CN112765441B - Enterprise policy information multiple dynamic intelligent matching recommendation method for digital government affairs - Google Patents
Enterprise policy information multiple dynamic intelligent matching recommendation method for digital government affairs Download PDFInfo
- Publication number
- CN112765441B CN112765441B CN202110370716.0A CN202110370716A CN112765441B CN 112765441 B CN112765441 B CN 112765441B CN 202110370716 A CN202110370716 A CN 202110370716A CN 112765441 B CN112765441 B CN 112765441B
- Authority
- CN
- China
- Prior art keywords
- policy
- enterprise
- data
- matching
- label
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000004458 analytical method Methods 0.000 claims abstract description 34
- 230000009193 crawling Effects 0.000 claims abstract description 4
- 238000011534 incubation Methods 0.000 claims description 29
- 238000003745 diagnosis Methods 0.000 claims description 21
- 238000012216 screening Methods 0.000 claims description 11
- 238000011161 development Methods 0.000 claims description 9
- 230000006399 behavior Effects 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 12
- 238000011160 research Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Human Resources & Organizations (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a multiple dynamic intelligent matching recommendation method for enterprise policy information of digital government affairs, which comprises the following steps: regularly crawling multiple groups of policy data from a webpage, performing text semantic analysis on the crawled data, performing modular disassembly and policy label setting on each group of policy data according to semantic analysis results, and storing the policy data in a policy database; enterprise data input by a user is received, enterprise label setting is respectively carried out on different types of content information in the enterprise data, and the different types of content information are stored in an enterprise database; and positioning the data category according to the set policy label and enterprise label, then performing multiple intelligent matching on enterprise data input by a user and multiple groups of policy data in the policy database, performing weight scoring on each round of matching result, and pushing a policy matching report to the user according to the final weight scoring result.
Description
Technical Field
The invention relates to the technical field of information processing, in particular to a multiple dynamic intelligent matching recommendation method for enterprise policy information of digital government affairs.
Background
With the rapid development of science and technology and economy of the country, in order to encourage innovation activities of enterprises, college scientific research institutions and the like, various departments of the country issue a plurality of policies every year to support and encourage innovation technology development of the enterprises and the college scientific research institutions. However, policy information is dispersed because the policy issuing departments, the issuing times, the issuing sites, and the like are different. Users want to know policy information in the industry field, often need to search a large amount of website information and analyze the acquired policy information to judge whether the users accord with declaration conditions, so that the complicated process occupies a large amount of time for the users such as enterprises. Users often need to analyze the policies one by one from massive policies to know the policies suitable for declaration, a large amount of labor time is occupied, and the efficiency is low. Enterprises cannot acquire policy information in time, cannot judge whether the enterprises accord with declaration conditions and finish a declaration process in time, cannot acquire support of government projects in time, and miss opportunities for acquiring policy assistance, so that the progress of national encouragement support is hindered.
Therefore, a method which can be connected with an enterprise and a policy bridge and can realize quick and accurate matching of enterprise data and policy data is urgently needed to be researched and developed, so that the policy can assist the enterprise development and protect the enterprise.
Disclosure of Invention
The object of the present invention is to solve at least one of the technical drawbacks mentioned.
Therefore, the invention aims to provide a method for recommending enterprise policy information for digital government affairs by multiple dynamic intelligent matching, which comprises the following steps:
step S1, crawling a plurality of groups of policy data from a webpage at regular time, performing text semantic analysis on the crawled data, performing modular disassembly and policy label setting on each group of policy data according to semantic analysis results, and storing the policy data in a policy database;
step S2, receiving enterprise data input by a user, respectively setting enterprise labels for different types of content information in the enterprise data, and storing the enterprise labels in an enterprise database; wherein the business data includes at least a business name;
step S3, positioning data types according to the set policy labels and enterprise labels, then performing multiple intelligent matching on enterprise data input by a user and multiple groups of policy data in the policy database, performing weight scoring on each round of matching results, and pushing a policy matching report to the user according to the final weight scoring result; wherein, the multiple intelligent matching comprises the following steps:
step S31, according to a preset policy text label system and an enterprise-policy layered incubation map, carrying out intelligent AI label setting on the enterprise data, wherein,
the policy text label system comprises: a universal map label, a structured layered incubation map label, a personalized special text label and an enterprise self-defined label, wherein,
(1) universal map label
Analyzing the enterprise data to obtain the development scale of the enterprise, matching corresponding universality policy content in the policy database according to the development scale of the enterprise, and setting a matched universality map label for the enterprise data;
(2) map label is hatched to structuralized layering
Presetting a structuralized layered incubation map, wherein the structuralized layered incubation map comprises: a hierarchical matching relationship map of policy categories and enterprise industry categories,
analyzing the enterprise data to obtain the industry category of the enterprise, matching corresponding policy data from the hierarchical data of the policy category according to the industry category of the enterprise, and setting a matched structured hierarchical incubation map label for the enterprise data; the structural layered incubation map is preset with the reportable policy content corresponding to each industry category;
(3) personalized special text label
(3.1) matching the enterprise data with the policy data, and respectively setting corresponding personalized special text labels for the enterprise data according to policies meeting conditions and policies not meeting conditions but having cultivation potential in matching results;
(3.2) carrying out background recording on the browsing behaviors of the users, analyzing the behavior habits of the enterprise users, positioning corresponding interest data sets, matching the interest data sets with the policy data, and setting matched personalized special text labels for the enterprise users;
(4) enterprise self-defined label
Receiving tag data actively input by an enterprise user, matching the tag data with the policy data, and setting a matched enterprise custom tag for the enterprise user;
step S32, searching matching policies in the policy database according to the universal map tag, the structured hierarchical incubation map tag, the enterprise custom tag and the personalized special text tag set in step S31, and performing weight scoring on each policy, including:
distributing different weight values for the universal map label, the structured layered incubation map label, the personalized special text label and the enterprise self-defined label;
searching all corresponding policy contents in the policy database according to the label category, and performing multiple rounds of automatic matching of enterprise data and keywords of policy data aiming at each policy, wherein the multiple rounds of automatic matching comprise the following steps: calculating the matching degree of each round, and performing weight scoring on the matching of the round according to the matching degree to obtain a final weight scoring result of the enterprise applying the policy; scoring is carried out according to the weighted value of the label category corresponding to the policy matching in the round, and then the scoring values of the policy matching results corresponding to all the label categories are accumulated to obtain the final scoring result of the policy;
and S33, obtaining the final weight scoring result of each policy declared by the enterprise in the mode of S32, screening out the policy names of which the final weight scoring result is higher than a threshold value, sequencing the policy names according to the final weight scoring result from high to low, generating a policy matching report and pushing the policy matching report to a user.
