracting sentence information of a preset part from the summary text, and using the sentence information as conclusion information in the summary text; using semantic network technology to construct a semantic relationship graph between conclusion texts, and using NLP technology to integrate several conclusion texts into a comprehensive conclusion text according to the semantic relationship graph.

[0009] As a further solution of the present invention: In S2, the calculation process of the duplicate check value is: The summary text and the comprehensive conclusion text are respectively converted into two groups of ASCII code values ​​in character order; the features of each character are represented by the ASCII code values ​​of the surrounding characters; the features of each character point are reduced in dimension by using an autoencoder; the similarity between the two texts is calculated using the Jaccard similarity coefficient for the reduced-dimensional features; The calculation formula for the text duplication value is: ; Among them, A is the feature set of the comprehensive conclusion text, B is the feature set of the abstract text corresponding to any pending document, and J is the duplicate check value.

[0010] As a further solution of the present invention: in S3, the specific process of correction is: Obtain all relevant technical fields corresponding to the pending document and the target document, and then obtain all overlapping technical fields and mark them as Y1, Y2, ..., YN, obtain the interest level of any overlapping technical field and mark it as P1, P2, ..., PN, and calculate according to the formula The corrected text duplication check value C' is calculated, where α is a preset correction coefficient and Pi is the interest level of any overlapping technical field.

[0011] A comprehensive management system for scientific and technological information, comprising: A data acquisition module is used to acquire the browsed data and unbrowsed data of the target object after retrieval. The browsed data is the result documents browsed after retrieval, the document browsing time and the document download record; the unbrowsed data is the result document set that has not been browsed after retrieval; A data analysis module is used to determine the target document according to the browsing data, obtain the relevant technical fields corresponding to the target document and mark them as y1, y2, ..., yn, calculate the interest of any of the relevant fields in turn and mark them as P1, P2, ..., Pn; extract the abstract text in the target document and perform preprocessing to obtain a comprehensive conclusion text; Select any pending document from the unbrowsed achievement document set, extract the abstract text in the pending document and perform preprocessing, perform text duplication calculation on the preprocessed abstract text and the comprehensive conclusion text, obtain the text duplication values ​​of all pending documents and mark them as C1, C2, ..., Cm; The result generation module is used to obtain the relevant technical field corresponding to any pending document, determine the interest of the pending document according to the relevant technical field, correct the text duplication value of the pending document according to the interest, and sort all the pending documents from large to small according to the corrected text duplication value and recommend pending documents to the target object according to the sorting result.

[0012] Beneficial effects of the present invention: The present invention first obtains the text browsing interest field of the target object by analyzing the browsing data of the user, and at the same time determines the target text and obtains the abstract text of the target text. It can be understood that the text abstract is intended to provide the main viewpoints, findings, methods and conclusions of the full text. The abstract text is preprocessed to obtain a comprehensive conclusion text containing key conclusion information from the target text. By calculating the similarity between the abstract text and the comprehensive conclusion text of the pending document, the document most relevant to the target document can be preferentially recommended. It can be understood that the classification and indexing of texts largely depend on the selection of keywords. Since keywords may have multiple meanings, similar texts may be classified into different technical fields. Therefore, only recommending based on the similarity between the pending text and the target text may result in that even if the recommended text is very similar to the target text, it may not be the technical field of interest to the user. Therefore, the present invention modifies the recommendation value of the pending document by considering the user's interest in different technical fields, so as to better reflect the real similarity between documents, improve the accuracy of the recommendation system and user experience, improve the retrieval efficiency, reduce the burden of users screening massive documents, and improve the efficiency of conversion of scientific and technological information results. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The present invention will be further described below in conjunction with the accompanying drawings.

