CN115309885A - A knowledge graph construction, retrieval and visualization method and system for scientific and technological services - Google Patents

Abstract

The invention relates to a knowledge graph construction, retrieval and visualization method and a system for scientific and technological service, wherein the method comprises the following steps: acquiring resource metadata of a scientific and technological service platform and preprocessing the metadata; performing data cleaning and knowledge extraction on the metadata to obtain a scientific and technological resource entity list, an entity attribute list and an entity relation list; carrying out knowledge correction on the scientific and technological resource entities; calculating the membership degree of the screened scientific and technological resource entities; importing the processed scientific and technological resources into a knowledge map database; determining scientific and technological resource categories, and designing accurate filtering conditions for user retrieval; responding to the personalized retrieval requirement of the user and completing the query operation; and presenting the user retrieval content in a visual map form. According to the invention, scientific and technological service resources on the scientific and technological service platform are subjected to knowledge mapping construction and storage, personalized retrieval service and visual presentation effect are provided for platform users, dynamic requirements of the users are responded more flexibly, and the resource conversion rate of the scientific and technological service platform is improved.

Description

A knowledge map construction, retrieval and visualization method for scientific and technological services and system

technical field

The present invention relates to the field of modern service industry, especially to the field of scientific and technological services, and specifically refers to a method and system for constructing, retrieving and visualizing knowledge graphs for scientific and technological services.

Background technique

The science and technology service industry is one of the important components of the modern service industry and an emerging industry that provides services for the entire chain of technological innovation. Compared with foreign science and technology service industries, our country has the characteristics of late start and fast development in this industry, and is currently in a stage of small overall development but fast growth.

The knowledge graph originated from Google Knowledge Graph, which describes entities, events, attributes and their relationships in the objective world in a structured form, and expresses information in a form closer to human cognition. At the same time, in terms of data storage, the extended graph database models and manages data in a simple and intuitive way, which can more easily miniaturize and standardize data units, realize rich relationship connections, and more clearly describe the relationship between data complicated relationship. This technology provides us with an ability to better organize, manage and understand vast amounts of information.

Applying the technology of knowledge graph to scientific and technological services can realize the integration and supplementation of various information such as knowledge, behavior, and data, and carry out knowledge graph modeling of scientific and technological service resources, so that intelligent retrieval services can be precise, scene-based, and personalized It can effectively improve the conversion rate of science and technology service resources on the science and technology service platform.

Contents of the invention

The purpose of the present invention is to overcome the above-mentioned shortcomings of the prior art, and provide a knowledge map construction, retrieval and visualization method and system for scientific and technological services that meet the requirements of high accuracy, good flexibility, and high resource conversion rate.

In order to achieve the above purpose, the knowledge map construction, retrieval and visualization method and system for scientific and technological services of the present invention are as follows:

The main feature of the method for constructing, retrieving and visualizing knowledge graphs for scientific and technological services is that the method includes the following steps:

(1) Collect the resource metadata of the technology service platform and perform word segmentation and part-of-speech tagging preprocessing;

(2) Perform data cleaning and knowledge extraction on the preprocessed metadata, and obtain the entity list, entity attribute list and entity relationship list of scientific and technological resources;

(3) Perform knowledge correction on the scientific and technological service resource entity, and determine the scientific and technological service resources that can be imported into the knowledge map;

(4) Carry out resource membership degree calculation for the scientific and technological service resources that have been screened and determined to be imported into the knowledge map;

(5) Import the processed scientific and technological service resources into the knowledge graph database;

(6) Determine the category of scientific and technological resources, and design the precise filter conditions for user retrieval;

(7) In response to the user's personalized retrieval needs, complete the query operation in the knowledge graph database;

(8) The user search content is presented in the form of a visual graph to improve the user search experience.

Preferably, the science and technology service platform is a science and technology service resource sharing platform with a technology service transaction function, which supports users to register and log in by roles, and the roles include technology service suppliers and technology service demanders.

Preferably, the resource metadata of the science and technology service platform includes service providers, service products, instruments and equipment, park services, intellectual property rights, investment and experts, etc.