Further, in the step S1, the method includes the steps of:
step S11, performing text semantic analysis on the policy data to obtain a semantic analysis result; modularly disassembling the semantic analysis result according to a plurality of preset module topics, wherein the preset module topics comprise: policy making department, effective time of policy, industry to which policy belongs, region to which policy belongs, policy declaration condition, policy declaration flow and policy reward content;
step S12, setting corresponding module subject labels as the policy labels for the disassembled modular text paragraphs;
step S13, store each set of policy data into the policy database as a unit, wherein each set of policy data includes a modular text paragraph with a plurality of policy labels.
Further, in the step S2, the enterprise data is tagged in one of the following two forms:
(1) receiving enterprise description full text data input by the user, performing semantic analysis on the full text data to obtain different types of content information in the enterprise data, and setting enterprise tags;
(2) and providing a preset enterprise data template for the user, wherein each theme in the enterprise data template corresponds to an enterprise tag, and after the user finishes inputting according to the enterprise data template, automatically setting the enterprise tags for the input enterprise data.
Further, in step S32, in the multi-round automatic matching, three modes of forward matching, reverse condition, and reverse diagnosis are adopted:
(1) forward matching
Comparing the enterprise data with the industry to which the policy belongs, the region to which the policy belongs and the policy declaration conditions in the policy data respectively, and when a plurality of policies which meet the conditions exist, weighting and scoring the policies which meet the conditions as the reportable policies;
(2) reverse condition
Comparing the enterprise data with the industry to which the policy belongs, the region to which the policy belongs and the policy declaration condition in the policy data respectively, and when the enterprise data is judged not to accord with at least one hard necessary condition in the policy data, no longer pushing the policy data to the enterprise user; each item of policy data is preset with one or more hard necessary conditions for reporting the policy, and the policy is not allowed to be reported as long as one condition is not met;
(3) inverse diagnosis
And carrying out reverse analysis and diagnosis on the policy declaration of the enterprise according to the following two directions:
1) analyzing the declared projects and the enterprise basic data recorded in the enterprise data, and carrying out policy reverse diagnosis, wherein the policy reverse diagnosis comprises the following steps: diagnosing policies that are not declared but that meet the declared conditions;
2) matching the enterprise data with the policy data, and reversely diagnosing the policy which does not meet the declaration condition but has a difference with the declaration condition within a preset interval, wherein the policy comprises the following steps: policies that are not declared but have growing prospects are diagnosed.
Further, in the setting of the personalized special text label in step S31, automatically recording the system browsing record data of each user, analyzing to obtain an interface theme with high browsing frequency of the user, and generating a user interest data set; and when the user login is detected, automatically recommending policy information associated with the user interest data set to the user.
Further, in the step S32, corresponding weight values are set for policies of different tag types, wherein,
the weight value of the policy corresponding to the universal map label is less than that of the policy corresponding to the structured layered incubation map label is less than that of the enterprise self-defined label is less than that of the personalized special text label.
Further, the method also comprises the following steps: when a user registers, role setting options are provided for the user, and the options comprise: enterprise users, channel users and government agency users, and sets corresponding operation function interfaces and authorities for users with different roles;
and configuring a plurality of role authorities belonging to the enterprise user aiming at the same enterprise user, wherein each role authority distributes different policy matching viewing authority ranges for the same enterprise user according to the level of the login role.
Further, the policy database has an automatic updating function, the effective time label of the policy of each group of policy data is detected regularly, and when the judgment time is a past formula, the policy data is automatically rejected;
the enterprise database has an updating reminding function, regularly reminds the user to update enterprise data, and automatically generates a latest policy matching report for the user according to the updated enterprise data after detecting that the user is updated.
Further, when the enterprise information input by the user is only an enterprise name, acquiring current published basic data of the enterprise by taking the enterprise name as a keyword, and performing policy matching on the enterprise according to the current published basic data to generate a preliminary matching result;
and simultaneously, sending a non-public basic data acquisition request to the user to acquire non-public basic data related to policy declaration of the enterprise, combining the non-public basic data and the public basic data, performing secondary policy matching, and generating a final matching result.
Further, the policy matching report provides a matching policy list to the enterprise user, and further provides a selection control whether the enterprise user has a declared item, and the item which has been declared is not recommended to the enterprise user.
The enterprise policy information multiple dynamic intelligent matching recommendation method for digital government affairs, provided by the embodiment of the invention, has the following beneficial effects:
(1) text semantic analysis and modular disassembly are carried out on the crawled policy data, so that the content of the policy text can be quickly analyzed, and the core content and the key theme in the policy text are captured, so that the quick data matching can be realized by using the theme in the subsequent matching with the enterprise data, the matching of the full policy text is not needed, and the matching efficiency is greatly improved;
(2) the enterprise data is analyzed and the category labels are set, and the core content and the attribute themes in the enterprise data are captured, so that the quick data matching can be realized by using the themes in the subsequent matching with the policy data, the matching of the enterprise data full text is not needed, and the matching efficiency is greatly improved. In addition, the invention can provide enterprise information research tables for enterprise users, and the enterprise users can input the information according to the actual conditions of the enterprise users, thereby realizing the rapid analysis of enterprise data.