[0014] Figure 1 It is a schematic diagram of a flow chart of a comprehensive management method for scientific and technological information of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] See also Figure 1 As shown, the present invention is a method for comprehensive management of scientific and technological information, comprising the following steps: S1, obtaining browsed data and unbrowsed data of the target object after retrieval, wherein the browsed data is the result documents browsed after retrieval, the document browsing time and the document download record; the unbrowsed data is the result document set that has not been browsed after retrieval; S2, determining the target document according to the browsing data, obtaining the relevant technical fields corresponding to the target document and marking them as y1, y2, ..., yn, calculating the interest of any of the relevant fields in turn and marking them as P1, P2, ..., Pn; extracting the abstract text in the target document and preprocessing it to obtain a comprehensive conclusion text; Select any pending document from the unbrowsed achievement document set, extract the abstract text in the pending document and perform preprocessing, perform text duplication calculation on the preprocessed abstract text and the comprehensive conclusion text, obtain the text duplication values ​​of all pending documents and mark them as C1, C2, ..., Cm; S3, obtaining the relevant technical field corresponding to any pending document, and determining the interest of the pending document according to the relevant technical field, correcting the text duplication value of the pending document according to the interest, sorting all pending documents from large to small according to the corrected text duplication value, and recommending pending documents to the target object according to the sorting result.

[0017] The present invention first obtains the text browsing interest field of the target object by analyzing the browsing data of the user, and at the same time determines the target text and obtains the abstract text of the target text. It can be understood that the text abstract is intended to provide the main viewpoints, findings, methods and conclusions of the full text. The abstract text is preprocessed to obtain a comprehensive conclusion text containing key conclusion information from the target text. By calculating the similarity between the abstract text and the comprehensive conclusion text of the pending document, the document most relevant to the target document can be preferentially recommended. It can be understood that the classification and indexing of texts largely depend on the selection of keywords. Since keywords may have multiple meanings, similar texts may be classified into different technical fields. Therefore, only recommending based on the similarity between the pending text and the target text may result in that even if the recommended text is very similar to the target text, it may not be the technical field of interest to the user. Therefore, the present invention modifies the recommendation value of the pending document by considering the user's interest in different technical fields, so as to better reflect the real similarity between documents, improve the accuracy of the recommendation system and user experience, improve the retrieval efficiency, reduce the burden of users screening massive documents, and improve the efficiency of conversion of scientific and technological information results.

[0018] In a preferred embodiment of the present invention, the S1 further includes: The literature database is screened according to the search terms entered by the target object to obtain the search data, which are then sorted according to the publication order of the literature, the relevance of the article title to the search terms, and the impact factor.

[0019] In a preferred embodiment of the present invention, in S2, the process of determining the target document is as follows: Obtain the document browsing time t and document download record S of any browsed document. If S is greater than 0, calculate the recommendation score W of all browsed documents according to the calculation formula W=b*t+d*S, where b and d are preset coefficients and b is less than d; if S is equal to 0, calculate the recommendation score of all browsed documents according to the calculation formula W=b*t; if there is any browsed document whose recommendation score W is greater than or equal to the preset threshold, mark the browsed document as the target text.

[0020] Normally, readers pay attention to some key information in the text, such as the title, keywords, abstract, etc., and judge whether it is worthwhile to further read the full text of the document based on this key information. For example, readers click on the corresponding link, open the corresponding document, and then read the abstract to find the key information. If they determine that the document is valuable, they will browse it. Some users may tend to browse for a long time without downloading, while others may tend to download without in-depth browsing. Whether the document meets the needs of the target object is measured by browsing time and number of downloads.

[0021] In a preferred embodiment of the present invention, in S2, the process of calculating the interest level of the technical field is as follows: S11, obtaining the recommendation scores of all browsed documents and summing them up to obtain a total recommendation score value W', calculating the ratio of any browsed document to the recommendation score, performing weighted summation based on the browsed document recommendation score and the ratio, and obtaining the interest level X of any browsed document; S12, obtaining all relevant technical fields corresponding to any browsed document and defining the interest degree of any relevant technical field corresponding to the browsed document as the interest degree of the browsed document; obtaining the relevant technical fields corresponding to the target document, and calculating the interest degree of each relevant technical field of the target document, the specific calculation is as follows: ; Pm represents the interest level of the target document corresponding to the relevant technical field m, i represents the browsed document corresponding to the relevant technical field m, n is the number of browsed documents corresponding to the relevant technical field m, and Xi represents the interest level corresponding to the browsed document i.