A science and technology service knowledge map construction, retrieval and visualization system applied to the method of construction, retrieval and visualization of the science and technology service knowledge map: including a data collection and preprocessing module, a knowledge extraction module, a knowledge correction module, a membership degree calculation module, and a map Import module, retrieval resource classification module, resource retrieval module and visualization module;

The data collection and preprocessing module is used to collect the resource information of the technology service platform, and divide it into structured data, semi-structured data and unstructured data, and perform word segmentation for semi-structured data and unstructured data and preprocessing of part-of-speech tagging;

The knowledge extraction module is used to perform data cleaning and traversal operations on structured data, and perform knowledge extraction operations on semi-structured data and unstructured data, specifically including entity recognition, relationship extraction and attribute extraction;

The knowledge correction module is used to correct the knowledge of the cleaned scientific and technological resources, and determine the complete scientific and technological resource information that can be imported into the knowledge graph database;

The membership degree calculation module is used to calculate the resource membership degree of the scientific and technological service resources determined to be imported into the knowledge map after knowledge correction;

The map import module is used to import the scientific and technological resources after data processing into the knowledge map database;

The retrieval resource classification module is used to classify the scientific and technological resources on the platform, and determine the filter conditions for user retrieval;

The resource retrieval module is used to respond to user retrieval requests and generate corresponding query statements, and complete query operations in the knowledge graph database;

The visualization module is used to present the user's retrieval results in the form of knowledge map visualization in the front-end interface.

Adopting the knowledge map construction, retrieval and visualization method and system for scientific and technological services of the present invention, the present invention stores the scientific and technological service resources on the scientific and technological service platform as a knowledge map. Compared with traditional relational database storage, it will Structured data is stored on the network to form a map instead of a table, which greatly improves the storage efficiency of scientific and technological resources. The invention uses the advantages of the node and relationship structure in the knowledge map for the retrieval of scientific and technological service resources. Through this retrieval method, the personalized needs of users can be accurately captured, the matching degree of supply and demand resources on the platform is improved, and the success rate of transaction of technological service resources on the platform is improved.

Description of drawings

FIG. 1 is a flow chart of the method for constructing, retrieving and visualizing a knowledge map for scientific and technological services according to the present invention.

Fig. 2 is a schematic structural diagram of the knowledge map construction, retrieval and visualization system for scientific and technological services of the present invention.

Fig. 3 is a brief flow chart of the method for constructing, retrieving and visualizing knowledge graphs for scientific and technological services according to the present invention.

Detailed ways

In order to describe the technical content of the present invention more clearly, further description will be given below in conjunction with specific embodiments.

The knowledge map construction, retrieval and visualization method for scientific and technological services of the present invention includes the following steps:

(1) Collect the resource metadata of the technology service platform and perform word segmentation and part-of-speech tagging preprocessing;

(2) Perform data cleaning and knowledge extraction on the preprocessed metadata, and obtain the entity list, entity attribute list and entity relationship list of scientific and technological resources;

(3) Perform knowledge correction on the scientific and technological service resource entity, and determine the scientific and technological service resources that can be imported into the knowledge map;

(4) Carry out resource membership degree calculation for the scientific and technological service resources that have been screened and determined to be imported into the knowledge map;

(5) Import the processed scientific and technological service resources into the knowledge graph database;

(6) Determine the category of scientific and technological resources, and design the precise filter conditions for user retrieval;

(7) In response to the user's personalized retrieval needs, complete the query operation in the knowledge graph database;

(8) The user search content is presented in the form of a visual graph to improve the user search experience.

Preferably, the science and technology service platform is a science and technology service resource sharing platform with a technology service transaction function, which supports users to register and log in by roles, and the roles include technology service suppliers and technology service demanders.

Preferably, the resource metadata of the science and technology service platform includes service providers, service products, instruments and equipment, park services, intellectual property rights, investment and experts, etc.

A science and technology service knowledge map construction, retrieval and visualization system applied to the method of construction, retrieval and visualization of the science and technology service knowledge map: including a data collection and preprocessing module, a knowledge extraction module, a knowledge correction module, a membership degree calculation module, and a map Import module, retrieval resource classification module, resource retrieval module and visualization module;

The data collection and preprocessing module is used to collect the resource information of the technology service platform, and divide it into structured data, semi-structured data and unstructured data, and perform word segmentation for semi-structured data and unstructured data and preprocessing of part-of-speech tagging;

The knowledge extraction module is used to perform data cleaning and traversal operations on structured data, and perform knowledge extraction operations on semi-structured data and unstructured data, specifically including entity recognition, relationship extraction and attribute extraction;