(3) An intelligent AI label system is established and divided into a universal map label, a structured hierarchical incubation map label, an enterprise self-defined label and an individualized special text label, the label hierarchical system can gradually realize the matching of enterprise data and policy data from shallow to deep, and a screening policy pool is gradually matched and reduced from multiple dimensions of universal policy content, enterprise industry categories, enterprise user self-defined labels, enterprise individualized data and the like, so that the final matching result has higher precision and better accords with the self-intention and the requirement of enterprise users.
(4) The method adopts three modes of forward matching, reverse condition and reverse diagnosis, except for pushing a matched policy for the user and shielding a unmatched policy, the method more innovatively provides a reverse diagnosis function for enterprise users, namely, the method matches the current policy which is not met but has a cultivation prospect of the enterprise according to the current condition of the enterprise, thereby providing long-term policy planning for the user and providing future cultivation direction and cultivation progress for the enterprise.
(5) The method and the system realize automatic collection of policy information on each website and classification and labeling of the policy information by adopting a semantic processing mode, realize efficient and accurate matching of enterprise data and policy data, solve the problems of low efficiency and poor accuracy of manual screening and matching of numerous and complicated policy information of the current enterprises or colleges, provide accurate intelligent policy matching and pushing service for users, provide policy guidance for the users, enable the users to timely know self-declared policies and provide policy cultivation guidance suggestions for the users.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart of a multiple dynamic intelligent matching recommendation method for enterprise policy information of digital government affairs according to an embodiment of the present invention;
fig. 2 is an architecture diagram of an intelligent AI tag architecture according to an embodiment of the present invention;
FIG. 3 is an exemplary diagram of policy information according to an embodiment of the present invention;
FIG. 4 is an interface diagram of a policy matching report according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
In order to solve the problems of low efficiency and poor analysis degree caused by the fact that enterprise-policy matching still stays in a manual matching stage in the prior art, the invention provides an enterprise policy information multiple dynamic intelligent matching recommendation method for digital government affairs.
As shown in fig. 1, the multiple dynamic intelligent matching recommendation method for enterprise policy information of digital government affairs according to the embodiment of the present invention includes the following steps:
and step S1, crawling multiple groups of policy data from the webpage at regular time, performing text semantic analysis on the crawled data, performing modular disassembly and policy label setting on each group of policy data according to semantic analysis results, and storing the policy data in a policy database.
Specifically, step S11, performing text semantic parsing on the policy data to obtain a semantic parsing result; modularly disassembling the semantic analysis result according to a plurality of preset module topics, wherein the preset module topics comprise: policy making department, effective time of policy, industry to which policy belongs, region to which policy belongs, policy declaration condition, policy declaration flow, policy reward content and the like. It should be noted that the module theme is not limited to the above example, and may also include other types of themes, which are set according to changes in policy content, and are not described herein again.
Taking the policy text in fig. 3 as an example, the policy text of the web page is crawled to perform text semantic analysis, semantic content of each paragraph of text is obtained through analysis, and then modular disassembly is performed. The specific disassembly is as follows:
and (4) policy making department: xxxxxxxx;
the policy effective time: by 2022 years;
the industry to which the policy belongs: new generation information technology, advanced manufacturing, new energy, new materials, unmanned aerial vehicles;
region to which policy belongs: tianjin;
policy reporting process: filling a' xxxxxxxxxx requirement characteristic set table (see annex) in 2021 according to actual requirements, and sending an electronic edition to xxxxx 6 days before 4 months in 2021;
policy declaration conditions: the method takes important economic value, important technical bottleneck, emerging industrial cultivation and supply chain safety as guidance, focuses on key core technical problems to be solved, explains research purposes, significance and necessity, clarifies main research contents, provides clear assessment targets and provides requirements for unveiling unit. The requirements for disclosing the control project are mainly provided by leading enterprises and scientific and technological enterprises in the Tianjin key field.
And step S12, respectively setting corresponding module subject labels as policy labels for the disassembled modular text paragraphs.
For example, referring to fig. 3, a label of "industry to which policy belongs" is set for the paragraph of "new generation information technology, advanced manufacturing, new energy, new material, unmanned aerial vehicle" according to the analysis result.
Step S13, store each set of policy data into a policy database, wherein each set of policy data includes a modular text paragraph with a plurality of policy labels.
Note that, although the policy text is divided into a plurality of sections, the policy text is stored in the database in units of each group of policies. That is, the division into paragraphs is for facilitating subsequent data matching, but when the policy is pushed to the user again, the policy is still pushed in a complete policy text.
And step S2, receiving the enterprise data input by the user, respectively setting enterprise labels for different types of content information in the enterprise data, and storing the enterprise labels in an enterprise database.
In this step, the enterprise information research form is provided to the user, and the user fills in the enterprise information according to the contents in the research form. For example, the enterprise data in the research sheet includes: enterprise name, enterprise type, registration address, main business, core technology, financial status, intellectual property status, declared project and the like.
It should be noted that the enterprise data in the present invention at least includes an enterprise name.
And when the enterprise information input by the user is only the enterprise name, acquiring the current published basic data of the enterprise by taking the enterprise name as a keyword, and performing policy matching on the enterprise according to the current published basic data to generate a preliminary matching result. And simultaneously, sending a non-public basic data acquisition request to the user to acquire non-public basic data related to policy declaration of the enterprise, combining the non-public basic data and the public basic data, performing secondary policy matching, and generating a final matching result.
Specifically, the user may only provide the name of the enterprise, and the backend system obtains other published basic data of the enterprise, such as the published information of the enterprise type, the registered address, the main business, the core technology, the financial status, the intellectual property status, and the like, from the existing enterprise database or network information according to the name of the enterprise, and then performs policy matching according to the data. But since some policy items require other data than the above-mentioned public data, such as financial data, etc., this kind of data is not public and needs to be provided by the user. And after the user provides the non-public data, combining the non-public basic data with the public basic data to perform secondary policy matching.
