Expert matching methods, devices, and storage media

Abstract

This disclosure provides an expert matching method, apparatus, and storage medium, relating to the field of computer technology. The method includes: constructing an expert technical capability knowledge graph based on a first entity and a first relationship; setting corresponding expert capability weights for first nodes based on a series; constructing an enterprise demand graph based on a second entity and a second relationship; determining the enterprise demand weights for corresponding second nodes based on the second entity; using the matching first and second nodes as matching node pairs; determining the matching degree based on the expert capability weights and enterprise demand weights corresponding to the matching node pairs; and determining recommended experts that match the enterprise demand information based on the matching degree. The method, apparatus, and medium of this disclosure utilize a knowledge graph to construct an information system, making the recommended expert information more accurate and complete; by setting weight information for keywords to calculate the comprehensive expert matching degree, it can better match enterprise needs and improve the user experience.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more particularly to an expert matching method, apparatus, and storage medium. Background Technology

[0002] When enterprises undertake technological innovation projects, especially those involving interdisciplinary and cross-domain integrated innovation, they often face problems such as insufficient internal expert reserves and a lack of adequate technical expert resources. Furthermore, existing expert matching methods suffer from incomplete expert databases, a lack of information on successfully completed projects by experts and the enterprise's historical project requirements, and simplistic processing methods for enterprise project needs, failing to differentiate between the requirements of different enterprises. Therefore, they cannot effectively determine the suitability between enterprises and experts. Thus, a new technical solution for expert matching is needed. Summary of the Invention

[0003] To address the aforementioned technical problems, this disclosure is proposed. Embodiments of this disclosure provide an expert matching method, apparatus, and storage medium.

[0004] According to a first aspect of the present disclosure, an expert matching method is provided, comprising: acquiring expert information, extracting a first entity from the expert information and determining a first relationship between the first entities, and constructing an expert technical capability knowledge graph based on the first entities and the first relationship; wherein the expert technical capability knowledge graph includes a plurality of first nodes for representing the first entities; determining a level corresponding to the first node, and setting a corresponding expert capability weight for the first node based on the level; acquiring enterprise demand information, extracting a second entity from the enterprise demand information and determining a second relationship between the second entities, and constructing an enterprise demand graph based on the second entities and the second relationship; wherein the enterprise demand graph includes: a plurality of second nodes for representing the second entities; determining an enterprise demand weight for the corresponding second node according to the second entity; searching in the expert technical capability knowledge graph whether there is a first node that matches the second node of the enterprise demand graph, and if so, taking the matching first node and second node as a matching node pair; determining a matching degree based on the expert capability weight and enterprise demand weight corresponding to the matching node pair, and determining a recommended expert that matches the enterprise demand information according to the matching degree.

[0005] Optionally, the step of extracting a first entity from the expert information and determining a first relationship between the first entities, and constructing an expert technical capability knowledge graph based on the first entity and the first relationship, includes: extracting a first keyword from the expert information as the first entity; extracting a first association relationship between the first keywords from the expert information as the first relationship; constructing a first triplet based on the first keyword and the first association relationship, and constructing the expert technical capability knowledge graph based on the first triplet; wherein, the expert information includes one or more of the following: structured, semi-structured, or unstructured introductions, projects, research, and professional information; the first node is used to represent the first keyword, and the connection between the first nodes is used to represent the corresponding first association relationship.

[0006] Optionally, determining the level corresponding to the first node and setting the corresponding expert capability weight for the first node based on the level includes: selecting a first node as the root node in the expert technical capability knowledge graph and setting the expert capability weight of the root node; sequentially determining the levels corresponding to the root node for other first nodes and setting the expert capability weight for the other first nodes based on the level.

[0007] Optionally, the first keyword represented by the root node includes: expert name or number; the larger the level corresponding to the first node, the lower the expert capability weight set for this first node.

[0008] Optionally, the step of extracting a second entity from the enterprise demand information and determining a second relationship between the second entities, and constructing an enterprise demand graph based on the second entity and the second relationship, includes: extracting a second keyword from the enterprise demand information as the second entity; extracting a second association relationship between the second keywords from the enterprise demand information as the second relationship; constructing a second triplet based on the second keyword and the second association relationship, and constructing the enterprise demand graph based on the second triplet; wherein, the enterprise demand information includes: structured, semi-structured, or unstructured enterprise demand information; the second node is used to represent the second keyword, and the connection between the second nodes is used to represent the corresponding second association relationship.

[0009] Optionally, determining the enterprise demand weight of the corresponding second node based on the second entity includes: obtaining feature information of the second keyword represented by the second node; determining the enterprise demand weight of the second node based on the feature information; wherein the feature information includes: the word frequency of the second keyword in the enterprise demand information, the weight information of the second keyword set by the user, or the attribute information corresponding to the second keyword.

[0010] Optionally, the step of searching in the expert technical capability knowledge graph whether there is a first node that matches the second node of the enterprise demand graph includes: sorting all the second keywords based on the enterprise demand weight; selecting a preset number of second keywords as matching keywords according to the sorting result; and searching in the expert technical capability knowledge graph whether there is a first node that matches the matching keywords; wherein the first keyword represented by this first node is the matching keyword or a synonym of the matching keyword.

[0011] Optionally, determining the matching degree based on the expert capability weights and enterprise demand weights corresponding to the matching node pair includes: determining the matching degree. Among them, A k Let α be the enterprise demand weight of the second keyword in the k-th matching node pair. k is the expert capability weight of the first keyword in the k-th matching node pair, and n is the number of matching node pairs.

[0012] Optionally, determining the recommended experts matching the enterprise's demand information based on the matching degree includes: determining the predicted workload based on the enterprise's demand information, and determining the predicted available time of the experts based on the expert information and the matching degree; ranking all experts according to the matching degree; determining a predetermined number of recommended experts based on the ranking result, the predicted workload, and the predicted available time; and ranking the recommended experts according to the predicted available time, wherein the predicted available time is the product of the expert's available working time and the corresponding matching degree.

