Scientific achievement management method and system with pushing function

Abstract

The invention provides a scientific research achievement management method and system with a pushing function, and the method comprises the steps: obtaining achievement data of different data sources, and carrying out the fusion of the achievement data; marking corresponding category labels for the fused achievement data, and performing parallel classification on the achievement data of different category labels by a distributed decision tree classification algorithm to obtain a plurality of hierarchical classification results of the achievement data of different category labels; analyzing the time sequence of the plurality of hierarchical classification results from the achievement data to obtain a change rule of the achievement data; based on an association rule mining algorithm, mining association relationships among the plurality of hierarchical classification results, and based on the plurality of hierarchical classification results, the change rule of the result data and the mined association relationships among the plurality of hierarchical classification results, forming an image-text report for visual display; and obtaining a user demand, extracting key features of the user demand, carrying out association matching on an image-text report of a corresponding theme based on the key features, and carrying out personalized pushing to the user.

Description

Technical Field

[0001] The present invention relates to the technical field of scientific research achievement management, and in particular to a scientific research achievement management method and system with a push function. Background Art

[0002] For a long time, the research work of scientific research departments has been more "plan-oriented and less productive", and the management and transformation of project results lack top-level considerations and rigid norms, as well as efficient and intelligent technical means to support them. A large number of project results are buried in filing cabinets due to improper management and cannot play their due role. Therefore, it is difficult for management departments at all levels to make scientific decisions based on the actual value of the results. At the same time, researchers are unable to effectively identify and quickly extract useful information that meets their needs from a large amount of complex project materials, thereby providing support for the transformation and application promotion of scientific research results.

[0003] In the context of today's technological progress, the promotion and application of scientific research results has become crucial, which is also of great significance to the improvement of the work efficiency of scientific research departments. To this end, people push scientific research results to researchers through active push. However, some results push systems have obvious problems of inaccurate push, lack of in-depth understanding and mining of scientific research results, and cannot provide users with personalized recommendations, thus failing to meet the diversified needs of users.

[0004] It can be seen that there is currently a lack of a strong technical guarantee to effectively solve the above problems. Therefore, it is imperative to develop a results push and management system that can achieve effective management of the entire process of scientific research results from output to application promotion and accurate push of scientific research results. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a scientific research achievement management method and system with a push function to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of the present invention provides a scientific research achievement management method with a push function, the method comprising the following steps:

[0007] Obtain user needs and achievement data from different data sources, and store the achievement data in the achievement database;

[0008] Extract, convert and load the achievement data, and integrate the achievement data from different data sources in the achievement database;

[0009] Based on the preset labeling strategy, the corresponding category labels are marked on the fused achievement data, and the achievement data marked with different category labels are classified in parallel at multiple levels based on the category labels through the distributed decision tree classification algorithm to obtain multiple multi-level classification results of the achievement data with different category labels;

[0010] Extracting a time series of multiple multi-level classification results of achievement data with different category labels from achievement data marked with different category labels based on multiple multi-level classification results, analyzing the time series of multiple multi-level classification results of achievement data with different category labels, and obtaining a change pattern of the achievement data;

[0011] Based on the association rule mining algorithm, the association relationship between multiple multi-level classification results of the achievement data with different category labels is mined, and charts and reports for visual display are formed based on the multiple multi-level classification results of the achievement data with different category labels, the change law of the achievement data and the association relationship between the multiple multi-level classification results of the achievement data with different category labels, and the charts and reports are stored in the achievement database according to the subject classification;

[0012] Obtain user needs, extract key features from user needs, associate and match charts and reports on corresponding topics in the results database based on key features, and push them to users in a personalized manner.

[0013] In some embodiments of the present invention, the data source includes part or all of scientific research reports, project completion reports, papers, journals and patent documents; the achievement data includes part or all of technical information, product information, technology maturity, technology advancement, product mass production, technology research methods, technology application fields, technology research results, technology research and development cycle and technology research unit information in various technical fields.