The knowledge correction module is used to correct the knowledge of the cleaned scientific and technological resources, and determine the complete scientific and technological resource information that can be imported into the knowledge graph database;

The membership degree calculation module is used to calculate the resource membership degree of the scientific and technological service resources determined to be imported into the knowledge map after knowledge correction;

The map import module is used to import the scientific and technological resources after data processing into the knowledge map database;

The retrieval resource classification module is used to classify the scientific and technological resources on the platform, and determine the filter conditions for user retrieval;

The resource retrieval module is used to respond to user retrieval requests and generate corresponding query statements, and complete query operations in the knowledge graph database;

The visualization module is used to present the user's retrieval results in the form of knowledge map visualization in the front-end interface.

In the specific embodiment of the present invention, it is a knowledge map construction, retrieval and visualization method for scientific and technological services, please refer to Figure 1 and Figure 3, the method of the present invention is implemented according to the following steps:

The first step is to collect the resource metadata of the technology service platform and perform word segmentation and part-of-speech tagging preprocessing.

Collect the metadata of scientific and technological resources from the resource library of the scientific and technological service platform, and divide it into structured data, semi-structured data, and unstructured data. Structured data includes tabular data, relational databases, etc. Semi-structured data includes log files, XML documents, JSON documents, etc. Unstructured data includes office documents in all formats, various reports, text, etc.

At the same time, word segmentation and part-of-speech tagging are preprocessed for semi-structured and unstructured data. Word segmentation and part-of-speech tagging are important preparations for complex NLP tasks such as knowledge extraction;

Different from the Latin language represented by English, it uses letters as the basic structure of words and spaces as separators between different words. When performing Chinese semantic recognition, Chinese sentences should first be segmented into Chinese character sequences, which will be segmented Sentences are divided into separate words so that the machine can correctly understand the meaning of the text;

Through part-of-speech tagging, the part-of-speech is determined for each word in each sentence in the text. The types of part-of-speech include noun (/n), verb (/v), gerund (/vn), adjective (/a), adverb (/ d) etc.;

The specific word segmentation and part-of-speech tagging steps are as follows:

Step 1.1: Determine the corresponding industry corpus dictionary according to the industry category to which different scientific and technological resources belong, and match the text to be segmented with the entries in the dictionary;

Step 1.2: Set the starting length of the longest forward match, divide the dictionary according to the length of different Chinese characters, and extract words in the order from front to back, and search the dictionary corresponding to the number of Chinese characters for each match;

Step 1.3: Set the starting length of the reverse longest match, divide the dictionary according to the length of different Chinese characters, and extract words according to the order from back to front, and search the dictionary corresponding to the number of Chinese characters for each match;

Step 1.4: Determine the optimal selection rule. If the number of words after the forward longest match and the reverse longest match are the same, use the least non-dictionary words and single-character words as the word segmentation result;

Step 1.5: Perform part-of-speech tagging according to the selected industry corpus dictionary, and design feature templates;

Step 1.6: Create a blank part-of-speech tagger, use the conditional random field model to set relevant parameters and complete the model training;

Step 1.7: Input the word-segmented text for part-of-speech tagging task.

Specifically, take the brief introduction of a certain project on the platform as an example for data preprocessing: "The project is in Shenzhen. The Internet APP project developed for schools and communities is currently in the research and development stage, and the equity financing is 3 million. The funds are mainly used for product research and development. , improve, follow the small-scale promotion in the local market in the later stage, try it out, and develop it to the whole country in the future.”

After word segmentation and part-of-speech tagging, the results are as follows: "[project/n, at/p, Shenzhen/ns,,/w, for/p, school/n,,/w, community/n, development/v, of/ uj, Internet/n, APP/nx, project/n,,/w, now/t, in/v, R&D/j, stage/n,,/w, equity/n, financing/vn, 3 million/m ,,/w, fund/n, main/b, used for/v, product/n, /uj, R&D/j,,/w, improvement/v,,,/w, following/p, later stage/f, Local /s, Market /n, /uj, Small /a, Range /n, Promotion /v, /w, Trial /v, /w, Future /t, Again /d, To /p, National / n, development /vn, ./w]".

The second step is to perform data cleaning and knowledge extraction on the preprocessed metadata, and obtain the list of scientific and technological resources entities, entity attributes and relationships between entities.