In this step, the enterprise data is tagged in one of the following two forms:
1. and receiving enterprise description full text data input by a user, performing semantic analysis on the full text data to obtain different types of content information in the enterprise data, and setting enterprise tags.
2. And providing a preset enterprise data template for a user, wherein each theme in the enterprise data template corresponds to an enterprise tag respectively, and after the user finishes inputting according to the enterprise data template, automatically setting the enterprise tags for the input enterprise data.
It should be noted that both the policy database and the enterprise database in the present invention can be updated automatically.
1. The policy database has an automatic updating function, the effective time label of the policy of each group of policy data is detected at regular time, and when the judgment time is a past formula, the policy data is automatically rejected.
Taking fig. 3 as an example, when the current date is detected to exceed 2021, 4/6, the policy is determined to be an expired policy, and the policy data is automatically removed from the policy database.
2. The enterprise database has an updating reminding function, regularly reminds the user to update enterprise data, and automatically generates a latest policy matching report for the user according to the updated enterprise data after detecting that the user is updated.
Specifically, as the enterprise itself is continuously developing, the enterprise data is also changing, and if policy matching is still performed on the initial enterprise data, the problem of inaccurate matching can be caused. Therefore, the invention sets a preset period of time, for example, 6 months, and automatically reminds the registered enterprise users in the system every 6 months whether the enterprise data needs to be updated or not. And if the fact that the user updates the enterprise data is detected, policy matching is carried out according to the updated enterprise data, and therefore the accuracy of policy matching is guaranteed.
Step S3, locating data categories according to the set policy labels and enterprise labels, performing multiple intelligent matching between enterprise data entered by the user and multiple sets of policy data in the policy database, performing weight scoring on each round of matching results, and pushing a policy matching report to the user according to the final weight scoring result, as shown in fig. 4.
And step S31, carrying out intelligent AI label setting on enterprise data according to a preset policy text label system and an enterprise-policy layered incubation map.
As shown in fig. 2, the policy text label system includes: the system comprises a universal map label, a structured layered incubation map label, an individualized special text label and an enterprise self-defined label. The following describes each type of label.
(1) Universal map label
And analyzing the enterprise data to obtain the development scale of the enterprise, matching corresponding universality policy content in the policy database according to the development scale of the enterprise, and setting a matched universality map label for the enterprise data.
Specifically, different policy projects have different declaration requirements on the enterprise scale, for example, declaration conditions of a type a policy require that the enterprise registration capital is higher than 100 ten thousand, declaration conditions of a type B policy require that the enterprise registration capital is between 100 ten thousand and 500 ten thousand, declaration conditions of a type C policy require that the enterprise registration capital is between 500 ten thousand and 1000 ten thousand, and the like.
Therefore, the enterprise data needs to be analyzed, and when the registered capital of the enterprise is judged to be between 100 million and 500 million, the policy with the declaration condition in the interval is screened to obtain a universal screening policy pool.
(2) Map label is hatched to structuralized layering
Predetermine the map is hatched in structuralized layering, the map is hatched in structuralized layering includes: and (4) a hierarchical matching relationship map of the policy category and the enterprise industry category.
Analyzing enterprise data to obtain an industry category of the enterprise, matching corresponding policy data from the hierarchical data of the policy category according to the industry category of the enterprise, and setting a matched structured hierarchical incubation map label for the enterprise data; and the structural layered incubation map is preset with the reportable policy content corresponding to each industry category.
Specifically, different policy terms have different declaration requirements for the enterprise industry category. For example, the declaration condition of the type a policy requires the enterprise industry category to be new energy and new materials, the declaration condition of the type B policy requires the enterprise industry category to be new internet technology, the declaration condition of the type C policy requires the enterprise industry category to be biological and new medical technology, and the like.
Therefore, it is necessary to analyze the enterprise data, and when the enterprise industry type is determined to be new energy, the policy information that "the industry to which the policy belongs" is new energy is screened from the obtained universal screening policy pool, so as to obtain a structured screening policy pool.
(3) Personalized special text label
And (3.1) matching the enterprise data with the policy data, and respectively setting corresponding personalized special text labels for the enterprise data according to the policies meeting the conditions and the policies not meeting the conditions but having cultivation potential in the matching result.
Specifically, the term "policy meeting the condition" means that if the enterprise data is the same as or similar to the keyword of the declaration condition in the policy data, it is determined that the declaration condition is met.
The policy which does not meet the conditions but has cultivation potential means that keywords of declaration conditions in the enterprise data policy data are different, but semantic analysis finds that the conditions can be met by cultivating the enterprise for a plurality of times.
For example: a policy claim requires that the core technician of the enterprise have 5 doctrines, 15 granted patents. But the enterprise only has 3 doctors and 2 issued patents at present. Through analysis, the enterprise finds that the intellectual property right can reach the declaration condition in the next year after the enterprise applies for the intellectual property right through the introduced talents.
And (3.2) carrying out background recording on the browsing behaviors of the enterprise users, analyzing the behavior habits of the enterprise users, positioning corresponding interest data sets, matching the interest data sets with policy data, and setting matched enterprise custom tags for the enterprise users.
In the embodiment of the invention, the system browsing record data of each user is automatically recorded, the interface theme with high browsing frequency of the user is obtained by analysis, and a user interest data set is generated; and when the user login is detected, automatically recommending policy information associated with the user interest data set to the user.
Specifically, the background system automatically records browsing record data of all registered users on the system in real time and then analyzes the browsing record data. For example, when it is recorded that a business user browses web contents identified by a high and new technology business for one week, the background automatically pushes policy contents related to the high and new technology business identification to the user after the user logs in.