[0013] According to a second aspect of the present disclosure, an expert matching apparatus is provided, comprising: an expert graph construction module, configured to acquire expert information, extract a first entity from the expert information and determine a first relationship between the first entities, and construct an expert technical capability knowledge graph based on the first entities and the first relationship; wherein the expert technical capability knowledge graph includes a plurality of first nodes for representing the first entities; an expert weight setting module, configured to determine a level corresponding to the first node, and set a corresponding expert capability weight for the first node based on the level; and a demand graph construction module, configured to acquire enterprise demand information, extract a second entity from the enterprise demand information and determine a second relationship between the second entities, and construct an expert technical capability knowledge graph based on the first entity and the first relationship; wherein the expert technical capability knowledge graph includes a plurality of first nodes for representing the first entities; an expert weight setting module, configured to determine a level corresponding to the first node, and set a corresponding expert capability weight for the first node; and a demand graph construction module, configured to acquire enterprise demand information, extract a second entity from the enterprise demand information and determine a second relationship between the second entities, and construct an expert technical capability knowledge graph based on the first entity and the first relationship. A business demand graph is constructed using two entities and the second relationship; wherein the business demand graph includes: multiple second nodes for representing the second entities; a demand weight setting module for determining the business demand weight of the corresponding second node based on the second entity; a graph matching processing module for searching in the expert technical capability knowledge graph whether there is a first node that matches the second node of the business demand graph, and if so, taking the matching first node and second node as a matching node pair; a matching degree determination module for determining the matching degree based on the expert capability weight and business demand weight corresponding to the matching node pair; and a recommended expert determination module for determining recommended experts that match the business demand information based on the matching degree.

[0014] Optionally, the expert graph construction module is used to extract a first keyword from the expert information as the first entity; extract a first association relationship between the first keywords from the expert information as the first relationship; construct a first triplet based on the first keyword and the first association relationship, and construct the expert technical capability knowledge graph based on the first triplet; wherein, the expert information includes one or more of the following: structured, semi-structured, or unstructured introductions, projects, research, and professional information; the first node is used to represent the first keyword, and the connection between the first nodes is used to represent the corresponding first association relationship.

[0015] Optionally, the expert weight setting module is used to select a first node as the root node in the expert technical capability knowledge graph and set the expert capability weight of the root node; sequentially determine the level of other first nodes corresponding to the root node, and set the expert capability weight of the other first nodes based on the level.

[0016] Optionally, the first keyword represented by the root node includes: expert name or number; the larger the level corresponding to the first node, the lower the expert capability weight set for this first node.

[0017] Optionally, the demand graph construction module is used to extract a second keyword from the enterprise demand information as the second entity; extract a second association between the second keywords from the enterprise demand information as the second relationship; construct a second triplet based on the second keyword and the second association, and construct the enterprise demand graph based on the second triplet; wherein, the enterprise demand information includes: structured, semi-structured, or unstructured enterprise demand information; the second node is used to represent the second keyword, and the connection between the second nodes is used to represent the corresponding second association.

[0018] Optionally, the demand weight setting module is used to obtain feature information of the second keyword represented by the second node; and determine the enterprise demand weight of the second node based on the feature information; wherein, the feature information includes: the word frequency of the second keyword in the enterprise demand information, the weight information of the second keyword set by the user, or the attribute information corresponding to the second keyword.

[0019] Optionally, the graph matching processing module is used to sort all the second keywords based on the enterprise demand weight; select a preset number of second keywords as matching keywords according to the sorting result; and search in the expert technical capability knowledge graph whether there is a first node that matches the matching keyword; wherein the first keyword represented by this first node is the matching keyword or a synonym of the matching keyword.

[0020] Optionally, the matching degree determination module is used to determine the matching degree. Among them, A k Let α be the enterprise demand weight of the second keyword in the k-th matching node pair. k is the expert capability weight of the first keyword in the k-th matching node pair, and n is the number of matching node pairs.

[0021] Optionally, the expert recommendation module is configured to determine the predicted workload based on the enterprise demand information, and determine the predicted available time of the experts based on the expert information and the matching degree; sort all experts according to the matching degree; determine a predetermined number of recommended experts based on the sorting result, the predicted workload, and the predicted available time; and sort the recommended experts according to the predicted available time, wherein the predicted available time is the product of the expert's available working time and the corresponding matching degree.

[0022] According to a third aspect of the present disclosure, an expert matching apparatus is provided, comprising: a processor; a memory for storing processor-executable instructions; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the method described above.

[0023] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, the storage medium storing a computer program for performing the above-described method.

[0024] Based on the expert matching method, apparatus, and storage medium provided in the above embodiments of this disclosure, an information system is constructed using a knowledge graph, making the recommended expert information more accurate and complete. Expert information can be collected comprehensively and dynamically, fully describing the experts' capabilities, better matching them with enterprise needs, and making more comprehensive and accurate technical expert recommendations. By judging the available time information of experts, the problem of experts being successfully matched but not having time to work can be avoided. By setting weight information for keywords to calculate the comprehensive matching degree of experts, the optimal recommended experts can be selected, improving the user experience.

[0025] The technical solutions of this disclosure will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0026] The above and other objects, features, and advantages of this disclosure will become more apparent from the more detailed description of the embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the embodiments of this disclosure to explain the disclosure and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0027] Figure 1 This is a flowchart illustrating one embodiment of the expert matching method disclosed herein;

[0028] Figure 2A This is a schematic diagram illustrating the process of constructing an expert technical capability knowledge graph and setting weights in one embodiment of the expert matching method disclosed herein. Figure 2B and Figure 2C A schematic diagram of an expert's technical capability knowledge graph;

[0029] Figure 3A This is a schematic diagram illustrating the process of constructing an enterprise demand graph and setting weights in one embodiment of the expert matching method disclosed herein; Figure 3B A schematic diagram of enterprise demand mapping;

[0030] Figure 4This is a schematic diagram of the process for determining recommended experts that match enterprise needs information in one embodiment of the expert matching method disclosed herein;

[0031] Figure 5 This is a schematic diagram of one embodiment of the expert matching device of this disclosure;

[0032] Figure 6 This is a schematic diagram of another embodiment of the expert matching device of this disclosure. Detailed Implementation

[0033] Those skilled in the art will understand that the terms "first," "second," etc., in the embodiments of this disclosure are only used to distinguish different steps, devices, or modules, and do not represent any specific technical meaning, nor do they indicate a necessary logical order between them.