Perform data cleaning and traversal operations on the structured data obtained in the first step to obtain the list of scientific and technological resource entities, entity attribute lists, and entity-to-entity relationship lists.

Perform knowledge extraction on the semi-structured data and structured data obtained in the first step, determine named entities, entity attributes, and inter-entity relationships, and further obtain the list of scientific and technological resource entities, entity attributes, and inter-entity relationships.

The entities that need to be extracted in the present invention are related named entities of scientific and technological service resources, such as service providers, service products, equipment, parks, patents, investment institutions, experts, and so on. Each extracted named entity may be finally added to the knowledge graph as a related entity of technology services.

The entity attributes that need to be extracted in the present invention are various attributes of different scientific and technological service resources, such as patent inventors, patent applicants, patent agencies, equipment owners, service products provided by service providers, experts' affiliated units, etc. Each extracted entity attribute may be finally added to the knowledge graph as a related entity attribute of the technology service.

The relationships between entities that need to be extracted in the present invention are various relationships between different scientific and technological service resources, such as the ownership of a patent by a scientific research institution, the ownership of a technology by an expert, the cooperative relationship between an investor and a university, When a service provider provides a technology service product, etc., each entity relationship extracted may be finally added to the knowledge graph as a related entity relationship of the technology service.

Optionally, the entity attribute is regarded as a nominal relationship between the entity and the attribute value, and the attribute extraction task is transformed into a relation extraction task.

Optionally, use the supervised Lattice-LSTM Chinese name recognition algorithm model to realize the extraction of technology service named entities:

Step 2.1: Manually label the scientific and technological resource data to be extracted, and perform Embedding operations on the data to extract text context features;

Step 2.2: Use the word vector training tool Word2Vec to train the preprocessed text and construct the word vector corresponding to the text characters;

Step 2.3: Use the LSTM neural network layer to calculate and obtain the context representation vectors, and finally use these representation vectors as features to predict the label classification results through the CRF label reasoning layer, "B" indicates the start word, "E" indicates the end word, " S" means a single word;

Step 2.4: Use 70% of the labeled corpus as the training set and 30% of the corpus as the test set. The dimension of the word vector is set to 8logN, N is the size of the vocabulary, and the number of neurons in the hidden layer of the LSTM model is set to 150. Based on Lattice -LSTM model realizes the naming recognition of entities;

Step 2.5: Extract the relationship between entities of different scientific and technological resources, and obtain the semantic similarity between the relationship and the relationship word through cosine similarity calculation, and select the relationship word with high semantic similarity as the relationship between entities;

Step 2.6: On the basis of extracting the relationship based on semantic similarity, manually verify the relationship, improve and correct the language description, standardize the name attribute of the relationship and other features.

The third step is to correct the knowledge of the technology service resource entity and determine the technology service resources that can be imported into the knowledge map.

The scientific and technological resource data extracted from the large-scale multi-source heterogeneous database of the scientific and technological service platform must have characteristics such as data duplication and unclear reference. is very necessary.

After obtaining the scientific and technological resource entity, the different expressions of the same entity of data from different sources are fused to complete entity recognition, relationship linking and ontology generation. For example, two patents with the same patent number and publication number are duplicate patents. Only one patent is reserved.

Optionally, use the Limes tool for knowledge correction. Limes is a metric space-based entity matching discovery framework, suitable for large-scale data links, and the programming language is Java. The specific steps are as follows:

Step 3.1: Write a configuration file, including data source, fusion algorithm, fusion conditions and other information;

Step 3.2: Given a source dataset S, a target dataset T, and a threshold θ;

Step 3.3: Calculate the distance m(s, e) between s∈S and e∈E, use the triangle inequality to filter, and filter out the entity pairs (s ,t);

Step 3.4: Calculate the distance m(s, t) of the remaining entity pair (s, t), and store it in a user-specified format;

Step 3.5: Complete knowledge fusion according to entity similarity, and determine the scientific and technological service resource data that can be imported into the knowledge graph database.

The fourth step is to calculate the resource membership degree for the scientific and technological service resources that have been screened and determined to be imported into the knowledge map.