[0014] In some embodiments of the present invention, a distributed decision tree classification algorithm is used to perform parallel multi-level classification on the achievement data marked with different category labels based on the category labels, and multiple multi-level classification results of the achievement data with different category labels are obtained, including:

[0015] Divide the achievement data marked with different category labels into multiple data sets according to the category labels, including specific technical field data sets, technical information data sets, product information data sets, technical maturity data sets, technical advancement data sets, product mass production data sets, technical research method data sets, technical application field data sets, technical research results data sets, technical R&D cycle data sets, and part or all of the technical research unit information data sets;

[0016] Based on the information gain criterion, multiple features with high correlation with the classification results are selected from each data set to form a feature set, and the feature with the largest information gain in the feature set is selected as the root node. For the root node, the feature with the largest information gain in the feature set is selected as multiple first-level child nodes, and for each first-level child node, the feature with the largest information gain in the feature set is selected as the second-level child node, until there is no feature to be selected in the feature set, and the corresponding multiple decision trees are generated;

[0017] Actively prune some branches of each decision tree to obtain multiple decision trees to represent multiple multi-level classification results of outcome data with different category labels.

[0018] In some embodiments of the present invention, the association relationship includes part or all of the following: the association between technical information in the same or different technical fields, between product information, and between technical information and product information, as well as the association between technical information, product information and technology maturity, technological advancement, product mass production, technology research methods, technology application fields, technology research results, technology R&D cycle, and technology research unit information.

[0019] In some embodiments of the present invention, mining the association relationship between multiple multi-level classification results of achievement data with different category labels based on an association rule mining algorithm includes:

[0020] Scan multiple multi-level classification results of achievement data with different category labels, search for a set of candidate achievement data and the corresponding support, prune and remove the achievement data set with lower support and confidence, and obtain a set of frequent achievement data;

[0021] Connect one item set of frequent result data to obtain two item sets of candidate result data, prune and remove the two item sets of candidate result data with lower support and confidence, and obtain the two item sets of frequent result data. Repeat this process for multiple iterations until the k+1 item set of frequent result data cannot be found, and output the corresponding set of k item sets of frequent result data to achieve the association between multiple multi-level classification results of result data with different category labels.

[0022] In some embodiments of the present invention, the key features in the user's needs are extracted, and the charts and reports of the corresponding topics in the matching result database are associated with the key features, including:

[0023] Divide user needs into word sequences;

[0024] Inputting a word sequence into a pre-trained feature extraction model so that the feature extraction model extracts keyword features from the word sequence required by the user, wherein the feature extraction model includes a BERT model and a conditional random field model, wherein the BERT model is used to extract keyword features from the input word sequence, and the conditional random field model is used to output keyword features and the probability of correct extraction of keyword features based on the extracted key features;

[0025] Select some keyword features with a higher probability of correct extraction from the output keyword features, and associate the charts and reports of corresponding topics in the matching result database based on the some keyword features with a higher probability of correct extraction.

[0026] In some embodiments of the present invention, the method further comprises:

[0027] Acquire historical user demand and historical access behavior data of multiple users, and extract key features of the historical user demand and historical access behavior data, wherein the access behavior data includes historical browsing, downloading and citing behavior data;

[0028] Obtain the access behavior data of the current user, extract the key features of the access behavior data of the current user, measure the similarity of the user needs and key features of the access behavior data between the current user and multiple users through the Euclidean distance algorithm, and obtain users with similar scientific research interests to the current user;

[0029] Actively push charts and reports of corresponding results data obtained by the current user to users with similar scientific research interests as the current user.

[0030] In some embodiments of the present invention, before storing the achievement data in the achievement database, the method further includes: performing a cleaning process on the achievement data.

[0031] In some embodiments of the present invention, cleaning the achievement data and storing the cleaned achievement data in an achievement database includes:

[0032] Remove redundancy and noise, remove duplication, repair erroneous data, convert formats and integrate the results data;

[0033] Data storage rules are defined based on the characteristics of the achievement data and the structure of the achievement database, and the integrated achievement data are stored in the achievement database based on the data storage rules. The data storage rules include field mapping relationships and data verification rules.

[0034] Another aspect of the present invention provides a scientific research achievement management system with a push function, the system comprising: a processor, a memory and computer instructions stored in the memory, the processor is used to execute the computer instructions, and when the computer instructions are executed, the system implements the steps of the aforementioned method.