[0034] It should also be understood that in the embodiments disclosed herein, "multiple" can refer to two or more, and "at least one" can refer to one, two or more.

[0035] It should also be understood that any component, data or structure mentioned in the embodiments of this disclosure can generally be understood as one or more unless expressly defined or given to the contrary in the context.

[0036] Furthermore, the term "and / or" in this disclosure is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this disclosure generally indicates that the preceding and following related objects have an "or" relationship.

[0037] It should also be understood that the description of the various embodiments in this disclosure emphasizes the differences between the various embodiments, and the similarities or commonalities can be referred to each other. For the sake of brevity, they will not be described in detail. At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0038] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use.

[0039] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0040] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0041] The embodiments of this disclosure can be applied to electronic devices such as terminal devices, computer systems, and servers, and can operate with a wide range of other general-purpose or special-purpose computing system environments or configurations. Examples of well-known terminal devices, computing systems, environments, and / or configurations suitable for use with electronic devices such as terminal devices, computer systems, or servers include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments that include any of the above systems, etc.

[0042] Electronic devices such as terminal devices, computer systems, and servers can be described in the general context of computer system executable instructions (such as program modules) executed by a computer system. Typically, program modules can include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types.

[0043] Exemplary methods

[0044] In existing expert matching solutions, matching projects based on manual information retrieval by both supply and demand sides or on platform tag data is limited by the keyword content of different online platforms and the source of expert information. This makes it impossible to effectively measure the expert's capabilities and their suitability for the company's technical needs. Furthermore, it does not focus on successful project experience. Therefore, it is difficult to comprehensively consider factors such as expert capabilities, R&D resources, time arrangements, and estimated costs to find the most suitable expert support.

[0045] Figure 1 This is a flowchart illustrating one embodiment of the expert matching method disclosed herein, as follows: Figure 1 The method shown includes steps S101-S106. Each step is explained below.

[0046] S101, Obtain expert information, extract first entities from the expert information and determine the first relationships between the first entities, and construct an expert technical capability knowledge graph based on the first entities and the first relationships; wherein, the expert technical capability knowledge graph includes multiple first nodes used to represent the first entities. The first entity can be a first keyword, etc.

[0047] S102, determine the level corresponding to the first node, and set the corresponding expert capability weight for the first node based on the level.

[0048] S103, Obtain enterprise demand information, extract second entities from the enterprise demand information and determine the second relationships between the second entities, and construct an enterprise demand graph based on the second entities and the second relationships; wherein, the enterprise demand graph includes: multiple second nodes used to represent the second entities; the second entities can be second keywords, etc.

[0049] S104, determine the enterprise demand weight of the corresponding second node based on the second entity.

[0050] S105: Search the expert technical capability knowledge graph for a first node that matches the second node of the enterprise demand graph. If a first node matches, then use the matching first node and second node as a matching node pair.

[0051] S106. The matching degree is determined based on the expert capability weight and enterprise demand weight corresponding to the matching node pair, and the recommended experts that match the enterprise demand information are determined according to the matching degree.

[0052] This disclosed expert matching method provides a weighted knowledge graph-based approach for matching technical needs with experts. The knowledge graph provides a graphical method for constructing directed or undirected graph networks, building a relationship network between entities using triples (entity-relationship-entity). The knowledge graph represents various entities and concepts existing in the real world, as well as the relationships between entities, effectively describing enterprise needs and expert research capabilities. By mining the relationships between enterprises and technical experts, it solves the problem of information asymmetry between enterprises and technical experts, enabling mutual recommendations between enterprise technical needs and technical experts, and improving the matching accuracy of enterprises and experts on the platform.

[0053] Expert information includes one or more types of information such as structured, semi-structured, or unstructured introductions, projects, research, and professions. Figure 2A This is a schematic diagram illustrating the process of constructing an expert technical capability knowledge graph and setting weights in one embodiment of the expert matching method disclosed herein, as shown below. Figure 2A The method shown includes steps S201-S205. Each step is explained below.

[0054] S201, extract the first keyword from the expert information and use it as the first entity.

[0055] S202, extract the first association between the first keywords from the expert information, and use it as the first relationship.

[0056] S203, construct the first triplet based on the first keyword and the first association, and construct an expert technical capability knowledge graph based on the first triplet. The first node in the expert technical capability knowledge graph is used to represent the first keyword, and the connection between the first nodes is used to represent the corresponding first association.

[0057] S204. In the expert technical capability knowledge graph, select a first node as the root node. The first keyword represented by the root node includes the expert's name or number, etc.

[0058] S205, set the expert capability weight for the root node, sequentially determine the level corresponding to the root node for each of the other first nodes, and set the expert capability weight for each of the other first nodes based on the level. The higher the level of the first node, the lower the expert capability weight set for that first node. Alternatively, no expert capability weight can be set for the root node.

[0059] In one embodiment, expert information is collected from multiple information channels. This expert information includes details about expert projects, research, and professional backgrounds. The information can be structured, semi-structured, or unstructured. Structured information or data typically refers to information or data in expert databases, knowledge bases, etc. Unstructured information data refers to information or data such as voice, images, and video. Semi-structured data refers to descriptive information or data with implicit but not rigorous structure, such as emails, web pages, and reports.