This system is developed based on the scientific and technological resource sharing service platform. The scientific and technological service sharing platform gathers all kinds of scientific and technological resource information needed for the development of small and medium-sized enterprises on the basis of traditional platforms. It has the characteristics of extensive data sources, messy data standards, and various fields involved. In order to meet the needs of platform business development, centering on platform operation and unified resource allocation, it is necessary to calculate the resource membership degree of its overall scientific and technological resources;

Taking the Hainan Science and Technology Island platform as an example, Hainan Science and Technology Island includes four major industries including modern service industry, marine economy, international medical care, and cultural tourism. The degree of membership is used for the classification of offline resources on the platform and the classification and display of online resources;

In order to accurately serve the construction of an open ecological service-oriented industrial system dominated by Hainan's modern service industry, marine economy, medical care, and cultural tourism, and to serve the high-quality development of the modern economic system in the Free Trade Zone, five major development models through the platform: Hainan Science and Technology Park Service Concentration , decentralized technology services for townships in Hainan, customized services for enterprises above the designated size in Hainan, professional services for Hainan's characteristic industries, and imported technology service resources from outside the island to jointly promote the development of comprehensive technology services in the Hainan Free Trade Zone, and classify the technology resource entities in the platform as resources And calculate the degree of membership of the different development models it belongs to, for the use of platform offline resource classification and online resource classification display;

Taking the calculation of the resource membership degree of the four major industries as an example, the calculation process of the resource membership degree is introduced in detail:

Step 4.1: Obtain the word segmentation and part-of-speech tagging results of each scientific and technological entity in the scientific and technological resource entity list obtained in step 1;

Step 4.2: Filter and segment the word segmentation results of the scientific and technological resource entity list according to the stop word dictionary;

Step 4.3: Accumulate the weights of each scientific and technological entity based on the classification weight of each scientific and technological resource vocabulary in the four major industry databases of the Science and Technology Island platform;

Step 4.4: Calculate the mean and standard deviation of the original data, subtract the difference from the mean from the original data and then divide by the standard deviation to normalize the data;

Step 4.5: Establish a relationship with the original scientific and technological entity with the calculation result as the membership degree attribute of the four major industries;

Step 4.6: Sort the membership degree of the four major industries to which the scientific and technological resources belong, and guide the background data scheduling and front-end operation management of the Kechuang Island platform.

The fifth step is to import the processed scientific and technological service resources into the knowledge graph database.

Neo4j is a high-performance NOSQL graph database that stores structured data on the network to form a graph instead of a table. It can also be regarded as a high-performance graph engine that has all the characteristics of a mature database.

A graph database contains two basic data types in a data unit: Nodes (nodes) and Relationships (relationships). Nodes and Relationships contain attributes in the form of key/value, and Nodes are connected through the relationships defined by Relationships to form a relational network structure.

Optionally, use the Neo4j graph database to store the obtained scientific and technological resource entities and the relationship between entities, traverse and read the processed scientific and technological resource names, attributes and relationships in turn, run the Neo4j graph database script to create scientific and technological resource entities and build the relationship between scientific and technological resource entities relationship between.

The specific process of importing the Neo4j database is as follows:

Step 5.1: Use Excsl, MySql and other tools to generate csv files from the processed scientific and technological resource data;

Step 5.2: Put the csv file into the import folder in Neo4j, and use the Load statement to read the data in Neo4jDesktop;

Step 5.3: Use the Create statement to create technology resource entity nodes, entity attributes, and relationships between entities;

Step 5.4: According to the size of the data, select the required node attributes to build an index to improve the efficiency of large-scale data retrieval;

Step 5.5: Use the Match statement to test whether the nodes and relationships are imported successfully.

The sixth step is to determine the category of scientific and technological resources, and design the precise filter conditions for user retrieval.

When responding to user search behavior, in order to finely determine user needs and accurately present scientific and technological service resources to users, according to the types of scientific and technological resources in the scientific and technological service platform, two levels of search categories are divided: first-level search categories and second-level search categories head;

According to the name of the scientific and technological resource entity in the imported graph database, determine the content of the first-level search category and the second-level search category to which different scientific and technological resources belong. The specific search categories are as shown in the following table:

First-level search category of scientific and technological resources Secondary search category of scientific and technological resources expert Experts, Universities, Research Teams technology Technical services, papers, patents, other intellectual property rights instrument Instrument equipment, instrument service finance Investors, investment projects, investment funds, investment institutions policy Policy information, policy consultation park Park, park service product service products, service providers

In order to maximize the excavation and meet the individual needs of users, based on the node-relationship-node architecture in the knowledge map, the filter conditions for user retrieval are established, including the content of the retrieval object, the content of the relationship object associated with the retrieval object, the retrieval object and the relationship The relationship between objects, as shown in the following table:

Specifically, the contents of the search object and the relationship object include subunit contents: the first-level search category, the second-level search category, the search (relationship) object content, and the search (relationship) object attributes, as shown in the following table :

First-level search category Secondary Search Category Retrieve (relational) object content Retrieve (relational) object properties

The seventh step is to respond to the user's personalized retrieval needs and complete the query operation in the knowledge graph database.