(4) Enterprise self-defined label
And receiving label data actively input by the enterprise user, matching the label data with the policy data, and setting a matched personalized special text label for the enterprise user. Namely, the invention supports the user to input the tag data in a customized manner, and actively sets the tag for the user, for example, the self-management content of the enterprise is software Internet, but part of the technology belongs to the electromechanical control class, so that the 'electromechanical control' pair tag can be set by the user, and the subsequent policy matching with the relevant 'electromechanical control' is facilitated.
Step S32, according to the universal map label, the structured hierarchical incubation map label, the enterprise self-defined label and the personalized special text label set in the step S31, matching policies are searched in a policy database, and weight scoring is carried out on each policy, wherein the steps comprise:
(1) different weighted values are distributed to the universal map label, the structured layered incubation map label, the personalized special text label and the enterprise self-defined label.
In the embodiment of the invention, the weight value of the policy corresponding to the universal map label < the weight value of the policy corresponding to the structured layered incubation map label < the weight value of the enterprise self-defined label < the weight value of the personalized special text label.
Because the enterprise self-defined label and the personalized special text label are data based on the self-intention of the user, the label weight values of the enterprise self-defined label and the personalized special text label are higher than the 'universal atlas label' and the 'structured layered incubation atlas label' for policy category classification, and the final policy matching report can better meet the self-intention and the self-demand of the enterprise through weight value distribution, and the experience of the user is higher.
(2) Searching all corresponding policy contents in a policy database according to the label category, and performing multiple rounds of automatic matching of enterprise data and keywords of policy data aiming at each policy, wherein the multiple rounds of automatic matching comprise the following steps: and calculating the matching degree of each round, and performing weight scoring on the matching of the round according to the matching degree to obtain a final weight scoring result of the enterprise applying the policy.
And then, accumulating the scoring values of the policy matching results corresponding to all the label categories to obtain the final scoring result of the policy. Namely, aiming at the policy matching, scores of the universal map label, the structured layered incubation map label, the enterprise self-defined label and the personalized special text label after being respectively matched are accumulated to obtain a final score result.
Specifically, in step S32, in the multi-round automatic matching, three ways of forward matching, reverse condition, and reverse diagnosis are adopted:
(1) forward matching
And comparing the enterprise data with the industry to which the policy belongs, the region to which the policy belongs and the policy declaration conditions in the policy data respectively, and when a plurality of policies which meet the conditions exist, weighting and scoring the policies which meet the conditions as the reportable policies.
(2) Reverse condition
Comparing the enterprise data with the industry to which the policy belongs, the region to which the policy belongs and the policy declaration condition in the policy data respectively, and when the enterprise data is judged not to accord with at least one hard necessary condition in the policy data, not pushing the policy data to the enterprise user any more; each item of policy data is preset with one or more hard necessary conditions for reporting the policy, and the policy is not allowed to be reported as long as one condition is not met.
For example: a policy declaration requires that the revenue of an enterprise reach 1000 ten thousand. But the enterprise currently only collects 100 million. And analyzing and finding that the enterprise can obviously not mention the revenue to the declaration condition in a short period, automatically screening the policy, and not pushing the policy to the user so as to avoid wasting the user time.
(3) Inverse diagnosis
And carrying out reverse analysis and diagnosis on the policy declaration of the enterprise according to the following two directions:
1) analyzing the declared projects and the enterprise basic data recorded in the enterprise data, and carrying out policy reverse diagnosis, wherein the policy reverse diagnosis comprises the following steps: the policy that is not declared but meets the declared conditions is diagnosed.
For example, if the enterprise data meets the conditions identified by the reporting high and new technology enterprise, but there is no item in the reported items entered by the enterprise, it is determined that the policy belongs to "diagnosis is not reported but meets the reporting conditions", and the policy is pushed to the user.
2) Matching the enterprise data with the policy data, and reversely diagnosing the policy which does not meet the declaration condition but has a difference with the declaration condition within a preset interval, wherein the policy comprises the following steps: policies that are not declared but have growing prospects are diagnosed.
For example: a policy claim requires that the core technician of the enterprise have 5 doctrines, 15 granted patents. But the enterprise only has 3 doctors and 2 issued patents at present. Through analysis, the enterprise finds that the intellectual property right can reach the declaration condition in the next year after the enterprise applies for the intellectual property right through the introduced talents. Judging that the policy belongs to 'diagnosis is not declared but has cultivation prospect', and pushing the policy to the user. And when the policy content is pushed to the user, a culture declaration is further pushed to the user. Explicitly listed to the user in the suggestion: the difference from the reporting condition, the direction needing to be cultivated in the future and the planning, thereby providing the policy reporting planning service of the whole flow for the user.
And S33, obtaining the final weight scoring result of each policy declared by the enterprise in the mode of S32, screening out the policy names of which the final weight scoring result is higher than a threshold value, sequencing the policy names according to the final weight scoring result from high to low, generating a policy matching report and pushing the policy matching report to a user.
Referring to fig. 4, in the policy matching report, information such as policy content, policy category, policy entry unit, and the like is presented to the user according to the scoring result, and the user can select to view detailed content of the policy by clicking. Meanwhile, the policy matching report provides the user with the content of the policy matching analysis result, the policy matching suggestion and the like.
In the embodiment of the invention, the policy matching report provides the enterprise user with a matching policy list and further provides the enterprise user with a selection control for judging whether the declared item exists, and the enterprise user is not recommended any more for the declared item selected by the detected user.
In addition, when a user registers, role setting options are provided for the user, and the options comprise: enterprise users, channel users and government agency users, and sets corresponding operation function interfaces and permissions for users with different roles. And configuring a plurality of role authorities belonging to the enterprise user aiming at the same enterprise user, wherein each role authority distributes different policy matching viewing authority ranges for the same enterprise user according to the level of the login role. That is, the same enterprise user may associate multiple registration ids, each id requiring selection of a corresponding role, for example: corporate, financial, legal, technical, etc. The invention provides different authorities for different users. After different role ids are logged in, corresponding pushed pages are displayed for the different role ids according to the corresponding permissions, so that the accurate requirements of the different roles on policies are met, and the user experience is improved.