[0060] Multiple methods are employed to parse structured or semi-structured expert information. For example, a trained neural network model is used to parse the expert information, obtaining entity-keyword (first keyword), entity-other information (first association)-keyword, and constructing a triplet knowledge graph, such as (Expert 1, technical keywords, light industry and chemical biotechnology, biocatalysis and fermentation, biological products, biopesticides); (Expert 1, institution name, Guangdong XX University); (Guangdong XX University, outstanding achievements, biocatalysis and fermentation, biological products, biopesticides); where Expert 1, Guangdong XX University, chemical biotechnology, etc. are the first keywords, and technical keywords, outstanding achievements, etc. are the first associations.

[0061] Various preprocessing techniques can be applied to expert information. For example, if the expert information is expert project information, word segmentation can be performed on the expert project information, followed by processes such as removing punctuation marks and invisible characters, Chinese word segmentation, removal of stop words, and removal of low-frequency words, transforming the expert project information into a sentence constructed from a sequence of keywords.

[0062] Various methods can be used to parse unstructured expert information. For example, a trained neural network model can be used to parse the expert information, processing it into a sentence constructed from a sequence of words. The word frequency of each word in each expert information document is calculated, and the words with the highest frequency are selected as the first keywords. The first association is then obtained and added to the knowledge graph.

[0063] In the constructed expert technical capability knowledge graph, different first nodes represent first keywords with different weights. Typically, the first node corresponding to the expert's name or number is selected as the root node, and the first node directly connected to the root node is a first-level node, and so on. Alternatively, the first node corresponding to the first keyword directly obtained from structured information can be used as a first-level node, and the first node corresponding to the first keyword interpreted from unstructured information can be used as a second-level node.

[0064] For example, the text information portion of expert Li Moufang's expert information is shown below:

[0065] During her time at a certain A-gene laboratory, Li Moufang's research interests included genome editing technology, construction of animal models of human diseases and research on pathogenesis, protein purification technology and functional research, high-throughput drug screening, evaluation of the activity and pharmacology of natural drugs, fish genetics research, and detection and mechanism research of aquatic diseases.

[0066] The keywords extracted from the research direction are: gene editing, genetic research, aquatic diseases, and drug activity.

[0067] In addition, the published research findings (expert information) of the A-gene laboratory where Li Moufang works include:

[0068] The study revealed the important role of molecule 000 in X cell-dependent humoral immune responses in small fish, providing a new theoretical basis for the development of fish vaccines and possessing significant theoretical innovation value.

[0069] A survey on the market value of variant ornamental fish, including some basic research.

[0070] Based on the information from a certain A gene laboratory, the keywords for the research field can be extracted as: humoral immunity, fish vaccines, and fish mutation.

[0071] The expert technical ability knowledge graph of Li Moufang constructed using the above method is as follows: Figure 2BAs shown, the first node corresponding to the first keyword "Li Moufang" is the root node, and the first nodes corresponding to the first keywords "gene editing," "genetic research," etc., directly connected to the root node are designated as first-level nodes; the first nodes corresponding to the first keywords "humoral immunity," "fishery vaccines," etc., directly connected to the first-level node corresponding to the first keyword "A gene laboratory," are designated as second-level nodes. The expert ability weight of the first-level nodes is defined as 1, and the expert ability weight of the second-level nodes is defined as 0.5. Therefore, the weight table of Li Moufang's expert technical ability knowledge graph is as follows:

[0072]

[0073] Table 1 - Expert Ability Weights Corresponding to Li Moufang's Expert Technical Ability Knowledge Graph

[0074] Regarding expert Liu Mouhao, the textual information about Liu Mouhao's expert status is described as follows:

[0075] His main research and development work includes: development of novel targeted antibody drugs for small cell lung cancer, application and industrialization of novel genetically engineered vaccines for classical swine fever, and design and development of pseudorabies protein vaccines. Extracting relevant technical keywords, Liu Mouhao's research areas are targeted antibodies, novel genes, classical swine fever vaccines, and pseudorabies protein.

[0076] Expert Liu Mouhao's workplace is a certain B biological laboratory, and his main research directions are: (1) biomolecular regulation of time medicine and stem cell differentiation. (2) intracellular vesicle transport mechanism and vaccine application industrialization research.

[0077] Based on the information from Biological Laboratory B above, the primary keywords for the research field can be extracted as: biomolecules, stem cell differentiation, and vaccine application.

[0078] The expert technical ability knowledge graph of Liu Mouhao constructed using the above method is as follows: Figure 2C As shown, the first node corresponding to the first keyword "Liu Mouhao" is the root node, and the first nodes corresponding to the first keywords "novel gene," "targeted antibody," etc., directly connected to the root node are designated as first-level nodes; the first nodes corresponding to the first keywords "biomolecule," "stem cell differentiation," etc., directly connected to the first-level node corresponding to the first keyword "Biological Laboratory B," are designated as second-level nodes. The expert ability weight of the first-level nodes is defined as 1, and the expert ability weight of the second-level nodes is defined as 0.5. Therefore, the weight table of Liu Mouhao's expert technical ability knowledge graph is as follows:

[0079]

[0080]

[0081] Table 2 - Expert Ability Weights in Liu Mouhao's Expert Technical Ability Knowledge Graph

[0082] Figure 3A This is a schematic diagram illustrating the process of constructing an enterprise demand graph and setting weights in one embodiment of the expert matching method disclosed herein, as shown below. Figure 3A The method shown includes steps S301-S305. Each step is explained below.

[0083] S301, extract the second keyword from the enterprise demand information and use it as the second entity.

[0084] S302, extract the second association between the second keywords from the enterprise demand information, and use it as the second relationship.

[0085] S303, construct a second triplet based on the second keyword and the second association, and construct an enterprise demand graph based on the second triplet; wherein, enterprise demand information includes structured, semi-structured or unstructured enterprise demand information; the second node is used to represent the second keyword, and the connection between the second nodes is used to represent the corresponding second association.

[0086] S304, Obtain the feature information of the second keyword represented by the second node; the feature information includes: the word frequency of the second keyword in the enterprise demand information, the weight information of the second keyword set by the user, or the attribute information corresponding to the second keyword, etc., and the attribute information includes information such as industry, main business, product characteristics, and technical field.