[0022] In a preferred embodiment of the present invention, in S2, the specific process of preprocessing is: Performing text analysis on the summary text to retrieve whether a preset signal word exists in the summary text; if the preset signal word exists in the summary text, extracting sentence information after the preset signal word, and using the sentence information as conclusion information in the summary text; if the preset signal word does not exist in the summary text, extracting sentence information of a preset part from the summary text, and using the sentence information as conclusion information in the summary text; using semantic network technology to construct a semantic relationship graph between conclusion texts, and using NLP technology to integrate several conclusion texts into a comprehensive conclusion text according to the semantic relationship graph.

[0023] It is understandable that the text summary is intended to provide the main ideas, findings, methods and conclusions of the full text, which is a summary of the literature. Therefore, the summary text information of the target text is extracted and the conclusion is extracted. It is understandable that the summary usually summarizes some existing literature technologies for the purpose of elicitation, which has little actual reference value. Therefore, conclusion extraction is performed and conclusion extraction is performed on the browsing data, and text duplication checking is performed based on the conclusion extraction. It is understandable that if the summary text conclusion is not extracted and the text duplication calculation is performed directly, it will result in the need to transcode each character in the summary text, which contains a large amount of unnecessary text, which not only leads to the amount of data processing tasks, but also increases the amount of calculation for subsequent text duplication checking. In addition, there are a large number of literature technologies written for the purpose of elicitation in the summary text. When there are similar literature technology summaries, it will affect the final text duplication checking results, resulting in inaccurate final recommendation results.

[0024] It is understandable that considering that in the abstracts of many documents, there are no clear signal words or keyword prompts in the conclusion part, or in the abstracts of some documents there is no conclusion part at all, and it is necessary to rely on semantic understanding to judge, when summarizing the contents of the abstracts of these documents, since the content of the abstract summary is generally placed at the end of the entire abstract, it is possible to consider obtaining it from the latter part of the abstract information, specifically, obtaining a preset number of sentences from the back to the front in the abstract information, and using the preset number of sentences (for example, 3 sentences, that is, the sentence information of the penultimate sentence, the second to last sentence and the third to last sentence) as the conclusion information in the abstract information.

[0025] In a preferred embodiment of the present invention, in S2, the calculation process of the duplicate check value is: The summary text and the comprehensive conclusion text are respectively converted into two groups of ASCII code values ​​in character order; the features of each character are represented by the ASCII code values ​​of the surrounding characters; the features of each character point are reduced in dimension by using an autoencoder; the similarity between the two texts is calculated using the Jaccard similarity coefficient for the reduced-dimensional features; The calculation formula for the text duplication value is: ; Among them, A is the feature set of the comprehensive conclusion text, B is the feature set of the abstract text corresponding to any pending document, and J is the duplicate check value.

[0026] In a preferred embodiment of the present invention, in S3, the specific process of correction is: Obtain all relevant technical fields corresponding to the pending document and the target document, and then obtain all overlapping technical fields and mark them as Y1, Y2, ..., YN, obtain the interest level of any overlapping technical field and mark it as P1, P2, ..., PN, and calculate according to the formula The corrected text duplication check value C' is calculated, where α is a preset correction coefficient and Pi is the interest level of any overlapping technical field.

[0027] It is understandable that similar texts may have different technical fields. It is understandable that the classification and indexing of texts largely depend on the selection of keywords. Since keywords may have multiple meanings, similar texts may be classified into different technical fields. Therefore, only recommending based on the similarity between the pending text and the target text may result in the recommended text being very similar to the target text but not being in the technical field required by the user. Taking student attendance as an example, in the achievement database, scientific and technological literature related to student attendance may include student attendance methods, student management, and teaching quality evaluation. If recommendations are simply made based on text similarity, the user's specific needs for a specific technical field may be ignored, resulting in the recommendation results being relevant but unable to accurately meet the user's query purpose. The present invention can more accurately adjust the results of text duplication checking to ensure that they are highly relevant to user needs by considering the user's interest in different technical fields. The corrected text duplication checking value can better reflect the true similarity between documents, which helps to improve the accuracy of the recommendation system and user experience.

[0028] A comprehensive management system for scientific and technological information, comprising: A data acquisition module is used to acquire the browsed data and unbrowsed data of the target object after retrieval. The browsed data is the result documents browsed after retrieval, the document browsing time and the document download record; the unbrowsed data is the result document set that has not been browsed after retrieval; A data analysis module is used to determine the target document according to the browsing data, obtain the relevant technical fields corresponding to the target document and mark them as y1, y2, ..., yn, calculate the interest of any of the relevant fields in turn and mark them as P1, P2, ..., Pn; extract the abstract text in the target document and perform preprocessing to obtain a comprehensive conclusion text; Select any pending document from the unbrowsed achievement document set, extract the abstract text in the pending document and perform preprocessing, perform text duplication calculation on the preprocessed abstract text and the comprehensive conclusion text, obtain the text duplication values ​​of all pending documents and mark them as C1, C2, ..., Cm; The result generation module is used to obtain the relevant technical field corresponding to any pending document, determine the interest of the pending document according to the relevant technical field, correct the text duplication value of the pending document according to the interest, and sort all the pending documents from large to small according to the corrected text duplication value and recommend pending documents to the target object according to the sorting result.

[0029] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for comprehensive management of scientific and technological information, characterized in that: The following steps are involved: S1, obtaining browsed data and unbrowsed data of the target object after retrieval, wherein the browsed data is the result documents browsed after retrieval, the document browsing time and the document download record; the unbrowsed data is the result document set that has not been browsed after retrieval; S2, determining the target document according to the browsing data, obtaining the relevant technical fields corresponding to the target document and marking them as y1, y2, ..., yn, calculating the interest of any relevant field in turn and marking them as P1, P2, ..., Pn; extracting the abstract text in the target document and preprocessing it to obtain a comprehensive conclusion text; Select any pending document from the unbrowsed achievement document set, extract the abstract text in the pending document and perform preprocessing, perform text duplication calculation on the preprocessed abstract text and the comprehensive conclusion text, obtain the text duplication values ​​of all pending documents and mark them as C1, C2, ..., Cm; S3, obtaining the relevant technical field corresponding to any pending document, and determining the interest of the pending document according to the relevant technical field, correcting the text duplication value of the pending document according to the interest, sorting all pending documents from large to small according to the corrected text duplication value, and recommending pending documents to the target object according to the sorting result.

2. A method for comprehensive management of scientific and technological information according to claim 1, characterized in that: Said S1 further includes: The literature database is screened according to the search terms entered by the target object to obtain the search data, which are then sorted according to the publication order of the literature, the relevance of the article title to the search terms, and the impact factor.

3. A method for comprehensive management of scientific and technological information according to claim 1, characterized in that: In S2, the process of determining the target document is as follows: Obtain the document browsing time t and document download record S of any browsed document. If S is greater than 0, calculate the recommendation score W of all browsed documents according to the calculation formula W=b*t+d*S, where b and d are preset coefficients and b is less than d; If S is equal to 0, the recommendation scores of all browsed documents are calculated according to the calculation formula W=b*t; if there is any browsed document whose recommendation score W is greater than or equal to the preset threshold, the browsed document is marked as the target text.