[0035] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the aforementioned method when executed by a processor.

[0036] Another aspect of the present invention provides a computer program product, comprising computer instructions, which implement the steps of the above method when executed by a processor.

[0037] The present invention provides a scientific research achievement management method and system with a push function, which can simultaneously realize the refined management and precise push of scientific research achievement data, cover the whole process management from achievement output to achievement application, support the deep mining of achievement data value, and enable scientific researchers to update, manage, apply and analyze various scientific research achievements more conveniently and quickly, thereby promoting the rapid transformation and application of scientific research achievements.

[0038] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention may be achieved and obtained by the structures specifically indicated in the specification and the accompanying drawings.

[0039] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention.

[0041] Figure 1 A schematic diagram of a flow chart of a scientific research achievement management method with a push function in one embodiment of the present invention;

[0042] Figure 2 It is a structural diagram of a scientific research achievement management system with a push function in one embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0044] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0045] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0046] It should also be noted that, unless otherwise specified, the term “connection” herein may refer not only to a direct connection but also to an indirect connection involving an intermediate.

[0047] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0048] The scientific research achievement management method and system with a push function provided by the embodiment of the present invention can, on the one hand, gather the achievement data of various professional and technical fields of scientific research units over the years, create an achievement data management platform, and further promote the transformation and application of achievements through the clustering effect of achievements. On the other hand, it uses advanced technologies such as artificial intelligence to create a service center for achievement transformation and application, and through a centralized achievement management model, matches the achievement management mechanism, and builds a benign achievement transformation service ecosystem, providing strong technical guarantees for effectively solving current problems.

[0049] Figure 1 FIG. 1 is a flow chart of a method for managing scientific research achievements with a push function in one embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0050] Step S110, obtaining achievement data from different data sources, and storing the achievement data in an achievement database.

[0051] Specifically, data sources include part or all of the content in scientific research reports, project completion reports, papers, journals and patent documents. Achievement data include part or all of the content in various technical fields, such as technical information, product information, technical maturity, technical advancement, product mass production, technical research methods, technical application fields, technical research results, technical R&D cycle and information on technical research units. Data sources and achievement data are not limited to the above listed contents.

[0052] In some embodiments, the method obtains user needs and achievement data from different data sources by importing documents such as scientific research reports, project completion reports, papers, journals and patent documents and user demand documents into a scientific research achievement management system in a pre-set data format, and storing them in a temporary database after review by the scientific research achievement management system to facilitate subsequent further processing of these data, such as cleaning various types of achievement data from different data sources stored in the temporary database, and extracting key features of user needs from user demand documents.

[0053] Step S120, extracting, converting and loading the achievement data, and integrating the achievement data from different data sources in the achievement database.

[0054] Through this step, the data extraction, conversion and loading rules are defined according to whether the achievement data represents a specific technical field or a specific product under a certain technical field, etc., so that various types of achievement data scattered in different data source files can be fused, aggregated or integrated, that is, all relevant content belonging to the A technical field can be fused, and the same approach is adopted for other specific categories.

[0055] Step S130, based on the preset labeling strategy, the corresponding category labels are marked on the fused results data, and the results data marked with different category labels are classified in parallel based on the category labels through a distributed decision tree classification algorithm to obtain multiple multi-level classification results of the results data with different category labels.

[0056] In this step, the preset labeling strategy specifically refers to identifying different categories of achievement data according to the content represented by the achievement data, and marking the corresponding category labels. Category labels include A technical field, B technical field, A1 subdivided technical field, B1 subdivided technical field, A product, A technology, principle stage, concept stage (technology maturity), etc. Specifically, the process of marking the corresponding category labels for the fused achievement data based on the preset labeling strategy can capture and identify the semantic information in the fused achievement data through supervised learning, semi-supervised learning or unsupervised learning, and output the category label result of the category to which the achievement data belongs.

[0057] In one embodiment of the present invention, a distributed decision tree classification algorithm is used to perform parallel multi-level classification on achievement data marked with different category labels based on category labels to obtain multiple multi-level classification results of achievement data with different category labels, including the following steps:

[0058] Step S132, dividing the achievement data marked with different category labels into multiple data sets according to the category labels, the data sets include specific technical field data sets, technical information data sets, product information data sets, technology maturity data sets, technology advancement data sets, product mass production data sets, technology research method data sets, technology application field data sets, technology research result data sets, technology R&D cycle data sets, and technology research unit information data sets, etc.;

[0059] Step S133, based on the information gain criterion, multiple features with high correlation with the classification results are selected from each data set to form a feature set, the feature with the largest information gain in the feature set is selected as the root node, the feature with the largest information gain in the feature set is selected from the root node as multiple first-level child nodes, and the feature with the largest information gain in the feature set is selected from each first-level child node as a second-level child node, until there is no feature to be selected in the feature set, and a corresponding plurality of decision trees are generated;

[0060] Step S134, actively pruning some branches of each decision tree to obtain multiple final decision trees to represent multiple multi-level classification results of the outcome data with different category labels.

[0061] The above steps can realize parallel classification operations through a distributed decision tree classification algorithm, and finally form multiple multi-level classification results such as A technology field and its sub-technology field, B technology field and its sub-technology field, A product and its sub-classification, etc.

[0062] Step S140, based on multiple multi-level classification results, extract the time series of multiple multi-level classification results of the achievement data with different category labels from the achievement data marked with different category labels, analyze the time series of multiple multi-level classification results of the achievement data with different category labels, and obtain the change pattern of the achievement data.

[0063] This step adds a time dimension to the result data, user demand data, and problem reporting data collected over the years on the basis of text analysis. By extracting the corresponding time series from the result data and performing time series analysis, the development and change laws of scientific research results are discovered and revealed, and the analysis results are used for knowledge and information prediction. For example, in the collection and demand of results classification in recent years, the change of results classification is analyzed by time series, such as data situation analysis, and it can be helpful to predict new technologies and new results that will appear in the short or long term in the next few years. Commonly used time series analysis includes white noise series, autoregressive model, moving average model, autoregressive moving average model and ARIMA model.

[0064] Step S150, based on the association rule mining algorithm, the association relationship between multiple multi-level classification results of the achievement data with different category labels is mined, and charts and reports for visual display are formed based on the multiple multi-level classification results of the achievement data with different category labels, the change law of the achievement data and the mined multiple multi-level classification results of the achievement data with different category labels. The charts and reports are stored in the achievement database according to thematic classification.

[0065] This step can realize the mining and analysis of the following association relationships, which specifically include part or all of the following contents: the relationship between technical information in the same or different technical fields, between product information, and between technical information and product information, as well as the relationship between technical information, product information and technology maturity, technological advancement, product mass production, technology research methods, technology application fields, technology research results, technology R&D cycle and technology research unit information. In this step, visualized charts and reports include but are not limited to text visualization, network diagram visualization, spatiotemporal data visualization, and multidimensional data visualization. Specific presentation forms include various data chart types, such as column charts, bar charts, line charts, pie charts, area charts, ring charts, radar charts, bubble charts, interactive dashboards, scatter plots, heat maps, statistical charts, multi-dimensional composite charts, correlation analysis charts, as well as special-shaped charts such as water polo charts and rose charts; vivid human-computer interaction presentation, such as support for data granularity adjustment, hierarchical roll-up, drill-down, slicing, zooming, mouse hotspots, multi-chart linkage, multi-perspective switching and other types of interactive presentation forms; and diverse data presentation dimensions: support for customizing the time and space presentation dimensions of achievement data, and multi-dimensional visualization presentation based on a hierarchical structure to achieve multi-mode display of instantaneous data, hourly data, and cumulative data in different regions and other spatial relationships.

[0066] In one embodiment of the present invention, the association relationship between multiple multi-level classification results of achievement data with different category labels is mined based on an association rule mining algorithm, which specifically includes the following steps:

[0067] Step S151, scanning multiple multi-level classification results of achievement data with different category labels, searching for a candidate achievement data item set and a corresponding support, pruning and removing achievement data item sets with lower support and confidence, and obtaining a frequent achievement data item set;

[0068] Step S152, connect one item set of frequent achievement data to obtain two item sets of candidate achievement data, prune the two item sets of candidate achievement data with lower support and confidence, and obtain the two item sets of frequent achievement data, and so on, through multiple iterations, until the k+1 item set of frequent achievement data cannot be found, output the corresponding set of k item sets of frequent achievement data, and realize the association between multiple multi-level classification results of achievement data with different category labels.

[0069] The above steps use the above association rule mining algorithm to perform correlation mining on multiple categories of achievement data. While searching for an achievement data set in a certain technical field, it is possible to frequently query an achievement data set in another technical field or technology, product, etc., and then the two technical fields or the technical field and technology, product can be associated, so as to analyze and mine the achievements and the technical fields, technologies, application fields, research units, technology maturity, research cycle and other information that match the achievements.

[0070] Step S160, obtaining user needs, extracting key features in the user needs, associating and matching charts and reports of corresponding topics in the result database based on the key features, and pushing them to the user in a personalized manner.

[0071] Specifically, user needs may be to obtain relevant technical information involved in the field of aviation science and technology, to obtain technical information related to drones, or to obtain results data related to scientific research work.

[0072] In one embodiment of the present invention, the key features in the user's needs are extracted, and the charts and reports of the corresponding topics in the matching result database are associated with the key features, which specifically includes the following steps:

[0073] Step S161, dividing the user's needs into word sequences;

[0074] Step S162, inputting the word sequence into a pre-trained feature extraction model, so that the feature extraction model extracts key word features from the word sequence required by the user, the feature extraction model includes a BERT model and a conditional random field model, the BERT model is used to extract key word features from the input word sequence, and the conditional random field model is used to output key word features and the correct extraction probability of the key word features according to the extracted key features;

[0075] Step S163, selecting some keyword features with a higher probability of correct extraction from the output keyword features, and associating the charts and reports of the corresponding topics in the matching result database according to the some keyword features with a higher probability of correct extraction.

[0076] In one embodiment of the present invention, the method further comprises the following steps:

[0077] Step S170: Obtain the historical user requirements and historical access behavior data of multiple users, and extract the key features of the historical user requirements and historical access behavior data. The access behavior data includes behavior data such as historical browsing duration, click-through rate, sharing times, downloads, and citations.

[0078] Step S180: Obtain the access behavior data of the current user, extract the key features of the access behavior data of the current user, and measure the similarity of the key features of the user requirements and access behavior data between the current user and multiple users through the Euclidean distance algorithm to obtain users with similar scientific research interests to the current user.

[0079] Step S190: Actively push the charts and reports of the corresponding achievement data obtained by the current user to the users with similar scientific research interests to the current user.

[0080] Through the above steps, the scientific research achievement management method with a push function can achieve precise active recommendation and passive push of scientific research achievement data.

[0081] In an embodiment of the present invention, before storing the achievement data in the achievement database, the method further includes: performing cleaning processing on the achievement data.

[0082] Specifically, performing cleaning processing on the achievement data and storing the cleaned achievement data in the achievement database specifically includes the following steps:

[0083] Step S112: Perform operations such as redundancy and noise removal, duplicate removal, error data repair, format conversion, and integration on the achievement data.

[0084] Specifically, remove noise such as HTML tags, special characters, and extra spaces from the specific text content such as technical information, product information, technology maturity, technology advancement, product mass production, technology research methods, technology application fields, technology research results, technology R & D cycles, and technology undertaking unit information in various technical fields in scientific research reports, project finalization reports, papers, periodicals, and patent documents, and remove redundancy such as some redundant and meaningless words such as "de" (的) and "le" (了), remove duplicate content information, and repair sentences that do not conform to common language usage habits, unsmooth sentences, and damaged error data in the text according to the context, context information, etc. Then, uniformly convert the format of the above data into a specific format convenient for storage in the achievement database, and then integrate and aggregate the achievement data after all the above operations.

[0085] Step S113, defining data storage rules based on the characteristics of the achievement data and the structure of the achievement database, and storing the integrated achievement data in the achievement database based on the data storage rules, wherein the data storage rules include field mapping relationships and data verification rules.

[0086] Specifically, the field mapping relationship can be a unified description of all relevant articles, paragraphs, sentences, vocabulary, etc. belonging to the technical field, technology, product, technology maturity, technology advancement, product mass production, technology research methods, technology application fields, technology research results, technology research and development cycle or technology research unit information, etc., converting the original description into a summary of the unified description as the field mapping relationship, and storing the corresponding content of the achievement data in the achievement database according to the mapped fields. The data verification rules include MD5 value integrity verification and cosine algorithm repeatability check of the achievement data to be stored to ensure the integrity, accuracy and consistency of the data.

[0087] Corresponding to the above method, an embodiment of the present invention also provides a scientific research achievement management system with a push function, the system includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, when the computer instructions are executed by the processor, the system implements the steps of the above method.

[0088] like Figure 2As shown, the system includes the following modules: user demand and problem reporting module, achievement data collection module, achievement data governance module, achievement data operation management module, achievement data display module, achievement data push module and achievement database, forming a complete process of scientific research achievement data collection, compilation and storage, intelligent analysis, docking application, transformation and promotion. Specifically, the achievement data collection module is used to obtain achievement data from different data sources; the achievement data governance module is used to clean the achievement data, as well as to extract, convert and load the achievement data, and merge the achievement data from different data sources in the achievement database; the achievement database is used to store the cleaned achievement data; the achievement data operation management module is used to mark the corresponding category labels for the fused achievement data based on the preset labeling strategy, and perform parallel multi-level classification of the achievement data marked with different category labels based on the category labels through the distributed decision tree classification algorithm, and obtain multiple multi-level classification results of the achievement data with different category labels; based on the multiple multi-level classification results, the time series of the multiple multi-level classification results of the achievement data with different category labels are extracted from the achievement data marked with different category labels, and the achievements with different category labels are classified. The time series of multiple multi-level classification results of the data are analyzed to obtain the changing rules of the achievement data; and the association relationship between the multiple multi-level classification results of the achievement data with different category labels is mined based on the association rule mining algorithm; the achievement data display module is used to form charts and reports for visual display based on the multiple multi-level classification results of the achievement data with different category labels, the changing rules of the achievement data and the mined multiple multi-level classification results of the achievement data with different category labels; the achievement database is also used to store the charts and reports for visual display according to thematic classification; the user demand and problem reporting module is used to obtain user needs and related problems reported; the achievement data push module is used to extract key features in user needs, and match the charts and reports of corresponding topics in the achievement database based on the key features and push them to users in a personalized manner.

[0089] In some embodiments, the system also includes a temporary database for temporarily storing user needs and related issues obtained by the user needs and problem reporting module, so that the results data push module can further process the needs and issues, and temporarily storing the results data from different data sources obtained in the results data collection module, so that the results data governance module can clean the temporarily stored results data.

[0090] Generally speaking, the user demand and problem reporting module and the achievement data collection module are both connected to the temporary database, the temporary database is also connected to the achievement data governance module and the achievement data push module, the achievement data governance module is also connected to the achievement database and the achievement data operation management module, the achievement data operation management module is also connected to the achievement data display module, the achievement data display module is also connected to the achievement database, and the achievement database is also connected to the achievement data push module.

[0091] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0092] An embodiment of the present invention further provides a computer program product, comprising computer instructions, which implement the steps of the aforementioned method when executed by a processor.

[0093] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0094] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0095] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0096] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A scientific research achievement management method with a push function, characterized in that: The method comprises: Obtain achievement data from different data sources and store the achievement data in the achievement database; Extract, convert and load the achievement data, and integrate the achievement data from different data sources in the achievement database; Based on the preset labeling strategy, the corresponding category labels are marked on the fused achievement data, and the achievement data marked with different category labels are classified in parallel at multiple levels based on the category labels through the distributed decision tree classification algorithm to obtain multiple multi-level classification results of the achievement data with different category labels; Extracting a time series of multiple multi-level classification results of achievement data with different category labels from achievement data marked with different category labels based on multiple multi-level classification results, analyzing the time series of multiple multi-level classification results of achievement data with different category labels, and obtaining a change pattern of the achievement data; Based on the association rule mining algorithm, the association relationship between multiple multi-level classification results of the achievement data with different category labels is mined, and charts and reports for visual display are formed based on the multiple multi-level classification results of the achievement data with different category labels, the change law of the achievement data and the association relationship between the multiple multi-level classification results of the achievement data with different category labels, and the charts and reports are stored in the achievement database according to the subject classification; Obtain user needs, extract key features from user needs, associate and match charts and reports on corresponding topics in the results database based on key features, and push them to users in a personalized manner.