[0087] S305, determine the enterprise demand weight of the second node based on feature information.

[0088] In one embodiment, enterprise demand information can be obtained from multiple channels, such as enterprise demand information filled in on a platform. Various methods can be used to extract secondary keywords and corresponding secondary relationships from structured, semi-structured, and unstructured enterprise demand information. For example, natural language processing methods can be used to extract secondary keywords, or the Jieba word segmentation tool can be used to extract secondary keywords according to word frequency thresholds. Alternatively, enterprises can fill in technical keywords (secondary keywords) and their weights. Secondary relationships include demand status, technical demand name, demand status, demand domain, technical keywords, technical demand category, specific demand content, deliverable requirements, reward amount, specific content, cooperation method, economic and technical indicator requirements, existing equipment and technological foundation, and other requirements.

[0089] Extract secondary keywords from enterprise requirement information (e.g., requirement documents), and construct an enterprise requirement graph based on "entity - secondary keyword" and "entity - other information (secondary relationship) - secondary keyword". For example, using the Jieba word segmentation tool to segment and count word frequencies of enterprise requirement information, the enterprise requirement information is as follows:

[0090] Enterprise Technology Requirements Description: Develop a rapid, non-invasive in vitro detection technology for common bacterial infectious diseases using high-throughput sequencing, immunofluorescence detection, quantitative PCR, and nucleic acid hybridization techniques, and guide clinical trials. This includes a rapid detection technology primarily focused on common aquatic bacterial pathogens and a reagent kit. The technical specifications for this technology include the ability to simultaneously detect at least 70 common aquatic bacterial pathogens, with at least 6 molecular markers detected for each pathogen, and the entire detection process should not exceed 8 hours, with an accuracy rate >85%. It is primarily intended for the rapid detection of common bacterial pathogens in aquariums, swimming pools, and natural water bodies.

[0091] Word segmentation and word frequency statistics were performed. The segmentation results and word frequency coefficients are as follows:

[0092] [('Technology', 1.0),('Detection', 0.9252364923108787),('Bacterial', 0.6160450005521897),('Pathogens', 0.5056300789700632),('Including', 0.3749314406643248),('Water Body', 0.37245144971491917),('Technical Means', 0.24685181307012455),('Swimming Pool', 0.24523137980952675),('Hour', 0.23726090824503224),('Technical Indicators', 0.23023807455641607),('Exceeding ',0.22655555062451288),('Sequencing',0.2136247117978059),('Nucleic Acid',0.2073195195221221),('Guidance',0.207067338895654),('Hybridization',0.20674274113427507),('Development',0.20528167697798613),('Quantitative',0.2012462468478737),('Aquarium',0.19032812844043437),('Application',0.18971836657499627),('Able',0.1868911890314467)).

[0093] The word segmentation results above can be used to remove verbs and general terms, such as "able to," "application," and "hours." The enterprise demand weight of the second keyword can be set based on its frequency in the enterprise demand information; alternatively, the second keyword and its weight can be obtained from the enterprise's technical requirements; or the enterprise demand weight can be set based on the attribute information corresponding to the second keyword, including industry, main business, product characteristics, and technical field.

[0094] The second keywords extracted from the enterprise demand information, the second association relationships between the second keywords, and the enterprise demand weights of the determined second nodes are shown in Table 3 below:

[0095]

[0096] Table 3 - Enterprise Demand Weighting Table for Second Keyword

[0097] The second keywords extracted from enterprise demand information include "technology demand Y", "biomedicine", and "vaccine application", etc. The second relationships between these keywords include "industry" and "main business". Second triples are constructed based on the second keywords and second relationships, and an enterprise demand graph is built based on these second triples, as shown below. Figure 3B As shown.

[0098] In one embodiment, all second keywords are sorted based on enterprise demand weights, and a preset number of second keywords are selected as matching keywords according to the sorting results; the preset number can be set, for example, 5, 6, etc. The system then searches the expert technical capability knowledge graph for a first node that matches the matching keyword. This first node represents a first keyword that is either the matching keyword or a synonym of the matching keyword.

[0099] The matching degree is determined as follows:

[0100]

[0101] Among them, A k Let α be the enterprise demand weight of the second keyword in the k-th matching node pair. k is the expert capability weight of the first keyword in the k-th matching node pair, and n is the number of matching node pairs.

[0102] Figure 4 This is a schematic diagram illustrating the process of determining recommended experts that match enterprise needs information in one embodiment of the expert matching method disclosed herein, such as... Figure 4 The method shown includes steps S401-S404. Each step is explained below.

[0103] S401, determine the predicted workload based on enterprise demand information, and determine the expert's predicted available time based on expert information and matching degree; the predicted available time is the product of the expert's available working time and the corresponding matching degree.

[0104] S402, sort all experts according to their matching degree.

[0105] S403 determines a predetermined number of recommendation experts based on the ranking results, predicted workload, and predicted available time. The preset number can be set, for example, to 2, 3, etc.

[0106] S404: Sort the recommended experts according to the predicted availability time.

[0107] In one embodiment, based on the second keyword in the enterprise demand graph, matching keyword groups or vectors can be generated. Keyword similarity analysis at different levels is performed on the expert technical capability knowledge graph of each expert. By applying the corresponding matching algorithm, keywords that are the same as or similar to the keywords in the matching keyword groups or vectors are matched and the matching degree is calculated. The technical experts are ranked according to the matching degree, and the current availability of the experts is judged to determine the optimal recommended expert.

[0108] You can use all the secondary keywords in the enterprise demand graph as matching keywords, or you can sort all the secondary keywords based on the enterprise demand weight, and select n secondary keywords as matching keywords according to the sorting results from high to low.

[0109] Various existing methods can be used to search the expert technical capability knowledge graph for the existence of a first node that matches each matching keyword. The first node represents the first keyword as the matching keyword or a synonym of the matching keyword. This allows us to obtain the keyword matching degree of different experts under the same needs, filter out the matching experts, and recommend them to the enterprise.

[0110] In one embodiment, if a first node in the expert technical capability knowledge graph matches a second node in the enterprise demand graph, then the matching first and second nodes are considered a matching node pair. For example, as shown in Table 4 below, the second keyword "drug activity" in the enterprise demand graph is the same as the first keyword "drug activity" in Li Moufang's expert technical capability knowledge graph. Therefore, the first node corresponding to "drug activity" in Li Moufang's expert technical capability knowledge graph and the second node corresponding to "drug activity" in the enterprise demand graph are considered a matching node pair, and the matching value A1α1 = 0.9 * 1 = 0.9 for this matching node pair.

[0111]

[0112] Table 4 - Matching Information Between the Secondary Keywords of Enterprise Demand Graph Y and the Expert Technical Capability Knowledge Graphs of Experts Li Moufang and Liu Mouhao

[0113] Based on the above method, after matching the enterprise demand graph with Li Moufang's expert technical capability knowledge graph, there are four matching node pairs, with matching values ​​of 0.9, 0.45, 0.7, and 0.7 respectively. Similarly, based on the above method, after matching the enterprise demand graph with Liu Mouhao's expert technical capability knowledge graph, there are four matching node pairs, with matching values ​​of 0.25, 0.25, 0.45, and 0.7 respectively.

[0114] The matching degree between enterprise technology requirement Y and technology expert Li Moufang A is: A k Let α be the enterprise demand weight of the second keyword in the k-th matching node pair. k Let be the expert capability weight of the first keyword in the k-th matching node pair, and n be the number of matching node pairs. Similarly, the matching degree between enterprise technology requirement Y and technical expert Liu Mouhao is calculated using the same method.

[0115] The matching degree M between the enterprise demand map and Li Moufang's expert technical capability knowledge map. Y-A for: The matching degree M between the enterprise demand map and Liu Mouhao's expert technical capability knowledge map. Y-B for: As shown above, expert Li Moufang has a higher degree of matching with the company's needs. This method can be used to calculate the matching degree between other experts and the company's needs.

[0116] All experts are ranked according to their matching degree. After selecting one or more experts with the highest matching degree, the enterprise's demand information is analyzed to obtain workload-related information, such as the research cycle of projects. Based on this workload-related information, the predicted workload of the demand is determined. Expert information is parsed to obtain the expert's available working time information, such as the number of projects currently under research by the expert and the cycle of each project. Based on this available working time information, the expert's available working time is determined. The product of the expert's available working time and the corresponding matching degree is used as the predicted available time.

[0117] Based on the experts' predicted available time, it was determined whether there was time to complete the company's requirements. All experts were then ranked from highest to lowest according to their matching degree, resulting in a Top [list of experts]. k We recommend experts and sort them from longest to shortest predicted availability. Finally, we send the recommended expert information to the project manager in charge of the enterprise.

[0118] For example, based on the company's needs, the predicted workload is 100 person-days, and the project cycle is one year. Technical expert Li Moufang has 50 days of available time per year. Therefore, according to the formula: Predicted Available Time = Expert's (Actual) Available Working Time × Expert's Matching Degree, the predicted available time for the expert regarding the technical requirement is calculated. Li Moufang's predicted available time for technical requirement Y is:

[0119] If the predicted available time is greater than the predicted workload of 100 person-days, then Li Moufang will be the final recommended expert. If Li Moufang's available workload is 10 days, then the predicted available time is 27.5 days, which cannot meet the resource requirements. In the case of the same matching value, experts with higher local matching values ​​for the technical requirement keywords will be given priority. After obtaining multiple recommended experts, they will be ranked according to their respective predicted available time, and the ranking results will be sent to the user for selection.

[0120] Exemplary device

[0121] In one embodiment, such as Figure 5 As shown, this disclosure provides an expert matching device, including an expert graph construction module 51, an expert weight setting module 52, a demand graph construction module 53, a demand weight setting module 54, a graph matching processing module 55, a matching degree determination module 56, and a recommended expert determination module 57. The expert graph construction module 51 acquires expert information, extracts first entities from the expert information, determines first relationships between the first entities, and constructs an expert technical capability knowledge graph based on the first entities and the first relationships; wherein, the expert technical capability knowledge graph includes multiple first nodes used to represent the first entities.

[0122] The expert weight setting module 52 determines the level corresponding to the first node and sets the corresponding expert capability weight for the first node based on the level. The demand graph construction module 53 acquires enterprise demand information, extracts second entities from the enterprise demand information, determines the second relationships between the second entities, and constructs an enterprise demand graph based on the second entities and the second relationships; wherein, the enterprise demand graph includes: multiple second nodes used to represent the second entities. The demand weight setting module 54 determines the enterprise demand weight of the corresponding second node according to the second entity.

[0123] The graph matching processing module 55 searches the expert technical capability knowledge graph for a first node that matches a second node in the enterprise demand graph. If a match is found, the matching first and second nodes are considered a matching node pair. The matching degree determination module 56 determines the matching degree based on the expert capability weight and enterprise demand weight corresponding to the matching node pair. The expert recommendation determination module 57 determines recommended experts that match the enterprise demand information based on the matching degree.

[0124] In one embodiment, the expert graph construction module 51 extracts a first keyword from the expert information as a first entity. The expert graph construction module 51 also extracts a first association between the first keywords from the expert information as a first relationship. Based on the first keyword and the first association, the expert graph construction module 51 constructs a first triplet, and then constructs an expert technical capability knowledge graph based on the first triplet.

[0125] In the expert technical capability knowledge graph, the expert weight setting module 52 selects a first node as the root node and sets the expert capability weight of the root node. The expert weight setting module 52 then sequentially determines the level corresponding to the root node for other first nodes and sets the expert capability weight for the other first nodes based on the level.

[0126] The demand graph construction module 53 extracts second keywords from the enterprise demand information, which serve as second entities. The demand graph construction module 53 also extracts second relationships between these second keywords, which serve as second relations. Based on the second keywords and second relationships, the demand graph construction module 53 constructs second triples, and then builds an enterprise demand graph based on these second triples.

[0127] The demand weight setting module 54 obtains the feature information of the second keyword represented by the second node, and determines the enterprise demand weight of the second node based on the feature information. The feature information includes the word frequency of the second keyword in the enterprise demand information, the weight information of the second keyword set by the user, or the attribute information corresponding to the second keyword, etc.

[0128] In one embodiment, the graph matching processing module 55 sorts all the second keywords based on the enterprise demand weight. The graph matching processing module 55 selects a preset number of second keywords as matching keywords based on the sorting results, and searches the expert technical capability knowledge graph for a first node that matches the matching keywords. This first node represents a first keyword that is either a matching keyword or a synonym of a matching keyword.

[0129] Match Determination Module 56 determines the match degree Among them, A k Let α be the enterprise demand weight of the second keyword in the k-th matching node pair. k is the expert capability weight of the first keyword in the k-th matching node pair, and n is the number of matching node pairs.

[0130] The expert recommendation module 57 determines the predicted workload based on the enterprise's needs information and determines the predicted available time for each expert based on expert information and matching degree. The module 57 then ranks all experts according to their matching degree. Based on the ranking results, predicted workload, and predicted available time, the module 57 determines a predetermined number of recommended experts. Finally, the module 57 ranks the recommended experts according to their predicted available time, where the predicted available time is the product of the expert's available working time and its corresponding matching degree.

[0131] Figure 6 This is a schematic diagram of another embodiment of the expert matching device of this disclosure, as shown below. Figure 6 As shown, the expert matching device 61 includes one or more processors 611 and a memory 612.

[0132] The processor 611 may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the expert matching device 61 to perform the desired functions.

[0133] The memory 612 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, and flash memory. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 611 may execute the program instructions to implement the expert matching methods of the various embodiments of this disclosure described above and / or other desired functions. Various contents such as input signals, signal components, and noise components may also be stored in the computer-readable storage medium.

[0134] In one example, the expert matching device 61 may further include an input device 613 and an output device 614, etc., these components being interconnected via a bus system and / or other forms of connection mechanisms (not shown). Furthermore, the input device 613 may include, for example, a keyboard, a mouse, etc. The output device 614 can output various information to the outside. The output device 614 may include, for example, a monitor, speakers, a printer, and a communication network and its connected remote output devices, etc.

[0135] Of course, for the sake of simplicity, Figure 6 Only some of the components of the expert matching device 61 relevant to this disclosure are shown in this illustration; components such as buses, input / output interfaces, etc., are omitted. In addition, the expert matching device 61 may include any other suitable components depending on the specific application.

[0136] In addition to the methods and apparatus described above, embodiments of this disclosure may also be computer program products comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the expert matching methods according to various embodiments of this disclosure as described in the "Exemplary Methods" section of this specification.

[0137] Computer program products can be written in any combination of one or more programming languages ​​to perform the operations of embodiments of this disclosure. These programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0138] Furthermore, embodiments of this disclosure may also be computer-readable storage media having computer program instructions stored thereon, which, when executed by a processor, cause the processor to perform the steps in the expert matching methods according to various embodiments of this disclosure as described in the "Exemplary Methods" section above.

[0139] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples (not an exhaustive list) of a readable storage medium may include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0140] The expert matching method, device, and storage medium in the above embodiments employ a knowledge graph to construct an information system, making the recommended expert information more accurate and complete. They can comprehensively and dynamically collect relevant internal data on expert capabilities, such as papers, patent websites, and other open scientific and technological achievement platforms. Furthermore, they can openly collect relevant external data from the industry chain or related resource ecosystem, forming a richer structured knowledge ecosystem. For example, platforms like Wanfang Database and Huazhuan.com, which include scientific and technological experts, academic experts, and technology transfer platforms, provide a more comprehensive description of expert capabilities and better match them with enterprise needs.

[0141] When calculating the similarity of different entities, adjusting the weight ratios for each entity makes the focus clearer and improves the accuracy of recommendations. Matching the knowledge graph of enterprise technical needs with the knowledge graph of expert technical capabilities enables more comprehensive and accurate recommendations of technical experts. By considering the expert's successful and ongoing projects and assessing the expert's project capabilities and available time and energy, we can avoid situations where a project expert is successfully matched due to cost, lack of practical project experience, or insufficient time, but the expert does not have the time or the company does not have sufficient salary to hire them, leading to secondary or even tertiary matching issues.

[0142] A new similarity standard is adopted to improve the matching degree between enterprise technical needs and experts. Different weights are assigned based on the enterprise's industry, main business, product keywords, and the specific content of the enterprise's needs. The overall matching degree of experts is calculated in the expert knowledge graph. Technical experts are ranked according to similarity; the number of projects currently under development by each expert is assessed, workload is predicted, or the availability of time for users in the system is obtained. These factors are then used to rank experts from highest to lowest similarity to select the best-matched experts who meet the time requirements, thereby improving the user experience.

[0143] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For system embodiments, since they largely correspond to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0144] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0145] The methods and apparatus of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.

[0146] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An expert matching method, comprising: Obtain expert information, extract a first entity from the expert information and determine a first relationship between the first entities, and construct an expert technical capability knowledge graph based on the first entity and the first relationship; wherein, the expert technical capability knowledge graph includes multiple first nodes for representing the first entity; Determine the level corresponding to the first node, and set the corresponding expert capability weight for the first node based on the level; In the expert technical capability knowledge graph, a first node is selected as the root node, and the expert capability weight of the root node is set; the levels corresponding to the root node of other first nodes are determined in turn, and the expert capability weights of the other first nodes are set based on the levels; the first keyword represented by the root node includes: expert name or number; the larger the level corresponding to the first node, the lower the expert capability weight set for this first node; Obtain enterprise demand information, extract a second entity from the enterprise demand information and determine a second relationship between the second entities, and construct an enterprise demand graph based on the second entity and the second relationship; The enterprise demand graph includes: multiple second nodes representing the second entity; second keywords extracted from the enterprise demand information as the second entity; second relationships extracted from the enterprise demand information as the second relationship; second triples constructed based on the second keywords and the second relationships, and the enterprise demand graph constructed based on the second triples; the enterprise demand information includes: structured, semi-structured, or unstructured enterprise demand information; the second nodes represent the second keywords, and the connections between the second nodes represent the corresponding second relationships; The enterprise demand weight of the corresponding second node is determined based on the second entity. Specifically, the process involves obtaining feature information of the second keyword represented by the second node; determining the enterprise demand weight of the second node based on the feature information; and the feature information including: the word frequency of the second keyword in the enterprise demand information, the weight information of the second keyword set by the user, or the attribute information corresponding to the second keyword. In the expert technical capability knowledge graph, search for whether there is a first node that matches the second node of the enterprise demand graph. If there is, the matching first node and second node are taken as a matching node pair. The matching degree is determined based on the expert capability weight and enterprise demand weight corresponding to the matching node pair, and the recommended expert matching the enterprise demand information is determined based on the matching degree. Among them, the matching degree is determined. A k Let α be the enterprise demand weight of the second keyword in the k-th matching node pair. k The expert capability weight of the first keyword in the k-th matching node pair is denoted as n, where n is the number of matching node pairs. The method of determining the recommendation experts who match the enterprise's needs based on the matching degree includes: The predicted workload is determined based on the enterprise demand information, and the predicted available time of each expert is determined based on the expert information and the matching degree. All experts are ranked according to the matching degree. Based on the ranking result, the predicted workload, and the predicted available time, a predetermined number of recommended experts are determined. The recommended experts are ranked according to the predicted available time. The predicted available time is the product of the expert's available working time and the corresponding matching degree.

2. The method as described in claim 1, wherein extracting a first entity from the expert information and determining a first relationship between the first entities, and constructing an expert technical capability knowledge graph based on the first entity and the first relationship, comprises: Extract the first keyword from the expert information and use it as the first entity; Extract the first association relationship between the first keywords from the expert information, and use it as the first relationship; A first triplet is constructed based on the first keyword and the first association relationship, and the expert technical capability knowledge graph is constructed based on the first triplet. The expert information includes one or more of the following: structured, semi-structured, or unstructured introductions, projects, research, and professional information; the first node is used to represent the first keyword, and the connection between the first nodes is used to represent the corresponding first association relationship.

3. The method as described in claim 1, wherein searching in the expert technical capability knowledge graph for a first node that matches the second node of the enterprise demand graph includes: All secondary keywords are sorted based on the aforementioned enterprise demand weights; A preset number of second keywords are selected as matching keywords based on the sorting results; Search the expert technical capability knowledge graph for a first node that matches the matching keyword; wherein the first node represents the first keyword that is the matching keyword or a synonym of the matching keyword.

4. An expert matching device, comprising: An expert graph construction module is used to acquire expert information, extract a first entity from the expert information and determine a first relationship between the first entities, and construct an expert technical capability knowledge graph based on the first entity and the first relationship; wherein, the expert technical capability knowledge graph includes multiple first nodes for representing the first entity; The expert weight setting module is used to determine the level corresponding to the first node, and set the corresponding expert ability weight for the first node based on the level; wherein, in the expert technical ability knowledge graph, a first node is selected as the root node, and the expert ability weight of the root node is set; the levels corresponding to the root node of other first nodes are determined sequentially, and the expert ability weights of the other first nodes are set based on the levels; the first keyword represented by the root node includes: expert name or number; the larger the level corresponding to the first node, the lower the expert ability weight set for this first node; A demand graph construction module is used to acquire enterprise demand information, extract second entities from the enterprise demand information, determine second relationships between the second entities, and construct an enterprise demand graph based on the second entities and the second relationships. The enterprise demand graph includes: multiple second nodes representing the second entities; second keywords extracted from the enterprise demand information as the second entities; second association relationships extracted from the enterprise demand information as the second relationships; second triples constructed based on the second keywords and the second association relationships; and the enterprise demand graph constructed based on the second triples. The enterprise demand information includes structured, semi-structured, or unstructured enterprise demand information; the second nodes represent the second keywords, and the connections between the second nodes represent the corresponding second association relationships. The demand weight setting module is used to determine the enterprise demand weight of the corresponding second node based on the second entity; wherein, the feature information of the second keyword represented by the second node is obtained; the enterprise demand weight of the second node is determined based on the feature information; the feature information includes: the word frequency of the second keyword in the enterprise demand information, the weight information of the second keyword set by the user, or the attribute information corresponding to the second keyword; The graph matching processing module is used to search in the expert technical capability knowledge graph whether there is a first node that matches the second node of the enterprise demand graph. If there is, the matching first node and second node are taken as a matching node pair. The matching degree determination module is used to determine the matching degree based on the expert capability weight and enterprise demand weight corresponding to the matching node pair; The expert recommendation module is used to determine experts that match the enterprise's needs based on the matching degree; wherein, determining the matching degree... A k Let α be the enterprise demand weight of the second keyword in the k-th matching node pair. k Here, n represents the expert capability weight of the first keyword in the k-th matching node pair, and n is the number of matching node pairs. The step of determining the recommended experts matching the enterprise's demand information based on the matching degree includes: determining the prediction workload based on the enterprise's demand information, and determining the predicted available time of the experts based on the expert information and the matching degree; ranking all experts according to the matching degree; determining a predetermined number of recommended experts based on the ranking result, the prediction workload, and the predicted available time; and ranking the recommended experts according to the predicted available time. The predicted available time is the product of the expert's available working time and the corresponding matching degree.

5. An expert matching device, comprising: processor; Memory used to store the processor's executable instructions; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method described in any one of claims 1-3.

6. A computer-readable storage medium storing a computer program for performing the method according to any one of claims 1-3.

Images