4. A method for comprehensive management of scientific and technological information according to claim 1, characterized in that: In S2, the calculation process of the interest degree of the technical field is: S11, obtaining the recommendation scores of all browsed documents and summing them up to obtain a total recommendation score value W', calculating the ratio of any browsed document to the recommendation score, performing weighted summation based on the browsed document recommendation score and the ratio, and obtaining the interest level X of any browsed document; S12, obtaining all relevant technical fields corresponding to any browsed document and defining the interest degree of any relevant technical field corresponding to the browsed document as the interest degree of the browsed document; obtaining the relevant technical fields corresponding to the target document, and calculating the interest degree of each relevant technical field of the target document, the specific calculation is as follows: ； Pm represents the interest level of the target document corresponding to the relevant technical field m, i represents the browsed document corresponding to the relevant technical field m, n is the number of browsed documents corresponding to the relevant technical field m, and Xi represents the interest level corresponding to the browsed document i.

5. A method for comprehensive management of scientific and technological information according to claim 1, characterized in that: In S2, the specific process of preprocessing is: Performing text analysis on the summary text to search whether a preset signal word exists in the summary text; if the preset signal word exists in the summary text, extracting sentence information after the preset signal word, and using the sentence information as conclusion information in the summary text; If there is no preset signal word in the summary text, sentence information of a preset part is extracted from the summary text, and the sentence information is used as the conclusion information in the summary text; a semantic relationship graph between conclusion texts is constructed using semantic network technology, and several conclusion texts are integrated into a comprehensive conclusion text based on the semantic relationship graph using NLP technology.

6. A method for comprehensive management of scientific and technological information according to claim 1, characterized in that: In S2, the calculation process of the duplicate check value is: The summary text and the comprehensive conclusion text are respectively converted into two groups of ASCII code values ​​in character order; the characteristics of each character are represented by the ASCII code values ​​of the surrounding characters; The autoencoder is used to reduce the dimension of each character point; the Jaccard similarity coefficient is used to calculate the similarity between two texts after the dimension reduction. The calculation formula for the text duplication value is: ； Among them, A is the feature set of the comprehensive conclusion text, B is the feature set of the abstract text corresponding to any pending document, and J is the duplicate check value.

7. A method for comprehensive management of scientific and technological information according to claim 1, characterized in that: In S3, the specific process of correction is: Obtain all relevant technical fields corresponding to the pending document and the target document, and then obtain all overlapping technical fields and mark them as Y1, Y2, ..., YN, obtain the interest level of any overlapping technical field and mark it as P1, P2, ..., PN, and calculate according to the formula The corrected text duplication check value C' is calculated, where α is a preset correction coefficient and Pi is the interest level of any overlapping technical field.

8. A comprehensive management system for scientific and technological information, characterized in that: include: A data acquisition module is used to acquire the browsed data and unbrowsed data of the target object after retrieval. The browsed data is the result documents browsed after retrieval, the document browsing time and the document download record; the unbrowsed data is the result document set that has not been browsed after retrieval; A data analysis module is used to determine the target document according to the browsing data, obtain the relevant technical fields corresponding to the target document and mark them as y1, y2, ..., yn, calculate the interest of any relevant field in turn and mark them as P1, P2, ..., Pn; extract the abstract text in the target document and perform preprocessing to obtain a comprehensive conclusion text; Select any pending document from the unbrowsed achievement document set, extract the abstract text in the pending document and perform preprocessing, perform text duplication calculation on the preprocessed abstract text and the comprehensive conclusion text, obtain the text duplication values ​​of all pending documents and mark them as C1, C2, ..., Cm; The result generation module is used to obtain the relevant technical field corresponding to any pending document, determine the interest of the pending document according to the relevant technical field, correct the text duplication value of the pending document according to the interest, and sort all the pending documents from large to small according to the corrected text duplication value and recommend pending documents to the target object according to the sorting result.

Images