2. The method according to claim 1, characterized in that The data sources include part or all of scientific research reports, project completion reports, papers, journals and patent documents; the achievement data include part or all of technical information, product information, technology maturity, technology advancement, product mass production, technology research methods, technology application fields, technology research results, technology R&D cycle and technology research unit information in various technical fields.

3. The method according to claim 2, characterized in that Through the distributed decision tree classification algorithm, the achievement data marked with different category labels are classified in parallel at multiple levels based on the category labels, and multiple multi-level classification results of the achievement data with different category labels are obtained, including: Divide the achievement data marked with different category labels into multiple data sets according to the category labels, including specific technical field data sets, technical information data sets, product information data sets, technical maturity data sets, technical advancement data sets, product mass production data sets, technical research method data sets, technical application field data sets, technical research results data sets, technical R&D cycle data sets, and part or all of the technical research unit information data sets; Based on the information gain criterion, multiple features with high correlation with the classification results are selected from each data set to form a feature set, and the feature with the largest information gain in the feature set is selected as the root node. For the root node, the feature with the largest information gain in the feature set is selected as multiple first-level child nodes, and for each first-level child node, the feature with the largest information gain in the feature set is selected as the second-level child node, until there is no feature to be selected in the feature set, and the corresponding multiple decision trees are generated; Actively prune some branches of each decision tree to obtain multiple decision trees to represent multiple multi-level classification results of outcome data with different category labels.

4. The method according to claim 2, characterized in that: The said association relationship includes part or all of the following contents: the association between technical information in the same or different technical fields, between product information, between technical information and product information, and the association between technical information, product information and technology maturity, technological advancement, product mass production, technology research methods, technology application fields, technology research results, technology R&D cycle and technology research unit information.

5. The method according to claim 4, characterized in that Based on the association rule mining algorithm, the association relationship between multiple multi-level classification results of the achievement data with different category labels is mined, including: Scan multiple multi-level classification results of achievement data with different category labels, search for a set of candidate achievement data and the corresponding support, prune and remove the achievement data set with lower support and confidence, and obtain a set of frequent achievement data; Connect one item set of frequent achievement data to obtain two item sets of candidate achievement data. Prune and remove the two item sets of candidate achievement data with lower support and confidence to obtain the two item sets of frequent achievement data. Through multiple iterations, until the k+1 item set of frequent achievement data cannot be found, the corresponding set of k item sets of frequent achievement data is output to realize the association between multiple multi-level classification results of achievement data with different category labels.

6. The method according to claim 1, characterized in that Extract key features from user needs, and generate charts and reports on corresponding topics in the database based on key features, including: Divide user needs into word sequences; Inputting a word sequence into a pre-trained feature extraction model so that the feature extraction model extracts keyword features from the word sequence required by the user, wherein the feature extraction model includes a BERT model and a conditional random field model, wherein the BERT model is used to extract keyword features from the input word sequence, and the conditional random field model is used to output keyword features and the probability of correct extraction of keyword features based on the extracted key features; Select some keyword features with a higher probability of correct extraction from the output keyword features, and associate the charts and reports of corresponding topics in the matching result database based on the some keyword features with a higher probability of correct extraction.

7. The method according to claim 1, characterized in that The method further comprises: Acquire historical user demand and historical access behavior data of multiple users, and extract key features of the historical user demand and historical access behavior data, wherein the access behavior data includes historical browsing, downloading and citing behavior data; Obtain the access behavior data of the current user, extract the key features of the access behavior data of the current user, measure the similarity of the user needs and key features of the access behavior data between the current user and multiple users through the Euclidean distance algorithm, and obtain users with similar scientific research interests to the current user; Actively push charts and reports of corresponding results data obtained by the current user to users with similar scientific research interests as the current user.

8. A scientific research achievement management system with a push function, comprising a processor, a memory and computer instructions stored in the memory, characterized in that: The processor is used to execute the computer instructions, and when the computer instructions are executed, the system implements the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

Images