According to the personalized retrieval needs of different users, using the complex relational network unique to the knowledge graph, the relationship between different scientific and technological resources can be clearly detected, and the relationship as the retrieval object actually achieves what traditional relational databases cannot do. Retrieve the effect.

The main idea of the query step is as follows:

Step 7.1: The user determines the search object content of scientific and technological service resources and its first-level and second-level search categories and search object content attributes according to the needs;

Step 7.2: The user determines the search content of other scientific and technological service resources associated with the search content according to his actual needs, and determines the search content attributes and primary and secondary search categories of other scientific and technological service resources associated with the search content;

Step 7.3: The background responds to user retrieval conditions and generates corresponding Cypher query statements;

Step 7.4: Use the generated Cypher query statement to search the Neo4j database and return the search result.

The following two examples specifically introduce the user retrieval process:

Example one:

The user needs a service provider in Shanghai that can provide face recognition access control systems, and requires that the service provider has provided services for Shanghai University. According to this requirement, the search criteria are determined as follows:

After the background receives the user's retrieval instruction, it generates the corresponding Cypher statement according to the template:

Match(n: service provider)-[:cooperation]-(p:university)

Wheren.name='Access Control System' and n.location='Shanghai' and p.name='Shanghai University'

Return n, Count(n);

The system uses the generated Cypher statement to retrieve the Neo4j database and returns the retrieved results.

Example two:

The user has medical endoscope detection technology and is now seeking to invest 15 million. The investor is required to be located in Beijing, have had cooperation projects with scientific research institutions of Shanghai University, and have participated in the construction of a medical equipment industrial park. According to this demand, the search conditions are determined as follows surface:

After the background receives the user's retrieval instruction, it generates the corresponding Cypher statement according to the template:

Match(p: scientific research institution)-[:cooperation]-(n:investor)-[:provide]-(q:park service)

Where n.name='Medical Endoscope'and n.location='Beijing'and p.name='Shanghai University'and q.name='Medical Equipment'

Return n, Count(n);

The system uses the generated Cypher statement to retrieve the Neo4j database and returns the retrieved results.

The eighth step is to present user retrieval content in the form of a visual map to improve user retrieval experience.

Through the sixth and seventh steps, the search and filter conditions for scientific and technological resources are established. After the user’s search instructions are completed, the content of the user’s search is displayed in a visual form of the knowledge map, and the service of scientific and technological resources is recommended to users more accurately, so as to improve the success of the transaction of technological resources on the platform. Rate.

In the form of visual graphs, any form of retrieval can comprehensively present the various external connections of retrieved scientific and technological resources, as well as the connections between different scientific and technological resources, which provides insights that traditional texts, tables, and pictures cannot provide power to improve decision-making efficiency for users.

Preferably, use Neovis.js to construct a front-end knowledge graph to visualize and present content:

This component seamlessly integrates JavaScript visualization with Neo4j, and is an embedded tool that can be directly connected to Neo4j.

This component is built on the basis of Neo4j's property graph model, so the Neovis.js data format is consistent with Neo4j, and defines custom coloring styles based on labels, attributes, nodes, and relationships in a single configuration object, and allows developers to customize according to Set visual styles for nodes, relationships, or specific properties.

The steps to set up the visual component are as follows:

Step 8.1: Configure the Neo4j server port number, user name, and password, and connect to the Neo4j database to obtain real-time data;

Step 8.2: Select the DOM element that presents the visualization effect and set the style of the visualization element (nodes and relationships);

Step 8.3: Specify the label and the attribute to be displayed, specify the node attribute for the URL of the image of the node, specify the edge attribute for the edge thickness, and specify the node attribute for the node size;

Step 8.4: Respond to the user's search specification, receive the search results returned by the database, and build a front-end visualization component;

Step 8.5: Dynamically interact with user instructions, and respond to user operations in real time at the front end.

Please refer to Figure 2, a knowledge map construction, retrieval and visualization system for scientific and technological services according to the present invention, including data acquisition and preprocessing modules, knowledge extraction modules, knowledge correction modules, membership degree calculation modules, map import Module, Retrieval Resource Classification Module, Resource Retrieval Module and Visualization Module.

The data collection and preprocessing module is used to collect the resource information of the technology service platform, and divide it into structured data, semi-structured data and unstructured data, and perform word segmentation for semi-structured data and unstructured data and preprocessing of part-of-speech tagging;

The knowledge extraction module is used to perform data cleaning and traversal operations on structured data, and perform knowledge extraction on semi-structured data and unstructured data, specifically including entity recognition, relationship extraction and attribute extraction;

The knowledge correction module is used to correct the knowledge of the cleaned scientific and technological resources, and determine the complete scientific and technological resource information that can be imported into the knowledge graph database;

The membership degree calculation module is used to calculate the resource membership degree of the scientific and technological service resources determined to be imported into the knowledge map after knowledge correction;

The map import module is used to import the scientific and technological resources after data processing into the knowledge map database;

The retrieval resource classification module is used to classify the scientific and technological resources on the platform, and determine the filter conditions for user retrieval;

The resource retrieval module is used to respond to user retrieval requests and generate corresponding query statements, and complete query operations in the knowledge graph database;

The visualization module is used to present the user's retrieval results in the form of knowledge map visualization in the front-end interface.

For the specific implementation solution of this embodiment, reference may be made to relevant descriptions in the foregoing embodiments, and details are not repeated here.

It can be understood that, the same or similar parts in the above embodiments can be referred to each other, and the content that is not described in detail in some embodiments can be referred to the same or similar content in other embodiments.

It should be noted that, in the description of the present invention, the terms "first", "second" and so on are only used for description purposes, and should not be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" means at least two.

Any process or method descriptions in flowcharts or otherwise described herein may be understood to represent modules, segments or portions of code comprising one or more executable instructions for implementing specific logical functions or steps of the process , and the scope of preferred embodiments of the invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including substantially concurrently or in reverse order depending on the functions involved, which shall It is understood by those skilled in the art to which the embodiments of the present invention pertain.

It should be understood that various parts of the present invention can be realized by hardware, software, firmware or their combination. In the above-described embodiments, various steps or methods may be implemented by software or firmware stored in a memory and executed by a suitable instruction execution device. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques known in the art: Discrete logic circuits, ASICs with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

Those of ordinary skill in the art can understand that all or part of the steps carried by the method of the above-mentioned embodiments can be completed by instructing related hardware through a program, and the corresponding program can be stored in a computer-readable storage medium. When, one or a combination of the steps of the method embodiment is included.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing module, each unit may exist separately physically, or two or more units may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. If the integrated modules are realized in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium.

The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, and the like.

In the description of this specification, descriptions referring to the terms "one embodiment", "some embodiments", "example", "specific examples", or "some examples" mean that specific features described in connection with the embodiment or example , structure, material or feature is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Adopting the knowledge map construction, retrieval and visualization method and system for scientific and technological services of the present invention, the present invention stores the scientific and technological service resources on the scientific and technological service platform as a knowledge map. Compared with traditional relational database storage, it will Structured data is stored on the network to form a map instead of a table, which greatly improves the storage efficiency of scientific and technological resources. The invention uses the advantages of the node and relationship structure in the knowledge map for the retrieval of scientific and technological service resources. Through this retrieval method, the personalized needs of users can be accurately captured, the matching degree of supply and demand resources on the platform is improved, and the success rate of transaction of technological service resources on the platform is improved.

In this specification, the invention has been described with reference to specific embodiments thereof. However, it is obvious that various modifications and changes can be made without departing from the spirit and scope of the invention. Accordingly, the specification and drawings are to be regarded as illustrative rather than restrictive.

Claims (6)

1. A knowledge map construction, retrieval and visualization method for scientific and technological services, characterized in that it comprises the following steps: (1) Collect metadata of technology service platform resources, and perform word segmentation and part-of-speech tagging preprocessing; (2) Perform data cleaning and knowledge extraction on the preprocessed metadata, and obtain the entity list, entity attribute list and entity relationship list of scientific and technological resources; (3) Perform knowledge correction on the scientific and technological service resource entity, and determine the scientific and technological service resources that can be imported into the knowledge map; (4) Carry out resource membership degree calculation for the scientific and technological service resources that have been screened and determined to be imported into the knowledge map; (5) Import the processed scientific and technological service resources into the knowledge graph database; (6) Determine the category of scientific and technological resources, and design the precise filter conditions for user retrieval; (7) In response to the user's personalized retrieval needs, complete the query operation in the knowledge graph database; (8) The user search content is presented in the form of a visual graph to improve the user search experience. 2. The knowledge map construction, retrieval and visualization method for scientific and technological services according to claim 1, wherein the scientific and technological service platform is a scientific and technological service resource sharing platform with a technological service transaction function, and supports user registration by role Log in, and the roles include technology service provider and technology service demander. 3. The knowledge map construction, retrieval and visualization method for scientific and technological services according to claim 1, characterized in that, the precise filtering conditions for user retrieval in step (6) are specifically: According to the types of scientific and technological resources in the scientific and technological service platform, two levels of search categories are divided: first-level search categories and second-level search categories; Based on the node-relationship-node architecture in the knowledge graph, filter conditions for user retrieval are established, specifically including the content of the retrieval object, the content of the relationship object associated with the retrieval object, and the relationship between the retrieval object and the relationship object; The content of the retrieval object and the content of the relationship object include subunit contents: the primary retrieval category, the secondary retrieval category, the content of the retrieval relationship object, and the attributes of the retrieval relationship object. 4. A method for constructing, retrieving and visualizing a knowledge map for scientific and technological services according to claim 1, wherein the resources of the scientific and technological service platform include intellectual property rights, scientific and technological policies, technological finance, technology transfer, research and development , Entrepreneurial incubation, inspection and testing certification, equipment, scientific research team and comprehensive scientific and technological services. 5. A method for constructing, retrieving and visualizing knowledge graphs for scientific and technological services according to claim 1, characterized in that, in the step (2), data cleaning and knowledge extraction are performed on metadata, specifically including the following Steps: perform traversal operations on structured data, perform knowledge extraction operations on unstructured and semi-structured data, determine named entities, entity attributes, and inter-entity relationships, and further obtain scientific and technological resource entity lists, entity attribute lists, and inter-entity relationships list; The entity named after the knowledge extraction operation is a related named entity of the technology service resource, and each extracted named entity can be optionally added to the knowledge map as a related entity of the technology service; The entity attributes named after the knowledge extraction operation are various attributes of different technological service resources, and each extracted entity attribute can be selectively added to the knowledge map as a relevant entity attribute of technological services; The inter-entity relationships named after the knowledge extraction operation are various relationships between different technological service resources, and each extracted entity relationship can be selectively added to the knowledge map as a related entity relationship of technological services. 6. A science and technology service knowledge map construction, retrieval and visualization system applied to the method described in claims 1 to 5, characterized in that the system includes a data collection and preprocessing module, a knowledge extraction module, and a knowledge correction module , membership degree calculation module, map import module, retrieval resource classification module, resource retrieval module and visualization module, The data collection and preprocessing module is used to collect the resource information of the technology service platform, and divide it into structured data, semi-structured data and unstructured data, and perform word segmentation for semi-structured data and unstructured data and preprocessing of part-of-speech tagging; The knowledge extraction module is connected with the data acquisition and preprocessing module, and is used to perform data cleaning and traversal operations on structured data, and perform knowledge extraction operations on semi-structured data and unstructured data, specifically including entity Recognition, relation extraction and attribute extraction; The knowledge correction module is connected with the knowledge extraction module, and is used to correct the knowledge of the cleaned scientific and technological resources, and determine the complete scientific and technological resource information that can be imported into the knowledge graph database; The membership degree calculation module is connected to the knowledge correction module, and is used to calculate the resource membership degree of the scientific and technological service resources determined to be imported into the knowledge map after knowledge correction; The map import module is connected with the membership degree calculation module, and is used to import the scientific and technological resources after data processing into the knowledge map database; The retrieval resource classification module is connected with the map import module, and is used to classify the scientific and technological resources on the platform and determine the filter conditions for user retrieval; The resource retrieval module is connected with the retrieval resource classification module, and is used to respond to user retrieval requests and generate corresponding query statements, and complete query operations in the knowledge graph database; The visualization module is connected with the resource retrieval module, and is used to present the user's retrieval results in the form of knowledge map visualization in the front-end interface.

Images