The enterprise policy information multiple dynamic intelligent matching recommendation method for digital government affairs, provided by the embodiment of the invention, has the following beneficial effects:
(1) text semantic analysis and modular disassembly are carried out on the crawled policy data, so that the content of the policy text can be quickly analyzed, and the core content and the key theme in the policy text are captured, so that the quick data matching can be realized by using the theme in the subsequent matching with the enterprise data, the matching of the full policy text is not needed, and the matching efficiency is greatly improved;
(2) the enterprise data is analyzed and the category labels are set, and the core content and the attribute themes in the enterprise data are captured, so that the quick data matching can be realized by using the themes in the subsequent matching with the policy data, the matching of the enterprise data full text is not needed, and the matching efficiency is greatly improved. In addition, the invention can provide enterprise information research tables for enterprise users, and the enterprise users can input the information according to the actual conditions of the enterprise users, thereby realizing the rapid analysis of enterprise data.
(3) An intelligent AI label system is established and divided into a universal map label, a structured hierarchical incubation map label, an enterprise self-defined label and an individualized special text label, the label hierarchical system can gradually realize the matching of enterprise data and policy data from shallow to deep, and a screening policy pool is gradually matched and reduced from multiple dimensions of universal policy content, enterprise industry categories, enterprise user self-defined labels, enterprise individualized data and the like, so that the final matching result has higher precision and better accords with the self-intention and the requirement of enterprise users.
(4) The method adopts three modes of forward matching, reverse condition and reverse diagnosis, except for pushing a matched policy for the user and shielding a unmatched policy, the method more innovatively provides a reverse diagnosis function for enterprise users, namely, the method matches the current policy which is not met but has a cultivation prospect of the enterprise according to the current condition of the enterprise, thereby providing long-term policy planning for the user and providing future cultivation direction and cultivation progress for the enterprise.
(5) The method and the system realize automatic collection of policy information on each website and classification and labeling of the policy information by adopting a semantic processing mode, realize efficient and accurate matching of enterprise data and policy data, solve the problems of low efficiency and poor accuracy of manual screening and matching of numerous and complicated policy information of the current enterprises or colleges, provide accurate intelligent policy matching and pushing service for users, provide policy guidance for the users, enable the users to timely know self-declared policies and provide policy cultivation guidance suggestions for the users.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made in the above embodiments by those of ordinary skill in the art without departing from the principle and spirit of the present invention. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (7)
1. A multi-dynamic intelligent matching recommendation method for enterprise policy information of digital government affairs is characterized by comprising the following steps:
step S1, crawling a plurality of groups of policy data from a webpage at regular time, performing text semantic analysis on the crawled data, performing modular disassembly and policy label setting on each group of policy data according to semantic analysis results, and storing the policy data in a policy database;
step S2, receiving enterprise data input by a user, respectively setting enterprise labels for different types of content information in the enterprise data, and storing the enterprise labels in an enterprise database; wherein the business data includes at least a business name; when the enterprise information input by the user is only an enterprise name, acquiring current published basic data of the enterprise by taking the enterprise name as a keyword, and performing policy matching on the enterprise according to the current published basic data to generate a preliminary matching result;
meanwhile, sending a non-public basic data acquisition request to the user to acquire non-public basic data related to policy declaration of the enterprise, combining the non-public basic data and the public basic data, performing secondary policy matching, and generating a final matching result;
step S3, positioning data types according to the set policy labels and enterprise labels, then performing multiple intelligent matching on enterprise data input by a user and multiple groups of policy data in the policy database, performing weight scoring on each round of matching results, and pushing a policy matching report to the user according to the final weight scoring result; wherein, the multiple intelligent matching comprises the following steps:
step S31, according to a preset policy text label system and an enterprise-policy layered incubation map, carrying out intelligent AI label setting on the enterprise data, wherein,
the policy text label system comprises: a universal map label, a structured layered incubation map label, a personalized special text label and an enterprise self-defined label, wherein,
(1) universal map label
Analyzing the enterprise data to obtain the development scale of the enterprise, matching corresponding universal policy content in the policy database according to the development scale of the enterprise, and setting a label of the matched universal policy for the enterprise data;
(2) map label is hatched to structuralized layering
Presetting a structuralized layered incubation map, wherein the structuralized layered incubation map comprises: a hierarchical matching relationship map of policy categories and enterprise industry categories,
analyzing the enterprise data to obtain the industry category of the enterprise, matching corresponding policy data from the hierarchical data of the policy category according to the industry category of the enterprise, and setting a matched structured hierarchical incubation map label for the enterprise data; the structural layered incubation map is preset with the reportable policy content corresponding to each industry category;
(3) personalized special text label
(3.1) matching the enterprise data with the policy data, and respectively setting corresponding personalized special text labels for the enterprise data according to policies meeting conditions and policies not meeting conditions but having cultivation potential in matching results;
(3.2) carrying out background recording on the browsing behaviors of the enterprise users, analyzing the behavior habits of the enterprise users, positioning corresponding interest data sets, matching the interest data sets with the policy data, and setting matched personalized special text labels for the enterprise users;
(4) enterprise self-defined label
Receiving label data actively input by an enterprise user;
