Project information matching method, device and storage medium

Abstract

The present invention provides a project information matching method, device and storage medium, which relate to the field of computer technology. The method comprises: obtaining corresponding channel keywords in a channel corpus and corresponding expert keywords in an expert corpus according to demand keywords, and calculating a first correlation; selecting candidate project matching combinations according to demand keywords, channel keywords and expert keywords, and calculating a second correlation of the candidate project matching combinations; calculating a third correlation corresponding to the candidate project matching combinations according to the first correlation and the second correlation, so as to determine a target project matching combination; the method, device and storage medium of the present invention realize automatic matching between supply and demand parties and third-party transfer channels, reduce manual search and matching processes, improve matching efficiency and accuracy, improve technology transfer conversion efficiency, and improve user experience.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a project information matching method, device and storage medium. Background Art

[0002] With the development of the Internet, technology transfer has gradually been transferred to online technology transfer platforms. Online technology transfer platforms can break through the limitations of time and space, make up for the imbalance of regional development, accelerate the marketization of scientific and technological innovation, and reduce the transaction costs of technology transfer. At the same time, online technology transfer platforms can save a large amount of data, including technical information and transaction content of both technology supply and demand parties. On the one hand, this is conducive to improving the timeliness of information and reducing the opacity of information; on the other hand, large amounts of information put forward higher requirements for the matching technology and connection technology of technology transfer platforms, so as to better promote the exchange and communication of information, match supply and demand parties and third-party technology transfer channels, and promote the occurrence of technology transfer. At present, although online technology transfer platforms have a lot of data, the matching process is still through manual retrieval of information by both parties or matching of corresponding projects according to platform tag data. The matching efficiency is low and even the matching accuracy is not high, and automatic matching of technology supply and demand has not been achieved. Summary of the invention

[0003] In order to solve the above technical problems, the present disclosure is proposed. The embodiments of the present disclosure provide a project information matching method, device and storage medium.

[0004] According to a first aspect of an embodiment of the present disclosure, a project information matching method is provided, comprising: receiving project requirement information, obtaining requirement keywords from a requirement corpus based on the project requirement information; obtaining corresponding channel keywords in a channel corpus and corresponding expert keywords in an expert corpus based on the requirement keywords; calculating a first correlation between the requirement keywords, the channel keywords and the expert keywords; selecting a candidate project matching combination based on the requirement keywords, the channel keywords and the expert keywords and calculating a second correlation of the candidate project matching combination; wherein the candidate project matching combination comprises: the project requirement information, current channel information and current expert information corresponding to the project requirement information; calculating a third correlation corresponding to the candidate project matching combination based on the first correlation and the second correlation, and determining a target project matching combination from the candidate project matching combination based on the third correlation.

[0005] Optionally, historical project information of successful transfer or conversion is obtained, the demand keywords, the channel keywords and the expert keywords are extracted from the historical project information and corresponding word vectors are calculated; the demand keywords, the channel keywords and the expert keywords and the corresponding word vectors are stored in the demand corpus, the channel corpus and the expert corpus, respectively.

[0006] Optionally, the extracting the demand keywords, the channel keywords, the expert keywords and the corresponding word vectors from the historical project information includes: performing word segmentation and preprocessing on the historical project information to obtain the demand keywords, the channel keywords and the expert keywords; performing vectorization processing on the demand keywords, the channel keywords and the expert keywords respectively to obtain word vectors; respectively calculating the word frequency TF and inverse document frequency IDF of the demand keywords, the channel keywords and the expert keywords to obtain the TF-IDF values ​​corresponding to the demand keywords, the channel keywords and the expert keywords; based on a preset sorting rule and according to the TF-IDF value, respectively selecting a preset number of demand keywords, channel keywords and expert keywords.

[0007] Optionally, the demand corpus includes: a demand attribute word library and a demand target word library; the acquiring demand keywords from the demand corpus based on the project demand information includes: acquiring demand project attribute keywords and demand project target keywords from the demand attribute word library and the demand target word library respectively based on the project demand information;

[0008] Optionally, the channel corpus includes: a channel attribute word lexicon, a channel target word lexicon and a channel success word lexicon; the expert corpus includes: an expert attribute word lexicon and an expert success word lexicon; the acquiring of corresponding channel keywords in the channel corpus and corresponding expert keywords in the expert corpus according to the demand keywords includes: setting a first library group and a second library group; wherein the first library group includes: the demand attribute word lexicon, the channel attribute word lexicon and the expert attribute word lexicon; the second library group includes: the demand target word lexicon, the channel target word lexicon, the channel success word lexicon and the expert success word lexicon; acquiring corresponding channel keywords in the channel corpus and corresponding expert keywords in the expert corpus according to the demand item attribute keywords from the channel attribute word lexicon and the expert attribute word lexicon in the first library group; The corresponding channel attribute keywords and expert attribute keywords are respectively obtained from the demand project attribute keywords, the channel attribute keywords and the expert attribute keywords; at least one first keyword group is established based on the correspondence between the demand project attribute keywords, the channel attribute keywords and the expert attribute keywords; according to the demand project target keywords, the corresponding channel target project keywords, channel success project keywords and expert success project keywords are respectively obtained from the channel target word library, the channel success word library and the expert success word library in the second library group; based on the correspondence between the demand project target keywords, the channel target project keywords, the channel success project keywords and the expert success project keywords, at least one second keyword group is established.

[0009] Optionally, the calculation of the first association between the demand keywords, the channel keywords and the expert keywords includes: obtaining the word vector of each keyword in the first keyword group, and calculating the attribute association corresponding to the first keyword group based on the word vector; obtaining the word vector of each keyword in the second keyword group, and calculating the target association corresponding to the second keyword group based on the word vector; calculating the first association based on the attribute association and the target association.

[0010] Optionally, the calculation of the attribute association corresponding to the first keyword group includes: calculating the first distance between the word vectors of each two keywords among the requirement project attribute keywords, the channel attribute keywords and the expert attribute keywords, and taking the product of three first distances as the attribute association; the calculation of the target association corresponding to the second keyword group includes: calculating the second distance between the word vectors of each two keywords among the requirement project target keywords, the channel target project keywords and the expert success project keywords, and taking the product of three second distances as the first sub-target association; calculating the third distance between the word vectors of each two keywords among the requirement project target keywords, the channel success project keywords and the expert success project keywords, and taking the product of three third distances as the second sub-target association; calculating the average sum of the first sub-target association and the second sub-target association as the target association; the calculation of the first association based on the attribute association and the target association includes: calculating the first association based on all attribute associations and all target associations.

[0011] Optionally, the selecting of candidate project matching combinations based on the demand keywords, the channel keywords and the expert keywords includes: generating a demand keyword set based on the demand project attribute keywords and the demand project target keywords; generating a channel keyword set based on the channel attribute keywords, the channel target project keywords and the channel successful project keywords; generating an expert keyword set based on the expert attribute keywords and the expert successful project keywords; selecting the current channel information based on the channel keyword set; wherein the current channel information includes at least one first keyword in the channel keyword set; selecting the current expert information based on the expert keyword set; wherein the current expert information includes at least one second keyword in the expert keyword set; generating the candidate project matching combination based on the project demand information, the current channel information and the current expert information.

[0012] Optionally, calculating the second correlation of the candidate item matching combination includes: obtaining all first keywords in the channel keyword set contained in the current channel information; determining the corresponding first correlation coefficient based on the vocabulary to which the first keyword belongs; calculating the first matching correlation of the current channel information based on all first correlation coefficients; obtaining all second keywords in the expert keyword set contained in the current expert information; determining the corresponding second correlation coefficient based on the vocabulary to which the second keyword belongs; calculating the second matching correlation of the current expert information based on all second correlation coefficients; calculating the second correlation based on the first matching correlation and the second matching correlation.

[0013] Optionally, calculating a third degree of association corresponding to the candidate item matching combination based on the first degree of association and the second degree of association, and determining a target item matching combination from the candidate item matching combination based on the third degree of association includes: taking the sum of the first degree of association and the second degree of association, or the product of the first degree of association and the second degree of association, as the third degree of association; and determining the candidate item matching combination corresponding to the largest third degree of association as the target item matching combination.

[0014] According to a second aspect of an embodiment of the present disclosure, a project information matching device is provided, comprising: a first keyword acquisition module, used to receive project demand information, and acquire demand keywords from a demand corpus based on the project demand information; a second keyword acquisition module, used to acquire corresponding channel keywords in a channel corpus and corresponding expert keywords in an expert corpus based on the demand keywords; a first correlation calculation module, used to calculate a first correlation between the demand keywords, the channel keywords and the expert keywords; a second correlation calculation module, used to select a candidate project matching combination and calculate a second correlation of the candidate project matching combination based on the demand keywords, the channel keywords and the expert keywords; wherein the candidate project matching combination includes: the project demand information, current channel information and current expert information corresponding to the project demand information; a target project matching determination module, used to calculate a third correlation corresponding to the candidate project matching combination based on the first correlation and the second correlation, and determine a target project matching combination from the candidate project matching combination based on the third correlation.

[0015] Optionally, a corpus construction module is used to obtain historical project information that has been successfully transferred or converted, extract the demand keywords, the channel keywords and the expert keywords from the historical project information and calculate corresponding word vectors; and store the demand keywords, the channel keywords and the expert keywords and the corresponding word vectors in the demand corpus, the channel corpus and the expert corpus, respectively.

[0016] Optionally, the corpus construction module includes: a keyword processing unit, used to perform word segmentation and preprocessing on the historical project information to obtain the demand keywords, the channel keywords and the expert keywords; respectively vectorize the demand keywords, the channel keywords and the expert keywords to obtain word vectors; a keyword selection unit, used to respectively calculate the term frequency TF and inverse document frequency IDF of the demand keywords, the channel keywords and the expert keywords to obtain the TF-IDF values ​​corresponding to the demand keywords, the channel keywords and the expert keywords; based on a preset sorting rule and according to the TF-IDF value, respectively select a preset number of demand keywords, channel keywords and expert keywords.

[0017] Optionally, the demand corpus includes: a demand attribute word library and a demand target word library; the first keyword acquisition module is used to acquire demand project attribute keywords and demand project target keywords from the demand attribute word library and the demand target word library respectively based on the project demand information.

[0018] Optionally, the channel corpus includes: a channel attribute word lexicon, a channel target word lexicon and a channel success word lexicon; the expert corpus includes: an expert attribute word lexicon and an expert success word lexicon; the second keyword acquisition module includes: a keyword matching unit, used to set up a first library group, and obtain corresponding channel attribute keywords and expert attribute keywords from the channel attribute word lexicon and the expert attribute word lexicon in the first library group according to the demand item attribute keywords; the first library group includes: the demand attribute word lexicon, the channel attribute word lexicon and the expert attribute word lexicon; a phrase establishment unit, used to match the corresponding keywords of the demand item attribute keywords, the channel attribute keywords and the expert attribute keywords to the keyword system, establish at least one first keyword group; the keyword matching unit is used to set up a second library group, and obtain the corresponding channel target project keywords, channel success project keywords and expert success project keywords from the channel target word library, the channel success word library and the expert success word library in the second library group according to the demand project target keyword; the second library group includes: a demand target word library, a channel target word library, a channel success word library and an expert success word library; the phrase establishment unit is used to establish at least one second keyword group based on the correspondence between the demand project target keywords, the channel target project keywords, the channel success project keywords and the expert success project keywords.

[0019] Optionally, the first association calculation module includes: an attribute association calculation unit, used to obtain the word vector of each keyword in the first keyword group, and calculate the attribute association corresponding to the first keyword group based on the word vector; a target association calculation unit, used to obtain the word vector of each keyword in the second keyword group, and calculate the target association corresponding to the second keyword group based on the word vector; and a comprehensive association calculation unit, used to calculate the first association based on the attribute association and the target association.

[0020] Optionally, the attribute association calculation unit is used to calculate the first distance between the word vectors of every two keywords among the requirement item attribute keywords, the channel attribute keywords and the expert attribute keywords, and take the product of three first distances as the attribute association; the target association calculation unit is used to calculate the second distance between the word vectors of every two keywords among the requirement item target keywords, the channel target project keywords and the expert success project keywords, and take the product of three second distances as the first sub-target association; calculate the third distance between the word vectors of every two keywords among the requirement item target keywords, the channel success project keywords and the expert success project keywords, and take the product of three third distances as the second sub-target association; calculate the average sum of the first sub-target association and the second sub-target association as the target association; the comprehensive association calculation unit is used to calculate the first association based on all attribute associations and all target associations.

[0021] Optionally, the second association calculation module includes: a project matching combination acquisition unit, used to generate a demand keyword set based on the demand project attribute keywords and the demand project target keywords; generate a channel keyword set based on the channel attribute keywords, the channel target project keywords and the channel successful project keywords; generate an expert keyword set based on the expert attribute keywords and the expert successful project keywords; select the current channel information based on the channel keyword set; wherein the current channel information contains at least one first keyword in the channel keyword set; select the current expert information based on the expert keyword set; wherein the current expert information contains at least one second keyword in the expert keyword set; generate the candidate project matching combination based on the project demand information, the current channel information and the current expert information.

[0022] Optionally, the second association calculation module includes: a matching association calculation unit, used to obtain all first keywords in the channel keyword set contained in the current channel information; determine the corresponding first association coefficient based on the vocabulary to which the first keyword belongs; calculate the first matching association of the current channel information based on all the first association coefficients; obtain all second keywords in the expert keyword set contained in the current expert information, and determine the corresponding second association coefficient based on the vocabulary to which the second keyword belongs; calculate the second matching association of the current expert information based on all the second association coefficients; calculate the second association based on the first matching association and the second matching association.

[0023] Optionally, the target project matching determination module is used to take the sum of the first degree of association and the second degree of association or the product of the first degree of association and the second degree of association as the third degree of association; and determine the candidate project matching combination corresponding to the largest third degree of association as the target project matching combination.

[0024] According to a third aspect of an embodiment of the present disclosure, there is provided a project information matching device, comprising: a processor; a memory for storing instructions executable by the processor; the processor is configured to read the executable instructions from the memory and execute the instructions to implement the method described above.

[0025] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the storage medium stores a computer program, and the computer program is used to execute the above method.

[0026] Based on the project information matching method, device and storage medium provided by the above embodiments of the present disclosure, the corresponding channel keywords in the channel corpus and the corresponding expert keywords in the expert corpus are obtained according to the demand keywords, and the first correlation is calculated; according to the demand keywords, the channel keywords and the expert keywords, the candidate project matching combination is selected and the second correlation of the candidate project matching combination is calculated; the third correlation corresponding to the candidate project matching combination is calculated according to the first correlation and the second correlation to determine the target project matching combination; the automatic matching of the supply and demand parties with the third-party transfer channels is realized, the manual search and matching process is reduced, the matching efficiency and accuracy are improved, the technology transfer conversion efficiency is improved, and the user experience is improved.

[0027] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] By describing the embodiments of the present disclosure in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present disclosure will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. In the drawings, the same reference numerals generally represent the same components or steps.

[0029] Figure 1 A flowchart of an embodiment of the project information matching method disclosed in the present invention;

[0030] Figure 2 A schematic diagram of a process of establishing a corpus in one embodiment of the project information matching method disclosed in the present invention;

[0031] Figure 3 A schematic diagram of a flow chart of extracting keywords in one embodiment of the project information matching method disclosed herein;

[0032] Figure 4A A schematic diagram of obtaining demand keywords from a demand corpus based on project demand information;

[0033] Figure 4B A schematic diagram of a flow chart of calculating a first degree of association in one embodiment of the project information matching method disclosed herein;

[0034] Figure 5 A schematic diagram of a flow chart of calculating a second degree of association in one embodiment of the project information matching method disclosed herein;

[0035] Figure 6 A schematic diagram of a process of determining a target project matching combination in one embodiment of the project information matching method disclosed herein;

[0036] Figure 7 A schematic diagram of modules of an embodiment of the project information matching device disclosed in the present invention;

[0037] Figure 8 A schematic diagram of modules of another embodiment of the project information matching device disclosed in the present invention;

[0038] Fig. 9 A schematic diagram of a corpus building module in one embodiment of the project information matching device disclosed herein;

[0039] Fig.10 A schematic diagram of a second keyword acquisition module in one embodiment of the project information matching device disclosed herein;

[0040] Fig.11 A schematic diagram of a first correlation degree calculation module in one embodiment of the project information matching device disclosed in the present invention;

[0041] Fig.12 A schematic diagram of a second correlation degree calculation module in one embodiment of the project information matching device disclosed herein;

[0042] Fig.13 It is a structural schematic diagram of another embodiment of the project information matching device disclosed in the present invention. DETAILED DESCRIPTION

[0043] The exemplary embodiments according to the present disclosure will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described here.

[0044] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0045] Those skilled in the art can understand that the terms "first" and "second" in the embodiments of the present disclosure are only used to distinguish different steps, devices or modules, etc., and neither represent any specific technical meaning nor indicate the necessary logical order between them.

[0046] It should also be understood that in the embodiments of the present disclosure, “plurality” may refer to two or more than two, and “at least one” may refer to one, two, or more than two.

[0047] It should also be understood that any component, data or structure mentioned in the embodiments of the present disclosure can generally be understood as one or more, unless explicitly limited or otherwise indicated in the context.

[0048] In addition, the term "and / or" in this disclosure is only a description of the association relationship of associated objects, indicating that there may be three relationships, such as A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this disclosure generally indicates that the associated objects before and after are in an "or" relationship.

[0049] It should also be understood that the description of the various embodiments in the present disclosure focuses on the differences between the various embodiments, and the same or similar aspects thereof can be referenced to each other, and for the sake of brevity, they will not be described one by one.

[0050] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0051] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0052] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0053] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0054] The embodiments of the present disclosure can be applied to electronic devices such as terminal devices, computer systems, servers, etc., which can operate with many other general or special 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 personal computers, small computer systems, large computer systems, and distributed cloud computing technology environments including any of the above systems, etc.

[0055] Electronic devices such as terminal devices, computer systems, servers, etc. can be described in the general context of computer system executable instructions (such as program modules) executed by computer systems. Generally, program modules can include routines, programs, target programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in a distributed cloud computing environment. In a distributed cloud computing environment, tasks can be performed by remote processing devices linked through a communication network. In a distributed cloud computing environment, program modules can be located on local or remote computing system storage media including storage devices.

[0056] Exemplary Methods

[0057] Figure 1 A flowchart of an embodiment of the project information matching method disclosed in the present invention is shown in FIG. Figure 1 The method shown includes steps: S101-S105. Each step is described below.

[0058] S101, receiving project requirement information, and acquiring requirement keywords from a requirement corpus based on the project requirement information.

[0059] In one embodiment, after the user logs in to the technology platform, he / she fills in the relevant project demand information. The project demand information may be project information that requires technology transfer, transaction, etc. The project demand information includes information such as the demand enterprise, demand status, technology demand name, demand field, technology keywords, technology demand category, and specific demand content. A demand corpus is established in advance, and demand keywords are stored in the demand corpus.

[0060] S102, acquiring corresponding channel keywords in the channel corpus and corresponding expert keywords in the expert corpus according to the demand keywords.

[0061] In one embodiment, a channel corpus and an expert corpus are pre-established, and channel keywords and expert keywords are stored in the channel corpus and the expert corpus respectively. A variety of existing matching methods can be used to match channel keywords and expert keywords similar to the demand keywords in the channel corpus and the expert corpus respectively according to the demand keywords; the matching method can be an existing semantic analysis matching method, etc.

[0062] S103, calculating a first correlation between the demand keyword, the channel keyword and the expert keyword.

[0063] S104, selecting a candidate item matching combination according to the demand keyword, the channel keyword and the expert keyword and calculating a second correlation degree of the candidate item matching combination.

[0064] In one embodiment, the candidate project matching combination includes: project demand information, current channel information corresponding to the project demand information, and current expert information. The current channel information is the channel information currently in a valid state, and the channel information can be technology broker information, etc.; the current expert information can be the expert information currently in a valid state, and the expert information can be technical expert information, etc.

[0065] For example, channel information is the information of technology brokers in the transfer and conversion channels, including personal resume, professional expertise, personal performance, etc. Expert information includes basic information, technical fields, technical keywords, main work and study experience, main research direction, project experience, etc.

[0066] S105 , calculating a third degree of association corresponding to the candidate item matching combination according to the first degree of association and the second degree of association, and determining a target item matching combination from the candidate item matching combination based on the third degree of association.

[0067] In one embodiment, after determining the target project matching combination, the current channel information and current expert information in the target project matching combination are obtained, and the current channel information and current expert information are pushed to the user who publishes the project demand information, so as to realize the automatic matching of the technology transfer supply and demand parties with the transfer channel, and recommend the technology brokers and experts corresponding to the matching results to the user.

[0068] Figure 2 FIG. 1 is a flow chart of establishing a corpus in one embodiment of the project information matching method disclosed in the present invention. Figure 2 The method shown includes steps: S201-S202. Each step is described below.

[0069] S201, obtaining historical project information that has been successfully transferred or converted, extracting demand keywords, channel keywords and expert keywords from the historical project information and calculating corresponding word vectors.

[0070] In one embodiment, the successfully transferred or converted historical project information includes the successfully transferred or converted historical project demand information, and the channel information and expert information corresponding to the historical project demand information. After a project demand information is successfully transferred or converted, the related information is stored.

[0071] S202, storing the demand keywords, channel keywords, expert keywords and corresponding word vectors in the demand corpus, channel corpus and expert corpus respectively.

[0072] Figure 3 FIG. 1 is a flowchart of extracting keywords in one embodiment of the project information matching method disclosed in the present invention. Figure 3 The method shown includes steps: S301-S304. Each step is described below.

[0073] S301, perform word segmentation and preprocessing on historical project information to obtain demand keywords, channel keywords and expert keywords.

[0074] S302 performs vectorization processing on demand keywords, channel keywords and expert keywords respectively to obtain word vectors.

[0075] S303, respectively calculating the word frequency TF and the inverse document frequency IDF of the demand keywords, channel keywords and expert keywords, and obtaining the TF-IDF values ​​corresponding to the demand keywords, channel keywords and expert keywords.

[0076] S304, based on the preset sorting rules and according to the TF-IDF value, select a preset number of demand keywords, channel keywords and expert keywords respectively. The preset number can be set, for example, to 300, etc., and the 300 demand keywords, channel keywords and expert keywords with the highest TF-IDF values ​​are stored in the demand corpus, channel corpus and expert corpus respectively.

[0077] You can use a variety of existing methods to segment and preprocess historical project information, obtain word vectors and TF-IDF values. For example, obtain historical project information that has been successfully transferred or converted. The historical project information is text data. Segment and preprocess the historical project information to obtain multiple words. Use the existing Word2vec Skip-gram model to vectorize multiple words and obtain the word vector of each word. Calculate the word frequency TF and inverse document frequency IDF of each word to obtain its TF-IDF value. Arrange the TF-IDF values ​​of all words in descending order, select the topK words as keywords, K is the number of selected keywords, such as 300. The specific calculation formula for TF-IDF of term w is as follows:

[0078] TF-IDF w =TF w *IDF w (1-1);

[0079]

[0080]

[0081] In one embodiment, the demand corpus includes: a demand attribute word library and a demand target word library; the demand keywords include: demand item attribute keywords and demand item target keywords; the demand item attribute keywords and demand item target keywords are stored in the demand attribute word library and the demand target word library respectively.

[0082] The channel corpus includes: channel attribute word lexicon, channel target word lexicon and channel success word lexicon; channel keywords include: channel attribute keywords, channel target project keywords, channel success project keywords; channel attribute keywords, channel target project keywords, channel success project keywords are stored in the channel attribute word lexicon, channel target word lexicon and channel success word lexicon respectively.

[0083] The expert corpus includes: an expert attribute word database and an expert success word database; expert keywords include expert attribute keywords and expert success project keywords; expert attribute keywords and expert success project keywords are stored in the expert attribute word database and the expert success word database respectively.

[0084] The demand project attribute keywords, channel attribute keywords and expert attribute keywords are used to represent the fixed attribute characteristics of the demand project, channel (technology broker), technical field, region, unit or unit type, technology keywords, etc. The demand project target keywords and channel target word library are keywords that represent the content of the project that the demand project and channel (technology broker) want to complete or connect with. The channel success project keywords and expert success project keywords are keywords that represent the projects that the channel (technology broker) and expert have successfully completed or connected with in the past.

[0085] In one embodiment, based on the project requirement information, the requirement project attribute keywords and the requirement project target keywords are obtained from the requirement attribute word library and the requirement target word library, respectively. For example, the project requirement information can be segmented using an existing method, and the obtained words can be matched with the keywords in the requirement attribute word library and the requirement target word library using an existing matching method to obtain the requirement project attribute keywords and the requirement project target keywords.

[0086] Set up a first library group and a second library group; the first library group includes: demand attribute word library, channel attribute word library and expert attribute word library; the second library group includes: demand target word library, channel target word library, channel success word library and expert success word library.

[0087] The keywords under the first group are easier to extract and have higher accuracy, while the keyword extraction accuracy under the second group is often lower than that of the first group, and its content is more redundant. If the first group and the second group are mixed for calculation, the matching result is often inaccurate.

[0088] According to the demand item attribute keywords, corresponding channel attribute keywords and expert attribute keywords are respectively obtained from the channel attribute word library and the expert attribute word library in the first library group, and at least one first keyword group is established based on the correspondence between the demand item attribute keywords, the channel attribute keywords and the expert attribute keywords.

[0089] According to the demand project target keywords, corresponding channel target project keywords, channel success project keywords and expert success project keywords are respectively obtained from the channel target word library, channel success word library and expert success word library in the second library group, and based on the correspondence between the demand project target keywords, channel target project keywords, channel success project keywords and expert success project keywords, at least one second keyword group is established.

[0090] In one embodiment, as shown in Figure 4, when a user publishes a new demand, the name of the demand is demand project A. Based on the project demand information such as the technical field, region, unit, technical keywords, etc. of demand project A, the demand project attribute keywords are matched from the demand attribute word library; based on the project demand information in the specific content text of demand project A, the demand project target keywords are matched from the demand target word library.

[0091] For example, three demand item attribute keywords are matched from the demand attribute word library, namely: "automatic welding", "automatic flaw detection", and "fusion equipment"; four demand item target keywords are matched from the demand target word library, namely: "automatic welding", "automatic flaw detection", "cable fusion", and "fusion equipment". The demand item attribute keywords and the demand item target keywords can be the same or different.

[0092] According to the three demand item attribute keywords "automatic welding", "automatic flaw detection", and "welding equipment", one or more corresponding channel attribute keywords and one or more expert attribute keywords are obtained from the channel attribute word library and the expert attribute word library in the first library group, respectively, and at least one first keyword group is established based on the correspondence between the demand item attribute keywords, the channel attribute keywords, and the expert attribute keywords. The existing multiple keyword matching methods can be used to obtain the corresponding channel attribute keywords and expert attribute keywords from the channel attribute word library and the expert attribute word library in the first library group.

[0093] For example, for the demand item attribute keyword "automatic welding", a corresponding channel attribute keyword "acoustic flaw detection" and an expert attribute keyword "acoustic welding" are obtained from the channel attribute word library and the expert attribute word library in the first library group respectively to establish the first keyword group {"automatic welding", "acoustic flaw detection", "acoustic flaw detection"}.

[0094] If the attribute keyword of the required item is "automatic welding", the corresponding multiple channel attribute keywords and multiple expert attribute keywords are respectively obtained from the channel attribute word library and the expert attribute word library in the first library group, then based on the correspondence between "automatic welding" and the multiple channel attribute keywords and the multiple expert attribute keywords, multiple first keyword groups in the above format are established, and the first keyword of each first keyword group is "automatic welding".

[0095] Based on the same method, one or more first keyword groups are respectively established for the other two demand item attribute keywords "automatic welding", "automatic flaw detection" and "welding equipment".

[0096] The channel attribute keywords and expert attribute keywords corresponding to different demand item attribute keywords can be the same or different. For example, if the channel attribute keywords and expert attribute keywords corresponding to the three demand item attribute keywords "automatic welding", "automatic flaw detection", and "welding equipment" are all one and the same, then a total of three first keyword groups are generated: {"automatic welding", "sonic flaw detection", "sonic flaw detection"}, {"automatic flaw detection", "sonic flaw detection", "sonic flaw detection"} and {"welding equipment", "sonic flaw detection", "sonic flaw detection"}.

[0097] According to the four demand project target keywords "automatic welding", "automatic flaw detection", "cable fusion" and "fusion equipment", the corresponding one or more channel target project keywords, one or more channel success project keywords and one or more expert success project keywords are obtained from the channel target word library, channel success word library and expert success word library in the second library group respectively.

[0098] For example, based on the demand project target keyword "automatic welding", obtain the corresponding channel target project keyword "radiographic flaw detection", channel success word library and expert success word library in the second library group respectively; establish a second keyword group {"automatic welding", "radiographic flaw detection", "acoustic flaw detection", "fiber fusion"}.

[0099] If the project target keyword "automatic welding" is required, the corresponding multiple channel target project keywords, multiple channel success project keywords and multiple expert success project keywords are respectively obtained from the channel target word library, channel success word library and expert success word library in the second library group, then based on the correspondence between "automatic welding" and the multiple channel target project keywords, the multiple channel success project keywords and the multiple expert success project keywords, multiple second keyword groups in the above format are established, and the first keyword of each second keyword group is "automatic welding".

[0100] Based on the same method, one or more second keyword groups are established for the other three demand project target keywords “automatic flaw detection”, “cable fusion”, and “fusion splicing equipment”. The channel target project keywords, channel success project keywords, and expert success project keywords corresponding to different demand project target keywords can be the same or different. For example, if the channel target project keywords, channel success project keywords, and expert success project keywords corresponding to the four demand project target keywords “automatic welding”, “automatic flaw detection”, “cable fusion”, and “fusion splicing equipment” are all one and the same, then a total of four second keyword groups are generated: {“automatic welding”, “X-ray flaw detection”, “Acoustic flaw detection”, “Fiber optic fusion”},

[0101] {“Automatic flaw detection”, “Radiographic flaw detection”, “Acoustic flaw detection”, “Fiber optic fusion”}, {“Cable fusion”, “Radiographic flaw detection”, “Acoustic flaw detection”, “Fiber optic fusion”}, {“Fusion equipment”, “Radiographic flaw detection”, “Acoustic flaw detection”, “Fiber optic fusion”}.

[0102] Figure 4B FIG4 is a flow chart of calculating the first correlation degree in one embodiment of the project information matching method disclosed in the present invention, and the method shown in FIG4 includes steps: S401 to S403. Each step is described below.

[0103] S401, obtaining a word vector of each keyword in the first keyword group, and calculating the attribute association corresponding to the first keyword group based on the word vector.

[0104] S402, obtaining a word vector of each keyword in the second keyword group, and calculating a target relevance corresponding to the second keyword group based on the word vector.

[0105] S403: Calculate a first correlation degree based on the attribute correlation degree and the target correlation degree.

[0106] There are many methods for calculating the attribute association corresponding to the first keyword group. For example, the first distance between the word vectors of every two keywords in the demand item attribute keywords, channel attribute keywords and expert attribute keywords is calculated, and the product of the three first distances is used as the attribute association.

[0107] There are many methods for calculating the target relevance corresponding to the second keyword group. For example, the second distance between the word vectors of every two keywords in the demand project target keywords, channel target project keywords, and expert success project keywords is calculated, and the product of the three second distances is used as the first sub-target relevance.

[0108] Calculate the third distance between the word vectors of every two keywords in the demand project target keywords, channel success project keywords, and expert success project keywords, take the product of the three third distances as the second sub-goal relevance, and calculate the average sum of the first sub-goal relevance and the second sub-goal relevance as the target relevance.

[0109] A variety of methods can be used to calculate the first degree of association based on the attribute degree of association and the target degree of association. For example, based on all attribute degrees of association and all target degrees of association, a comprehensive average sum is calculated to obtain the first degree of association.

[0110] In one embodiment, the Word2vec tool is used in advance to calculate the vector form of demand keywords, channel keywords and expert keywords using the Skip-gram model, i.e., word vectors, which are stored in the demand corpus, channel corpus and expert corpus respectively. For example, the cosine similarity of each keyword in the keyword set A of the demand corpus, the keyword set B of the channel corpus and the keyword set C of the expert corpus is calculated to obtain a similarity matrix, and this similarity is the correlation between the keywords:

[0111]

[0112]

[0113]

[0114] Among them, the calculation method of cosine similarity is: v i 、v j 、v k Keyword A i , B j , C k The word vector (i.e. TF-IDF value) of keyword set A, B and C are represented by m, n and l respectively.

[0115] Calculate the first distance between the word vectors of every two keywords in the demand project attribute keywords, channel attribute keywords and expert attribute keywords. The first distance can be the cosine similarity (cosine distance) between the word vectors; calculate the second distance between the word vectors of every two keywords in the demand project target keywords, channel target project keywords and expert success project keywords. The second distance can also be the cosine similarity (cosine distance) of the word vectors.

[0116] For example, the required project attribute keyword set t 1 Including three keywords "automatic welding", "automatic flaw detection", and "welding equipment". For each demand item attribute keyword, the corresponding channel attribute keyword set is obtained from the channel attribute word library and the expert attribute word library in the first library group. 2 and expert attribute keyword t 3 .

[0117] Based on each demand item attribute keyword and channel attribute keyword set t 2 and expert attribute keyword t 3 The corresponding relationship between them is used to establish the first keyword group.

[0118] The number of the first keyword group established based on a demand item attribute keyword is: 2Number of channel attribute keywords in *t 3 The number of expert attribute keywords in the; add the number of first keyword groups established based on all demand item attribute keywords to obtain the total number of first keyword groups n 属性 .

[0119] For example, for t 1 Generate multiple first keyword groups, including three first keyword groups: {"automatic welding", "acoustic flaw detection", "acoustic flaw detection"}, {"automatic flaw detection", "acoustic flaw detection", "acoustic flaw detection"} and {"welding equipment", "acoustic flaw detection", "acoustic flaw detection"}.

[0120] t 1 The first distance between the three keywords in and the channel attribute keyword "acoustic flaw detection" is [0.94, 0.92, 0.95]; 1 The first distance between the three keywords in and the expert attribute keyword "acoustic welding" is [0.89, 0.95, 0.90]. The matching result between the channel attribute keyword "acoustic flaw detection" and the expert attribute keyword "acoustic welding" is [0.94].

[0121] The following takes a group of keyword associations in the first keyword group {"automatic welding", "sonic flaw detection", "sonic flaw detection"} as an example to illustrate the calculation of the attribute association corresponding to the first keyword group.

[0122] Calculate the attribute association degree corresponding to the phrase {"automatic welding", "sonic flaw detection", "sonic flaw detection"}, that is, the association degree r 11 For: r 11 [(Automatic welding) 需求项目属性关键词 ,(Acoustic flaw detection) 渠道属性关键词 ,(Acoustic Welding) 专家属性关键词 ]=r[(Automatic welding) 需求项目属性关键词 ,(Acoustic flaw detection) 渠道属性关键词 ]*r[(Acoustic flaw detection) 渠道属性关键词 ,(Acoustic Welding) 专家属性关键词 ]*r[(Sonic welding) 专家属性关键词 ,(Automatic welding) 需求项目属性关键词 ];

[0123] in,

[0124] r[(Automatic welding) 需求项目属性关键词 ,(Acoustic flaw detection) 渠道属性关键词 ]

[0125] =S 需求项目属性关键词，渠道属性关键词 ("Automatic welding", "Acoustic flaw detection")

[0126] Among them, S 需求项目属性关键词，渠道属性关键词("automatic welding", "sonic flaw detection") is the first distance between the corresponding keywords, that is, the cosine similarity between the keywords.

[0127] Based on the same method, the attribute associations corresponding to the remaining first keyword groups can be calculated, and the number of the obtained attribute associations is the total number of the first keyword groups.

[0128] Requirement project target keyword set t 4 Including four keywords {"automatic fusion", "radio flaw detection", "sonic flaw detection", "optical fiber fusion"}. For each demand project target keyword, the corresponding channel target project keyword set is obtained from the channel target word word library, channel success word word library and expert success word word library in the second library group. 5 , Channel Success Project Keyword Collection 6 And expert successful project keyword collection 7 .

[0129] Based on each demand project target keyword and channel target project keyword set t 5 , Channel Success Project Keyword Collection 6 And expert successful project keyword collection 7 The corresponding relationship between them is used to establish the second keyword group.

[0130] The number of second keyword groups established based on a target keyword of a demand project is: 5 Number of channel target project keywords in *t 6 The number of successful channel project keywords in *t 7 The number of successful project keywords of experts in the project; add the number of second keyword groups established based on all the target keywords of the required projects to obtain the total number of second keyword groups n 目标 .

[0131] Calculate the target relevance r of a group of keywords {“automatic fusion”, “radiographic flaw detection”, “sonic flaw detection”, “optical fiber fusion”} in the second keyword group 12 for:

[0132]

[0133] in,

[0134] r[(Automatic welding) 需求项目目标关键词 ,(Radiographic flaw detection) 渠道目标项目关键词 ,(Fiber optic fusion splicing) 专家成功项目关键词 ] is the first sub-goal relevance,

[0135] r[(Automatic welding) 需求项目目标关键词 ,(Acoustic flaw detection) 渠道成功项目关键词 ,(Fiber optic fusion splicing)专家成功项目关键词 ] is the second sub-goal association degree, and the specific calculation method of the first sub-goal association degree and the second sub-goal association degree is the same as the calculation method of the attribute association degree.

[0136] Based on the same method, the target relevances of the remaining second keyword groups are calculated respectively, and the number of the obtained target relevances is the total number of the second keyword groups.

[0137] Calculate the first correlation r based on all attribute correlations and all target correlations 1 For example, all attribute associations and all target associations can be added together to obtain the first association r 1 :

[0138]

[0139] Among them, t a,b is the keyword group in the demand database corresponding to the project demand information, t a,b By the keyword group t corresponding to the demand attribute a , and the keyword t corresponding to the demand target b Composition; i1 (t a ) is the attribute association degree corresponding to the i-th first keyword group, r i2 (t b ) is the target relevance corresponding to the i-th second keyword group; a is the first library group, b is the second library group; n 属性 is the number of all first keyword groups, n 目标 is the number of all second keyword groups.

[0140] In one embodiment, a demand keyword set is generated based on demand project attribute keywords and demand project target keywords. A channel keyword set is generated based on channel attribute keywords, channel target project keywords and channel successful project keywords. An expert keyword set is generated based on expert attribute keywords and expert successful project keywords.

[0141] Select the current channel information based on the channel keyword set; wherein the current channel information includes at least one first keyword in the channel keyword set. Select the current expert information based on the expert keyword set; wherein the current expert information includes at least one second keyword in the expert keyword set. Generate a candidate project matching combination based on the project demand information, the current channel information and the current expert information.

[0142] For example, based on the demand project attribute keywords and demand project target keywords, the demand keyword set {"automatic fusion", "automatic flaw detection", "fusion equipment", "cable fusion"} is generated. Based on the channel attribute keywords, channel target project keywords and channel success project keywords, the channel keyword set {"acoustic flaw detection", "radiographic flaw detection"} is generated. Based on the expert attribute keywords and expert success project keywords, the expert keyword set {"acoustic fusion", "fiber fusion"} is generated.

[0143] Based on the channel keyword set, current channel information B2 and B2 are selected; wherein the current channel information B2 and B2 contain at least one first keyword of {“acoustic flaw detection”, “radiographic flaw detection”}. Based on the expert keyword set, current expert information C1 and C2 are selected, wherein the current expert information C1 and C2 contain at least one second keyword of {“acoustic fusion”, “optical fiber fusion”}.

[0144] Generate candidate project matching combinations based on project demand information A, current channel information B2 and B2, and current expert information C1 and C2. Since the project demand information A is the same, there are four candidate project matching combinations, namely: {project demand information A, current channel information B1, current expert information C1}, {project demand information A, current channel information B1, current expert information C2}, {project demand information A, current channel information B2, current expert information C1}, {project demand information A, current channel information B2, current expert information C2}.

[0145] Figure 5 FIG. 1 is a flow chart of calculating the second correlation degree in one embodiment of the project information matching method disclosed in the present invention. Figure 5 The method shown includes steps: S501-S507. Each step is described below.

[0146] S501, obtaining all first keywords in a channel keyword set included in current channel information.

[0147] S502: Determine a corresponding first association coefficient based on the vocabulary to which the first keyword belongs.

[0148] S503: Calculate the first matching correlation degree of the current channel information based on all the first correlation coefficients.

[0149] S504: Acquire all second keywords in the expert keyword set included in the current expert information.

[0150] S505: Determine a corresponding second association coefficient based on the vocabulary to which the second keyword belongs.

[0151] S506: Calculate the second matching correlation degree of the current expert information based on all the second correlation coefficients.

[0152] S507: Calculate a second correlation degree according to the first matching correlation degree and the second matching correlation degree.

[0153] In one embodiment, all the first keywords in {“sound wave flaw detection”, “radio flaw detection”} contained in the current channel information B1 are “sound wave flaw detection”, that is, only one first keyword is contained. All the first keywords in {“sound wave flaw detection”, “radio flaw detection”} contained in the current channel information B2 are “sound wave flaw detection”, “radio flaw detection”, and two first keywords are contained.

[0154] The word library to which "sonic wave flaw detection" belongs is the channel target word library, and the word library to which "radio flaw detection" belongs is the channel success word library. The first correlation coefficient of the keywords in the channel target word library is preset to 0.5, and the first correlation coefficient of the keywords in the channel success word library is preset to 0.7. Different word libraries have different first correlation coefficients for keywords in the word library, which can be set according to importance.

[0155] The first matching correlation degree of the current channel information B1 is calculated to be 0.5, and the first matching correlation degree of the current channel information B2 is calculated to be 0.5+0.7=1.2.

[0156] All the second keywords in {"acoustic welding", "optical fiber welding"} contained in the current expert information C1 are obtained as "acoustic welding", and all the second keywords in {"acoustic welding", "optical fiber welding"} contained in the current expert information C2 are obtained as "acoustic welding", "optical fiber welding".

[0157] The word library of "sonic welding" is the expert attribute word library, and the word library of "optical fiber welding" is the expert success word library. The second correlation coefficient of the keywords in the expert attribute word library is preset to 0.4, and the second correlation coefficient of the keywords in the expert success word library is preset to 0.6.

[0158] The second matching correlation degree of the current expert information C1 is calculated to be 0.4, and the second matching correlation degree of the current expert information C2 is calculated to be 0.4+0.6=1.0.

[0159] The second correlation degree can be calculated by adding the first matching correlation degree and the second matching correlation degree. For example, the second correlation degree of the candidate project matching combination {project requirement information A, current channel information B2, current expert information C2} is 1.2+1.0=2.2. Since each project matching combination includes project requirement information A, project requirement information A can be ignored when calculating the second correlation degree.

[0160] The second correlation is mainly based on whether fuzzy matching is performed when matching the keyword with the current channel information and the current expert information, which is determined by whether fuzzy matching is performed on each keyword contained therein.

[0161] Figure 6 FIG. 1 is a flow chart of determining a target project matching combination in one embodiment of the project information matching method disclosed in the present invention. Figure 6 The method shown includes steps: S601-S602. Each step is described below.

[0162] S601: taking the sum of the first degree of association and the second degree of association or the product of the first degree of association and the second degree of association as the third degree of association.

[0163] S601: Determine the candidate item matching combination corresponding to the largest third correlation degree as the target item matching combination.

[0164] For example, a weighted sum of the first degree of association and the second degree of association is calculated, or a weighted product of the first degree of association and the second degree of association is calculated, and the weighted sum or weighted product is used as the third degree of association. The third degree of association can be selected for sorting, and the i candidate project matching combinations with the highest scores are selected as the target project matching combinations, and the target project matching combinations are used as the recommendation results to recommend the current channel information (technology broker information) and current expert information in the target project matching combinations to the user, where i can be 1, 2, 3, etc.

[0165] Exemplary Devices

[0166] In one embodiment, Figure 7 As shown, the present disclosure provides a project information matching device, including: a first keyword acquisition module 701, a second keyword acquisition module 702, a first relevance calculation module 703, a second relevance calculation module 704 and a target project matching determination module 705.

[0167] The first keyword acquisition module 701 receives project demand information and acquires demand keywords from the demand corpus based on the project demand information. The second keyword acquisition module 702 acquires corresponding channel keywords in the channel corpus and corresponding expert keywords in the expert corpus according to the demand keywords. The first correlation calculation module 703 calculates the first correlation between the demand keywords, the channel keywords and the expert keywords.

[0168] The second correlation calculation module 704 selects candidate project matching combinations according to demand keywords, channel keywords and expert keywords and calculates the second correlation of the candidate project matching combinations; wherein the candidate project matching combinations include: project demand information, current channel information corresponding to the project demand information and current expert information.

[0169] The target item matching determination module 705 calculates a third degree of association corresponding to the candidate item matching combination according to the first degree of association and the second degree of association, and determines the target item matching combination from the candidate item matching combination based on the third degree of association. For example, the target item matching determination module 705 uses the sum of the first degree of association and the second degree of association, or the product of the first degree of association and the second degree of association, as the third degree of association; and determines the candidate item matching combination corresponding to the largest third degree of association as the target item matching combination.

[0170] In one embodiment, Figure 8 As shown, the corpus construction module 706 obtains the historical project information that has been successfully transferred or converted, extracts the demand keywords, channel keywords and expert keywords from the historical project information and calculates the corresponding word vectors. The corpus construction module 706 stores the demand keywords, channel keywords and expert keywords and the corresponding word vectors in the demand corpus, channel corpus and expert corpus respectively.

[0171] like Fig. 9 As shown, the corpus construction module 706 includes: a keyword processing unit 7061 and a keyword selection unit 7062. The keyword processing unit 7061 performs word segmentation and preprocessing on the historical project information, obtains demand keywords, channel keywords and expert keywords, and vectorizes the demand keywords, channel keywords and expert keywords respectively to obtain word vectors. The keyword selection unit 7062 calculates the word frequency TF and inverse document frequency IDF of the demand keywords, channel keywords and expert keywords respectively, obtains the TF-IDF values ​​corresponding to the demand keywords, channel keywords and expert keywords, and selects a preset number of demand keywords, channel keywords and expert keywords based on the preset sorting rules and according to the TF-IDF values.

[0172] In one embodiment, the demand corpus includes: a demand attribute word library and a demand target word library. The first keyword acquisition module 701 acquires demand project attribute keywords and demand project target keywords from the demand attribute word library and the demand target word library respectively based on the project demand information. The channel corpus includes: a channel attribute word library, a channel target word library and a channel success word library; the expert corpus includes: an expert attribute word library and an expert success word library.

[0173] like Fig.10, the second keyword acquisition module 702 includes: a library group division unit 7021, a keyword matching unit 7022, and a phrase establishment unit 7023. The library group division unit 7021 sets a first library group and a second library group; wherein the first library group includes: a demand attribute word library, a channel attribute word library, and an expert attribute word library; the second library group includes: a demand target word library, a channel target word library, a channel success word library, and an expert success word library. The first library group and the second library group can also be set by the keyword matching unit 7022. The keyword matching unit 7022 obtains the corresponding channel attribute keywords and expert attribute keywords from the channel attribute word library and the expert attribute word library in the first library group according to the demand project attribute keywords. The phrase establishment unit 7023 establishes at least one first keyword group based on the correspondence between the demand project attribute keywords, the channel attribute keywords, and the expert attribute keywords.

[0174] The keyword matching unit 7022 obtains the corresponding channel target project keywords, channel success project keywords and expert success project keywords from the channel target word library, channel success word library and expert success word library in the second library group according to the demand project target keywords. The phrase building unit 7023 builds at least one second keyword group based on the correspondence between the demand project target keywords, channel target project keywords, channel success project keywords and expert success project keywords.

[0175] like Fig.11 As shown, the first relevance calculation module 703 includes: an attribute relevance calculation unit 7031, a target relevance calculation unit 7032 and a comprehensive relevance calculation unit 7033. The attribute relevance calculation unit 7031 obtains the word vector of each keyword in the first keyword group, and calculates the attribute relevance corresponding to the first keyword group based on the word vector.

[0176] The target relevance calculation unit 7032 obtains the word vector of each keyword in the second keyword group, and calculates the target relevance corresponding to the second keyword group based on the word vector. The comprehensive relevance calculation unit 7033 calculates the first relevance based on the attribute relevance and the target relevance.

[0177] The attribute association calculation unit 7031 calculates the first distance between the word vectors of every two keywords among the demand item attribute keywords, channel attribute keywords and expert attribute keywords, and takes the product of the three first distances as the attribute association degree.

[0178] In one embodiment, the target association calculation unit 7032 calculates the second distance between the word vectors of every two keywords in the demand project target keywords, channel target project keywords, and expert success project keywords, and takes the product of the three second distances as the first sub-target association degree.

[0179] The target association calculation unit 7032 calculates the third distance between the word vectors of each two keywords in the target keywords of the demand project, the channel success project keywords, and the expert success project keywords, takes the product of the three third distances as the second sub-target association, and calculates the average sum of the first sub-target association and the second sub-target association as the target association. The comprehensive association calculation unit 7033 calculates the first association based on all attribute associations and all target associations.

[0180] like Fig.12 As shown, the second correlation calculation module 704 includes: a project matching combination acquisition unit 7041 and a matching correlation calculation unit 7042. The project matching combination acquisition unit 7041 generates a demand keyword set according to the demand project attribute keywords and the demand project target keywords; the project matching combination acquisition unit 7041 generates a channel keyword set according to the channel attribute keywords, the channel target project keywords and the channel success project keywords.

[0181] The project matching combination acquisition unit 7041 generates an expert keyword set based on the expert attribute keywords and the expert successful project keywords, and selects the current channel information based on the channel keyword set; wherein the current channel information includes at least one first keyword in the channel keyword set. The project matching combination acquisition unit 7041 selects the current expert information based on the expert keyword set; wherein the current expert information includes at least one second keyword in the expert keyword set. The project matching combination acquisition unit 7041 generates a candidate project matching combination based on the project demand information, the current channel information and the current expert information.

[0182] The matching association calculation unit 7042 obtains all first keywords in the channel keyword set included in the current channel information, determines the corresponding first association coefficient based on the vocabulary to which the first keyword belongs, and calculates the first matching association degree of the current channel information based on all the first association coefficients. The matching association calculation unit 7042 obtains all second keywords in the expert keyword set included in the current expert information, determines the corresponding second association coefficient based on the vocabulary to which the second keyword belongs, and calculates the second matching association degree of the current expert information based on all the second association coefficients. The matching association calculation unit 7042 calculates the second association degree based on the first matching association degree and the second matching association degree.

[0183] Fig.13 FIG. 1 is a schematic diagram of a structure of another embodiment of the project information matching device disclosed in the present invention. Fig.13 As shown, the project information matching device 131 includes one or more processors 1311 and a memory 1312 .

[0184] The processor 1311 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the item information matching device 131 to perform desired functions.

[0185] The memory 1312 may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory, for example, may include: random access memory (RAM) and / or cache memory (cache), etc. Non-volatile memory, for example, may include: read-only memory (ROM), hard disk, and flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 1311 may run the program instructions to implement the project information matching method of each embodiment of the present disclosure above and / or other desired functions. Various contents such as input signals, signal components, noise components, etc. may also be stored in the computer-readable storage medium.

[0186] In one example, the project information matching device 131 may also include: an input device 1313 and an output device 1314, etc., which are interconnected through a bus system and / or other forms of connection mechanisms (not shown). In addition, the input device 1313 may also include, for example, a keyboard, a mouse, etc. The output device 1314 may output various information to the outside. The output device 1314 may include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, etc.

[0187] Of course, to simplify, Fig.13 Only some of the components related to the present disclosure in the project information matching device 131 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the project information matching device 131 may also include any other appropriate components.

[0188] In addition to the above-mentioned methods and devices, an embodiment of the present disclosure may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the project information matching method according to various embodiments of the present disclosure described in the above-mentioned "Exemplary Method" section of this specification.

[0189] The computer program product may be written in any combination of one or more programming languages ​​to write program code for performing the operations of the disclosed embodiments, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0190] In addition, an embodiment of the present disclosure may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the project information matching method according to various embodiments of the present disclosure described in the above "Exemplary Method" section of this specification.

[0191] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive enumeration) of readable storage media can include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0192] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, rather than limitation, and the above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0193] The project information matching method, device and storage medium in the above-mentioned embodiments obtain the corresponding channel keywords in the channel corpus and the corresponding expert keywords in the expert corpus according to the demand keywords, and calculate the first correlation; select the candidate project matching combination according to the demand keywords, channel keywords and expert keywords, and calculate the second correlation of the candidate project matching combination; calculate the third correlation corresponding to the candidate project matching combination according to the first correlation and the second correlation, so as to determine the target project matching combination; realize automatic matching between the supply and demand parties and the third-party transfer channels, reduce the manual search and matching process, improve the matching efficiency and accuracy, improve the technology transfer conversion efficiency, and improve the user experience.

[0194] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0195] The block diagrams of the devices, apparatuses, equipment, and systems involved in the present disclosure are intended only as illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems may be connected, arranged, and configured in any manner. Words such as "include," "comprise," "have," and the like are open words, meaning "including but not limited to," and may be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or," and may be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and may be used interchangeably therewith.

[0196] The method and apparatus of the present disclosure may be implemented in many ways. For example, the method and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

[0197] It should also be noted that in the apparatus, device and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0198] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects, etc., will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0199] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A project information matching method, comprising: Receiving project requirement information, and acquiring requirement keywords from a requirement corpus based on the project requirement information; Acquire corresponding channel keywords in the channel corpus and corresponding expert keywords in the expert corpus according to the demand keywords; Calculating a first correlation between the demand keyword, the channel keyword, and the expert keyword; According to the demand keyword, the channel keyword and the expert keyword, a candidate project matching combination is selected and a second correlation degree of the candidate project matching combination is calculated; wherein the candidate project matching combination includes: the project demand information, current channel information corresponding to the project demand information and current expert information; Calculating a third degree of association corresponding to the candidate item matching combination according to the first degree of association and the second degree of association, and determining a target item matching combination from the candidate item matching combination based on the third degree of association; wherein, historical project information of successful transfer or conversion is obtained, the demand keywords, the channel keywords and the expert keywords are extracted from the historical project information and corresponding word vectors are calculated; the demand keywords, the channel keywords and the expert keywords and the corresponding word vectors are stored in the demand corpus, the channel corpus and the expert corpus respectively; The demand corpus includes: a demand attribute word library and a demand target word library; the obtaining of demand keywords from the demand corpus based on the project demand information includes: obtaining demand project attribute keywords and demand project target keywords from the demand attribute word library and the demand target word library respectively based on the project demand information.

2. The method according to claim 1, wherein extracting the demand keywords, the channel keywords, the expert keywords and the corresponding word vectors from the historical project information comprises: Performing word segmentation and preprocessing on the historical project information to obtain the demand keywords, the channel keywords, and the expert keywords; Performing vectorization processing on the demand keywords, the channel keywords, and the expert keywords respectively to obtain word vectors; Calculate the word frequency TF and the inverse document frequency IDF of the demand keyword, the channel keyword and the expert keyword respectively, and obtain the TF-IDF values ​​corresponding to the demand keyword, the channel keyword and the expert keyword; Based on the preset sorting rules and according to the TF-IDF value, a preset number of demand keywords, channel keywords and expert keywords are selected respectively.

3. The method according to claim 1 or 2, wherein: The channel corpus includes: a channel attribute word lexicon, a channel target word lexicon and a channel success word lexicon; the expert corpus includes: an expert attribute word lexicon and an expert success word lexicon; the method of obtaining corresponding channel keywords in the channel corpus and corresponding expert keywords in the expert corpus according to the demand keywords includes: A first library group is set, and corresponding channel attribute keywords and expert attribute keywords are respectively obtained from the channel attribute word library and the expert attribute word library in the first library group according to the attribute keywords of the demand items; the first library group includes: the demand attribute word library, the channel attribute word library and the expert attribute word library; Establishing at least one first keyword group based on the correspondence between the demand item attribute keywords, the channel attribute keywords and the expert attribute keywords; Set a second library group, and obtain corresponding channel target project keywords, channel success project keywords and expert success project keywords from the channel target word library, the channel success word library and the expert success word library in the second library group according to the demand project target keywords; the second library group includes: demand target word library, channel target word library, channel success word library and expert success word library; At least one second keyword group is established based on the correspondence between the demand project target keywords, the channel target project keywords, the channel success project keywords and the expert success project keywords.

4. The method according to claim 3, wherein the calculating the first degree of association between the demand keyword, the channel keyword and the expert keyword comprises: Obtaining a word vector of each keyword in the first keyword group, and calculating an attribute association degree corresponding to the first keyword group based on the word vector; Obtaining a word vector of each keyword in the second keyword group, and calculating a target relevance corresponding to the second keyword group based on the word vector; The first degree of association is calculated based on the attribute degree of association and the target degree of association.

5. The method according to claim 4, wherein the calculating the attribute association degree corresponding to the first keyword group comprises: Calculating the first distance between the word vectors of every two keywords among the demand item attribute keyword, the channel attribute keyword and the expert attribute keyword, and taking the product of three first distances as the attribute association degree; The calculating of the target relevance corresponding to the second keyword group comprises: Calculate the second distance between the word vectors of every two keywords among the demand project target keyword, the channel target project keyword, and the expert success project keyword, and use the product of the three second distances as the first sub-goal relevance; Calculate the third distance between the word vectors of every two keywords among the target keyword of the demand project, the keyword of the channel success project, and the keyword of the expert success project, and use the product of the three third distances as the second sub-goal relevance; Calculating an average sum of the first sub-goal relevance and the second sub-goal relevance as the goal relevance; The calculating the first degree of association based on the attribute degree of association and the target degree of association comprises: The first degree of association is calculated based on all attribute degrees of association and all target degrees of association.

6. The method according to claim 5, wherein the selecting a candidate item matching combination according to the demand keyword, the channel keyword and the expert keyword comprises: Generate a demand keyword set according to the demand item attribute keywords and the demand item target keywords; Generate a channel keyword set according to the channel attribute keywords, the channel target project keywords and the channel successful project keywords; Generate an expert keyword set according to the expert attribute keywords and the expert successful project keywords; Selecting the current channel information based on the channel keyword set; wherein the current channel information includes at least one first keyword in the channel keyword set; Selecting the current expert information based on the expert keyword set; wherein the current expert information includes at least one second keyword in the expert keyword set; The candidate project matching combination is generated according to the project demand information, the current channel information and the current expert information.

7. The method according to claim 6, wherein the calculating the second degree of association of the candidate item matching combination comprises: Acquire all first keywords in the channel keyword set contained in the current channel information; Determine a corresponding first association coefficient based on the vocabulary to which the first keyword belongs; Calculating a first matching correlation degree of the current channel information based on all first correlation coefficients; Acquire all second keywords in the expert keyword set included in the current expert information, and determine corresponding second association coefficients based on the vocabulary to which the second keywords belong; Calculate the second matching correlation degree of the current expert information based on all second correlation coefficients; The second degree of association is calculated according to the first matching degree of association and the second matching degree of association.

8. The method according to any one of claims 1 to 7, wherein the calculating a third degree of association corresponding to the candidate item matching combination according to the first degree of association and the second degree of association, and determining a target item matching combination from the candidate item matching combination based on the third degree of association comprises: taking the sum of the first degree of association and the second degree of association or the product of the first degree of association and the second degree of association as the third degree of association; The candidate item matching combination corresponding to the largest third correlation degree is determined as the target item matching combination.

9. A project information matching device, comprising: A first keyword acquisition module, configured to receive project requirement information, and acquire requirement keywords from a requirement corpus based on the project requirement information; A second keyword acquisition module is used to acquire corresponding channel keywords in the channel corpus and corresponding expert keywords in the expert corpus according to the demand keywords; A first correlation calculation module, used to calculate a first correlation between the demand keyword, the channel keyword and the expert keyword; A second correlation calculation module is used to select a candidate project matching combination and calculate a second correlation of the candidate project matching combination according to the demand keyword, the channel keyword and the expert keyword; wherein the candidate project matching combination includes: the project demand information, the current channel information and the current expert information corresponding to the project demand information; a target item matching determination module, configured to calculate a third degree of association corresponding to the candidate item matching combination according to the first degree of association and the second degree of association, and determine a target item matching combination from the candidate item matching combination based on the third degree of association; A corpus construction module is used to obtain historical project information that has been successfully transferred or converted, extract the demand keywords, the channel keywords, and the expert keywords from the historical project information, and calculate corresponding word vectors; store the demand keywords, the channel keywords, the expert keywords, and the corresponding word vectors in the demand corpus, the channel corpus, and the expert corpus, respectively; Among them, the demand corpus includes: a demand attribute word library and a demand target word library; the first keyword acquisition module is used to obtain demand project attribute keywords and demand project target keywords from the demand attribute word library and the demand target word library respectively based on the project demand information.

10. The device of claim 9, wherein: The corpus construction module includes: A keyword processing unit, used to perform word segmentation and preprocessing on the historical project information to obtain the demand keywords, the channel keywords and the expert keywords; respectively perform vectorization processing on the demand keywords, the channel keywords and the expert keywords to obtain word vectors; The keyword selection unit is used to respectively calculate the term frequency TF and the inverse document frequency IDF of the demand keywords, the channel keywords and the expert keywords, and obtain the TF-IDF values ​​corresponding to the demand keywords, the channel keywords and the expert keywords; based on the preset sorting rules and according to the TF-IDF values, a preset number of demand keywords, channel keywords and expert keywords are selected respectively.

11. The device according to claim 9 or 10, wherein: The channel corpus includes: a channel attribute word library, a channel target word library and a channel success word library; the expert corpus includes: an expert attribute word library and an expert success word library; The second keyword acquisition module includes: The keyword matching unit is used to set a first library group, and obtain corresponding channel attribute keywords and expert attribute keywords from the channel attribute word library and the expert attribute word library in the first library group according to the demand item attribute keyword; the first library group includes: the demand attribute word library, the channel attribute word library and the expert attribute word library; A phrase building unit, configured to build at least one first keyword group based on the correspondence between the demand item attribute keywords, the channel attribute keywords and the expert attribute keywords; The keyword matching unit is used to set a second library group, and obtain the corresponding channel target project keywords, channel successful project keywords and expert successful project keywords from the channel target word library, the channel successful word library and the expert successful word library in the second library group according to the demand project target keyword; the second library group includes: the demand target word library, the channel target word library, the channel successful word library and the expert successful word library; The phrase building unit is used to build at least one second keyword group based on the correspondence between the demand project target keywords, the channel target project keywords, the channel success project keywords and the expert success project keywords.

12. The device of claim 11, wherein: The first association degree calculation module includes: an attribute association calculation unit, configured to obtain a word vector of each keyword in the first keyword group, and calculate an attribute association degree corresponding to the first keyword group based on the word vector; a target association calculation unit, configured to obtain a word vector of each keyword in the second keyword group, and calculate a target association degree corresponding to the second keyword group based on the word vector; A comprehensive association calculation unit is used to calculate the first association degree based on the attribute association degree and the target association degree.

13. The device of claim 12, wherein: The attribute association calculation unit is used to calculate the first distance between the word vectors of each two keywords among the demand item attribute keyword, the channel attribute keyword and the expert attribute keyword, and take the product of three first distances as the attribute association degree; The target association calculation unit is used to calculate the second distance between the word vectors of each two keywords among the demand project target keyword, the channel target project keyword, and the expert success project keyword, and use the product of the three second distances as the first sub-target association degree; Calculate the third distance between the word vectors of every two keywords among the target keyword of the demand project, the keyword of the channel success project, and the keyword of the expert success project, and use the product of the three third distances as the second sub-goal relevance; Calculating an average sum of the first sub-goal relevance and the second sub-goal relevance as the goal relevance; The comprehensive association calculation unit is used to calculate the first association degree based on all attribute association degrees and all target association degrees.

14. The device of claim 13, wherein: The second association degree calculation module includes: The project matching combination acquisition unit is used to generate a demand keyword set according to the demand project attribute keywords and the demand project target keywords; generate a channel keyword set according to the channel attribute keywords, the channel target project keywords and the channel successful project keywords; generate an expert keyword set according to the expert attribute keywords and the expert successful project keywords; select the current channel information based on the channel keyword set; wherein the current channel information contains at least one first keyword in the channel keyword set; select the current expert information based on the expert keyword set; wherein the current expert information contains at least one second keyword in the expert keyword set; generate the candidate project matching combination according to the project demand information, the current channel information and the current expert information.

15. The device of claim 14, wherein: The second association degree calculation module includes: A matching association calculation unit is used to obtain all first keywords in the channel keyword set contained in the current channel information; determine the corresponding first association coefficient based on the vocabulary to which the first keyword belongs; calculate the first matching association degree of the current channel information based on all the first association coefficients; obtain all second keywords in the expert keyword set contained in the current expert information, and determine the corresponding second association coefficient based on the vocabulary to which the second keyword belongs; calculate the second matching association degree of the current expert information based on all the second association coefficients; calculate the second association degree based on the first matching association degree and the second matching association degree.

16. The device according to any one of claims 9 to 15, wherein: The target item matching determination module is configured to use the sum of the first degree of association and the second degree of association or the product of the first degree of association and the second degree of association as the third degree of association; The candidate item matching combination corresponding to the largest third correlation degree is determined as the target item matching combination.

17. A project information matching device, comprising: processor; a memory for storing instructions executable by the processor; The processor is used to read the executable instructions from the memory and execute the instructions to implement the method described in any one of claims 1 to 8.

18. A computer-readable storage medium storing a computer program, wherein the computer program is used to execute the method according to any one of claims 1 to 8.

Images