step S32, searching matching policies in the policy database according to the universal map tag, the structured hierarchical incubation map tag, the enterprise custom tag and the personalized special text tag set in step S31, and performing weight scoring on each policy, including:
distributing different weight values for the universal map label, the structured layered incubation map label, the personalized special text label and the enterprise self-defined label;
searching all corresponding policy contents in the policy database according to the label categories, and performing multiple rounds of automatic matching of enterprise data and keywords of policy data aiming at each policy, wherein the multiple rounds of automatic matching comprise the following steps: calculating the matching degree of each round, and performing weight scoring on the matching of the round according to the matching degree to obtain a final weight scoring result of the enterprise applying the policy; the weight scoring is to set corresponding weight scores for policies of different label types, wherein the weight score of a policy corresponding to a universal atlas label < the weight score of a policy corresponding to a structured hierarchical incubation atlas label < the weight score of a policy corresponding to an enterprise custom label < the weight score of a policy corresponding to an individualized special text label, the weights of label categories corresponding to the matching of the policies are scored according to the matching of the policies, and then the scoring values of the matching results of the policies corresponding to all the label categories are accumulated to obtain a final scoring result of the policies;
wherein, in the many rounds of automatic matching, adopt forward matching, reverse condition, reverse diagnosis three kinds of modes:
(1) forward matching
Comparing the enterprise data with the industry to which the policy belongs, the region to which the policy belongs and the policy declaration conditions in the policy data respectively, and when a plurality of policies which meet the conditions exist, weighting and scoring the policies which meet the conditions as the reportable policies;
(2) reverse condition
Comparing the enterprise data with the industry to which the policy belongs, the region to which the policy belongs and the policy declaration condition in the policy data respectively, and when the enterprise data is judged not to accord with at least one hard necessary condition in the policy data, no longer pushing the policy data to the enterprise user; each item of policy data is preset with one or more hard necessary conditions for reporting the policy, and the policy is not allowed to be reported as long as one condition is not met;
(3) inverse diagnosis
And carrying out reverse analysis and diagnosis on the policy declaration of the enterprise according to the following two directions:
1) analyzing the declared projects and the enterprise basic data recorded in the enterprise data, and carrying out policy reverse diagnosis, wherein the policy reverse diagnosis comprises the following steps: diagnosing policies that are not declared but that meet the declared conditions;
2) matching the enterprise data with the policy data, and reversely diagnosing the policy which does not meet the declaration condition but has a difference with the declaration condition within a preset interval, wherein the policy comprises the following steps: the policy which is not declared but has cultivation prospect is diagnosed;
and S33, obtaining the final weight scoring result of each policy declared by the enterprise in the mode of S32, screening out the policy names of which the final weight scoring result is higher than a threshold value, sequencing the policy names according to the final weight scoring result from high to low, generating a policy matching report and pushing the policy matching report to a user.
2. The multiple dynamic intelligent matching recommendation method for enterprise policy information of digital government according to claim 1, wherein in the step S1, the method comprises the steps of:
step S11, performing text semantic analysis on the policy text to obtain a semantic analysis result; modularly disassembling the semantic analysis result according to a plurality of preset module topics, wherein the preset module topics comprise: policy making department, effective time of policy, industry to which policy belongs, region to which policy belongs, policy declaration condition, policy declaration flow and policy reward content;
step S12, setting corresponding module subject labels as the policy labels for the disassembled modular text paragraphs;
step S13, store each set of policy data into the policy database as a unit, wherein each set of policy data includes a modular text paragraph with a plurality of policy labels.
3. The multiple dynamic intelligent matching recommendation method for enterprise policy information of digital government according to claim 1, wherein in said step S2, the enterprise data is tagged in one of the following two forms:
(1) receiving enterprise description full text data input by the user, performing semantic analysis on the full text data to obtain different types of content information in the enterprise data, and setting enterprise tags;
(2) and providing a preset enterprise data template for the user, wherein each theme in the enterprise data template corresponds to an enterprise tag, and after the user finishes inputting according to the enterprise data template, automatically setting the enterprise tags for the input enterprise data.
4. The method according to claim 1, wherein the system browsing record data of each user is automatically recorded, and the interface theme with high browsing frequency of the user is analyzed to generate a user interest data set; and when the user login is detected, automatically recommending policy information associated with the user interest data set to the user.
5. The multiple dynamic intelligent matching recommendation method for enterprise policy information of digital government according to claim 1, further comprising the steps of: when a user registers, role setting options are provided for the user, and the options comprise: enterprise users, channel users and government agency users, and sets corresponding operation function interfaces and authorities for users with different roles;
and configuring a plurality of role authorities belonging to the enterprise user aiming at the same enterprise user, wherein each role authority distributes different policy matching viewing authority ranges for the same enterprise user according to the level of the login role.
6. The multiple dynamic intelligent matching recommendation method for enterprise policy information of digital government according to claim 1, wherein the policy database has an automatic updating function, the effective time label of the policy of each group of policy data is detected regularly, and when the judgment time is a past formula, the policy data is automatically rejected;
the enterprise database has an updating reminding function, regularly reminds users of updating enterprise data, and automatically generates a latest policy matching report for the users according to the updated enterprise data after the users are detected to be updated.
7. The method as claimed in claim 1, wherein the policy matching report provides a matching policy list to enterprise users, and further provides the enterprise users with a selection control whether there is a declared item, and the enterprise users will not be recommended any more for the declared item.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110370716.0A CN112765441B (en) | 2021-04-07 | 2021-04-07 | Enterprise policy information multiple dynamic intelligent matching recommendation method for digital government affairs |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110370716.0A CN112765441B (en) | 2021-04-07 | 2021-04-07 | Enterprise policy information multiple dynamic intelligent matching recommendation method for digital government affairs |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112765441A CN112765441A (en) | 2021-05-07 |
CN112765441B true CN112765441B (en) | 2021-11-02 |
Family
ID=75691250
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110370716.0A Active CN112765441B (en) | 2021-04-07 | 2021-04-07 | Enterprise policy information multiple dynamic intelligent matching recommendation method for digital government affairs |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112765441B (en) |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113435197B (en) * | 2021-06-25 | 2023-01-03 | 平安国际智慧城市科技股份有限公司 | Data pushing method and device, pushing server and storage medium |
CN113568873B (en) * | 2021-07-01 | 2024-03-22 | 浪潮软件股份有限公司 | Intelligent policy file matching method and device |
CN113569108B (en) * | 2021-09-23 | 2022-08-16 | 南京国睿信维软件有限公司 | Knowledge recommendation method based on project plan content and user behavior |
CN113870083A (en) * | 2021-09-27 | 2021-12-31 | 中关村意谷(北京)科技服务有限公司 | Policy matching method, device and system, electronic equipment and readable storage medium |
CN114422585A (en) * | 2021-12-27 | 2022-04-29 | 航天信息股份有限公司 | Message pushing method and system for enterprise service platform |
CN114398477A (en) * | 2022-01-19 | 2022-04-26 | 平安国际智慧城市科技股份有限公司 | Policy recommendation method based on knowledge graph and related equipment thereof |
CN114780735B (en) * | 2022-02-21 | 2022-11-25 | 北京融信数联科技有限公司 | Policy matching method, system and readable storage medium based on data analysis |
CN114297515B (en) * | 2022-03-10 | 2022-06-03 | 成都明途科技有限公司 | Information recommendation method and device, electronic equipment and storage medium |
CN114663208A (en) * | 2022-03-28 | 2022-06-24 | 奕资武汉商务咨询有限公司 | Enterprise tax intelligent management platform based on big data analysis |
CN115147081B (en) * | 2022-07-12 | 2023-10-27 | 芜湖中艺科技服务有限公司 | Policy matching method, system and storage medium based on artificial intelligence |
CN115576901A (en) * | 2022-11-10 | 2023-01-06 | 山东立信华创信息科技咨询有限公司 | Science and technology project information service system and method based on big data analysis |
CN118051607A (en) * | 2024-02-21 | 2024-05-17 | 北京市大数据中心 | Deep learning-based policy information service recommendation method, system and storage medium |
CN118096267A (en) * | 2024-04-29 | 2024-05-28 | 山东铂明网络科技有限公司 | Personalized advertisement delivery system and method based on data analysis |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107633033A (en) * | 2017-09-08 | 2018-01-26 | 成都链科信息科技有限公司 | A kind of policy big data intelligent Matching system and matching process |
CN110457696A (en) * | 2019-07-31 | 2019-11-15 | 福州数据技术研究院有限公司 | A kind of talent towards file data and policy intelligent Matching system and method |
CN111652524A (en) * | 2020-06-11 | 2020-09-11 | 中力数创(重庆)科技有限公司 | Method and device for intelligently matching policy and guiding improvement path |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080306750A1 (en) * | 2007-06-08 | 2008-12-11 | Adp, Inc. | Employment screening system and method |
CN111159630A (en) * | 2019-12-31 | 2020-05-15 | 科技谷(厦门)信息技术有限公司 | Park policy matching and evaluating method based on multi-standard decision model |
CN112380318A (en) * | 2020-11-12 | 2021-02-19 | 中国科学技术大学智慧城市研究院(芜湖) | Enterprise policy matching method based on label similarity |
-
2021
- 2021-04-07 CN CN202110370716.0A patent/CN112765441B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107633033A (en) * | 2017-09-08 | 2018-01-26 | 成都链科信息科技有限公司 | A kind of policy big data intelligent Matching system and matching process |
CN110457696A (en) * | 2019-07-31 | 2019-11-15 | 福州数据技术研究院有限公司 | A kind of talent towards file data and policy intelligent Matching system and method |
CN111652524A (en) * | 2020-06-11 | 2020-09-11 | 中力数创(重庆)科技有限公司 | Method and device for intelligently matching policy and guiding improvement path |
Also Published As
Publication number | Publication date |
---|---|
CN112765441A (en) | 2021-05-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112765441B (en) | Enterprise policy information multiple dynamic intelligent matching recommendation method for digital government affairs | |
Yang et al. | Citation analysis: a comparison of Google Scholar, Scopus, and Web of Science | |
Meho et al. | A new era in citation and bibliometric analyses: Web of Science, Scopus, and Google Scholar | |
Mayr et al. | Scientometrics and information retrieval: weak-links revitalized | |
Müller et al. | Towards a typology of business process management professionals: identifying patterns of competences through latent semantic analysis | |
US9613321B2 (en) | System and method for establishing a dynamic meta-knowledge network | |
Romanelli et al. | Four challenges when conducting bibliometric reviews and how to deal with them | |
CN106682150B (en) | Information processing method and device | |
Chang et al. | The research on the critical success factors of knowledge management and classification framework project in the Executive Yuan of Taiwan Government | |
Markham et al. | Unstructured text analytics to support new product development decisions | |
US20090144295A1 (en) | Apparatus and method for associating unstructured text with structured data | |
Bojović et al. | An overview of forestry journals in the period 2006–2010 as basis for ascertaining research trends | |
US20060026174A1 (en) | Patent mapping | |
El-Korany | Integrated expert recommendation model for online communities | |
Henning et al. | Building compatible and dynamic character matrices–Current and future use of specimen-based character data | |
Nguyen et al. | Ontologies supporting research-related information foraging using knowledge graphs: literature survey and holistic model mapping | |
Yang et al. | Performing literature review using text mining, Part I: Retrieving technology infrastructure using Google Scholar and APIs | |
Mazov et al. | Publication databases of research organizations as a tool for information studies | |
Fong | Framework of competitor analysis by monitoring information on the web | |
US20140222788A1 (en) | Research recommendation system | |
Junger | Can indexing be automated? The example of the Deutsche Nationalbibliothek | |
Dorn et al. | Meta-search in human resource management | |
Vardigan et al. | Creating Rich, Structured metadata: lessons learned in the metadata portal project | |
CN102541883A (en) | Library teaching material retrieval and management system | |
Carlstone | Scoring the quality of e-serials MARC records using Java |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |