Precise matching algorithm based on big data

Abstract

The invention relates to the technical field of big data analysis, in particular to an accurate matching algorithm based on big data, which comprises the following steps: acquiring basic attributes of an entrepreneurship project, creating an I-type database, and storing the basic attributes of the entrepreneurship project by using the I-type database; uploading the intent investment target attribute of the investment institution, creating a II-type database, and storing the intent investment target attribute of the investment institution by using the II-type database; configuring weights for basic attributes of each entrepreneurship project in the I-type database and attributes of an intentional investment target of each investment institution in the II-type database; selecting an entrepreneurship project from the I-type database as a to-be-processed target, traversing the II-type database based on the to-be-processed target, and screening to obtain a basically matched investment institution group; in the investment institution group, based on attribute similarity analysis, capturing an optimal adaptive investment institution for the to-be-processed target; the invention provides an intelligent and high-precision matching method for both the startup project and the investment institution in combination with a semantic analysis technology.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to a precise matching algorithm based on big data. Background Art

[0002] Investment institutions are organizations that provide financial support for various projects. They look for entrepreneurial projects with potential, just like digging for treasures. Entrepreneurial projects are carriers of creativity and dreams. The two are closely linked. The financial support of investment institutions can enable entrepreneurial projects to thrive, and high-quality projects can also bring rich returns to investment institutions.

[0003] The invention patent with application number 202311781701.9 discloses a precise matching system based on policy and enterprise big data, which includes the following steps: S1: Establish a network platform based on blockchain: Establish a platform that provides policy data to enterprises through blockchain technology; S2: Policy data integration: Classify policies according to policy requirements, and enter the classified policy information and enterprise demand information into the blockchain-based network platform; S3: Establish an information tree: Encode enterprise data according to policy type and policy, and generate a QR code or code for policy information, and then establish an information tree according to the policy information QR code or code when the policy information is entered; S4: Map management Management: When establishing an information tree, the information tree is established in a one-to-many manner, that is, the policy information QR code or code is used as the trunk of the information tree, and information trees are established corresponding to other information branches of the policy information; S5: Information query integration: When querying information, information is obtained by querying information on different blockchain points. When querying information, the information is queried by entering the QR code or code of the product, and then the policy information is retrieved from the database based on each blockchain point for display; S6: Data management: When modifying data, the information is modified on the blockchain point of the policy information entry department, and other blockchain points are restricted to only have viewing rights, and have no right to modify the data.

[0004] This application aims to solve the problem that "existing technology is unable to integrate policy information with corporate needs."

[0005] However, in the matching scenario between entrepreneurial projects and investment institutions, the matching between entrepreneurial projects and investment institutions mostly relies on evaluation by full-time personnel. The accuracy of the evaluation matching is affected by the cognitive experience of the evaluators, and the evaluation efficiency is relatively low and the evaluation accuracy is relatively poor. Summary of the invention

[0006] In view of the above-mentioned shortcomings of the prior art, the present invention provides a precise matching algorithm based on big data, which solves the technical problems raised in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] A precise matching algorithm based on big data, including:

[0009] Obtain basic attributes of entrepreneurial projects, create a Type I database, and use the Type I database to store basic attributes of entrepreneurial projects; upload attributes of investment institutions' intended investment targets, create a Type II database, and use the Type II database to store attributes of investment institutions' intended investment targets; configure weights for each basic attribute of entrepreneurial projects in the Type I database and each attribute of investment institutions' intended investment targets in the Type II database; select entrepreneurial projects as targets to be processed in the Type I database, traverse the Type II database based on the targets to be processed, and screen and obtain a group of investment institutions that basically match; based on attribute similarity analysis in the group of investment institutions, capture the best-fit investment institution for the target to be processed; output stage of the best-fit investment institution.

[0010] Furthermore, the basic attributes of the entrepreneurial project are composed of a number of Chinese words, and the Chinese words are no less than three, and the Chinese words are related to the content of the entrepreneurial project. The basic attributes of the entrepreneurial project are manually edited by the user. When storing the basic attributes of the entrepreneurial project, the Type I database simultaneously creates a data set based on the Chinese words contained in the basic attributes of the entrepreneurial project. The data set is simultaneously marked based on the name of the user who originated the entrepreneurial project and the amount of funds required to start the entrepreneurial project. The data sets corresponding to the basic attributes of each entrepreneurial project perform differentiated storage operations in the Type I database based on their marking information.

[0011] Furthermore, the investment institution's intended investment target attributes are composed of several Chinese words, and the Chinese words are no less than three, and the Chinese words are related to the content of the entrepreneurial project. The investment institution's intended investment target attributes are manually edited by the user end. When the Type II database stores the investment institution's intended investment target attributes, it simultaneously creates a data set based on the Chinese words contained in the intended investment target attributes. The data set is simultaneously marked based on the source investment institution name of its intended investment target attributes and the acceptable entrepreneurial project start-up funding threshold. The corresponding data sets of each investment institution's intended investment target attributes perform differentiated storage operations in the Type II database based on its marking information.

[0012] Furthermore, weights are assigned to the basic attributes of each entrepreneurial project and the attributes of the intended investment target of each investment institution, and weights are assigned to each Chinese word;

[0013] The weights include: 0.1, 0.2, and 0.3. The weight of each Chinese word is manually configured by the Chinese word source user during the Chinese word editing stage.

[0014] Furthermore, weights are assigned to the basic attributes of each entrepreneurial project and the attributes of the intended investment target of each investment institution, and weights are assigned to each Chinese word;

[0015] The weights include: 0.1, 0.2, and 0.3. The weight of each Chinese word is manually configured by the Chinese word source user during the Chinese word editing stage.

[0016] Furthermore, the attribute similarity analysis logic between each investment institution in the investment institution group and the target to be processed is expressed as:

[0017]

[0018] Where: SIMM (A, B) is the attribute similarity between the target A to be processed and the investment institution B; is the average vector of the Chinese word set corresponding to the target A to be processed, and the average vector of the Chinese word set corresponding to the investment institution B; For vector Model;

[0019] in, d is the dimension of the word vector; μ A,k , μ B,k For vector The value of the kth dimension, vector The value of the kth dimension, and is calculated in the same way.

[0020] Furthermore, based on the logical calculation formula of SIMM (A, B), the attribute similarity between each investment institution and the target to be processed is requested, and the calculation results of each attribute similarity are corrected. The correction logic is expressed as:

[0021] Get the configuration weight of each Chinese word in the Chinese word set corresponding to the target to be processed, select the Chinese word with the largest configuration weight, and calculate the similarity between the selected Chinese word and each Chinese word in the Chinese word set corresponding to the investment institution. The similarity calculation logic follows the calculation logic of SIMM (A, B), and all similarity calculation results are summed up. The sum result is recorded as Q, and the correction result of the attribute similarity calculation result is:

[0022] SIMM(A,B)′=SIMM(A,B)×Q A,B ;

[0023] Where: SIMM(A,B)′ is the modified attribute similarity between the target A and the investment institution B; Q A,B Q corresponding to the target A to be processed and the investment institution B;

[0024] Among them, based on the above formula, the attribute similarity between each investment institution in the investment institution group and the target to be processed is corrected, and the investment institution corresponding to the correction result with the largest value is used as the best-fit investment institution to perform the capture operation.

[0025] Furthermore, in the output stage of the best-fit investment institutions, the user terminal customizes the output number of the best-fit investment institutions. When the number of the best-fit investment institutions set by the user terminal is greater than one, the corrected attribute similarity calculation results are arranged in descending order, and the investment institutions corresponding to the number of results corresponding to the set output number at the front of the descending order queue are selected as output targets;

[0026] Among them, the investment institutions that serve as output targets still maintain their relative positions in the descending queue.

[0027] Furthermore, when the number of the best-matched investment institutions set on the user side is less than one, the basic attributes of the target to be processed are reset and edited or deleted.

[0028] Furthermore, each time the target to be processed completes the output of the best-fit investment institution, it jumps to the selection stage of the target to be processed, and selects a new entrepreneurial project in the Type I database as the target to be processed, until all entrepreneurial projects corresponding to the basic attributes of all entrepreneurial projects in the Type I database are selected as the targets to be processed and the output of the best-fit investment institution is completed, and then the process ends.

[0029] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0030] The present invention provides a precise matching algorithm based on big data. During the execution of the algorithm, the algorithm combines semantic analysis technology to provide an intelligent and highly accurate matching method for both entrepreneurial projects and investment institutions. During the execution phase of the algorithm, the attributes of entrepreneurial projects and investment institutions are distinguished and stored by constructing two databases, and then a data set is generated for each. Through a comprehensive evaluation of the similarity of the data sets and further correction of the configuration weights, the optimal investment institution is configured for each entrepreneurial project to ensure the good "development" of the entrepreneurial project.

[0031] At the same time, by recommending customized outputs of matching investment institutions, more choices and opportunities are provided to users of entrepreneurial projects, effectively providing entrepreneurial projects with more "growth" opportunities. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 The figure is a flowchart of a precise matching algorithm based on big data. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] The present invention will be further described below in conjunction with the embodiments.

[0036] Example:

[0037] In this embodiment, a precise matching algorithm based on big data is used. Figure 1 As shown, including:

[0038] Obtain basic attributes of entrepreneurial projects, create a Type I database, and use the Type I database to store basic attributes of entrepreneurial projects;

[0039] Upload the investment target attributes of the investment institution, create a Type II database, and use the Type II database to store the investment target attributes of the investment institution;

[0040] Assign weights to the basic attributes of each entrepreneurial project in the Type I database and the attributes of each investment institution's intended investment target in the Type II database;

[0041] The operation of assigning weights to the basic attributes of each entrepreneurial project and the attributes of the intended investment target of each investment institution, and the operation of assigning weights to each Chinese word;

[0042] The weights include: 0.1, 0.2, 0.3. The weight of each Chinese word is manually configured by the Chinese word source user during the Chinese word editing stage;

[0043] Select entrepreneurial projects as the targets to be processed in the Type I database, traverse the Type II database based on the targets to be processed, and screen and obtain a group of investment institutions that basically match them;

[0044] The logic of traversing the Type II database based on the target to be processed and screening to obtain a group of investment institutions that basically match is expressed as follows:

[0045] The amount of startup funds required for the entrepreneurial project in the tag information of the target to be processed is obtained. Based on the amount of startup funds required for the entrepreneurial project as the query target, the amount of startup funds required for the entrepreneurial project is compared with the threshold of acceptable startup funds required for the entrepreneurial project marked by the investment institution's intended investment target attribute, and the investment institution that meets the amount of startup funds required for the entrepreneurial project is queried in the type II database, and all the queried investment institutions are recorded as an investment institution group;

[0046] Among them, the operation of searching the Type II database for investment institutions that meet the amount of startup funds required for entrepreneurial projects, that is, searching for investment institutions whose amount of startup funds required for entrepreneurial projects is less than or equal to the threshold of acceptable startup funds required for entrepreneurial projects;

[0047] Based on the attribute similarity analysis among the investment institutions, the best matching investment institutions are captured for the target to be processed;

[0048] The attribute similarity analysis logic between each investment institution in the investment institution group and the target to be processed is expressed as:

[0049]

[0050] Where: SIMM (A, B) is the attribute similarity between the target A to be processed and the investment institution B; is the average vector of the Chinese word set corresponding to the target A to be processed, and the average vector of the Chinese word set corresponding to the investment institution B; For vector Model;

[0051] in, d is the dimension of the word vector; μ A,k , μ B,k For vector The value of the kth dimension, vector The value of the kth dimension, and is calculated in the same way;

[0052] Through the above logical formula, the method of calculating the attribute similarity between the target to be processed and the investment institution is limited to ensure that the investment institution that best matches the entrepreneurial project is stably output in this algorithm.

[0053] Based on the logical calculation formula of SIMM (A, B), the attribute similarity between each investment institution and the target to be processed is requested, and the calculation results of each attribute similarity are corrected. The correction logic is expressed as:

[0054] Get the configuration weight of each Chinese word in the Chinese word set corresponding to the target to be processed, select the Chinese word with the largest configuration weight, and calculate the similarity between the selected Chinese word and each Chinese word in the Chinese word set corresponding to the investment institution. The similarity calculation logic follows the calculation logic of SIMM (A, B), and all similarity calculation results are summed up. The sum result is recorded as Q, and the correction result of the attribute similarity calculation result is:

[0055] SIMM(A,B)′=SIMM(A,B)×Q A,B ;

[0056] Where: SIMM(A,B)′ is the modified attribute similarity between the target A and the investment institution B; Q A,B Q corresponding to the target A to be processed and the investment institution B;

[0057] Among them, based on the above formula, the attribute similarity between each investment institution in the investment institution group and the target to be processed is corrected, and the investment institution corresponding to the correction result with the largest value is the best-fit investment institution to perform the capture operation;

[0058] Through the above logic formula, the attribute similarity calculation result is further provided with a specified correction logic.

[0059] The output stage of the best-fit investment institution;

[0060] In the output stage of the best-fit investment institution, the user terminal customizes the output number of the best-fit investment institution. When the number of the best-fit investment institution set by the user terminal is greater than one, the corrected attribute similarity calculation results are arranged in descending order, and the investment institution corresponding to the number of results corresponding to the set output number at the front of the descending order queue is selected as the output target;

[0061] Among them, the investment institutions that are output as output targets still maintain their relative positions in the descending queue;

[0062] When the number of best-matched investment institutions set on the user side is less than one, the basic attributes of the target to be processed are reset and edited or deleted;

[0063] After each target to be processed completes the output of the best-fit investment institution, it jumps to the selection stage of the target to be processed, and selects a new entrepreneurial project in the Type I database as the target to be processed, until all entrepreneurial projects corresponding to the basic attributes of all entrepreneurial projects in the Type I database are selected as the targets to be processed and the output of the best-fit investment institution is completed, then the process ends.

[0064] In this embodiment, through the execution of the algorithm in the above embodiment, a matching service with better adaptability is provided for both entrepreneurial projects and investment institutions, ensuring that each entrepreneurial project can find its own "mentor", thereby comprehensively improving industrial diversification and economic diversification, and ensuring that each high-quality emerging enterprise in the society can grow healthily.

[0065] like Figure 1 As shown, the basic attributes of the entrepreneurial project are composed of several Chinese words, and there are no less than three Chinese words, and the Chinese words are related to the content of the entrepreneurial project. The basic attributes of the entrepreneurial project are manually edited by the user. When the I-type database stores the basic attributes of the entrepreneurial project, it simultaneously creates a data set based on the Chinese words contained in the basic attributes of the entrepreneurial project. The data set is synchronously marked based on the name of the user who originated the entrepreneurial project and the amount of funds required for the startup of the entrepreneurial project. The corresponding data set of the basic attributes of each entrepreneurial project performs a distinguished storage operation in the I-type database based on its marking information;

[0066] The attributes of an investment institution's intended investment target are composed of several Chinese words, and there are no less than three Chinese words, and the Chinese words are related to the content of the entrepreneurial project. The attributes of an investment institution's intended investment target are manually edited by the user. When storing the attributes of an investment institution's intended investment target, the Type II database simultaneously creates a data set based on the Chinese words contained in the attributes of the intended investment target. The data set is synchronously marked based on the name of the source investment institution of its intended investment target attributes and the acceptable threshold of funds required to start up the entrepreneurial project. The corresponding data sets of the attributes of each investment institution's intended investment target perform differentiated storage operations in the Type II database based on its marking information.

[0067] Through the above settings, the attribute data stored in the Type I database and the Type II database are further limited, and the storage contents of the two are limited, providing necessary execution data and logical support for the execution of the steps of the algorithm in the above embodiment.

[0068] In summary, during the execution of the algorithm in the above embodiment, combined with semantic analysis technology, an intelligent and highly accurate matching method is provided for both entrepreneurial projects and investment institutions. During the execution stage of the algorithm, the attributes of entrepreneurial projects and investment institutions are differentiated and stored by constructing two databases, and then a data set is generated for each. Through the comprehensive evaluation of the similarity of the data sets and further correction of the configuration weights, the optimal investment institution is configured for each entrepreneurial project to ensure the good development of the entrepreneurial project. At the same time, by recommending customized outputs of matching investment institutions, more choices and opportunities are provided to users of the entrepreneurial project, effectively providing more growth opportunities for the entrepreneurial project.

[0069] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A precise matching algorithm based on big data, characterized in that: include: Obtain basic attributes of entrepreneurial projects, create a Type I database, and use the Type I database to store basic attributes of entrepreneurial projects; Upload the investment target attributes of the investment institution, create a Type II database, and use the Type II database to store the investment target attributes of the investment institution; Assign weights to the basic attributes of each entrepreneurial project in the Type I database and the attributes of each investment institution's intended investment target in the Type II database; Select entrepreneurial projects as the targets to be processed in the Type I database, traverse the Type II database based on the targets to be processed, and screen and obtain a group of investment institutions that basically match them; Based on the attribute similarity analysis among the investment institutions, the best matching investment institutions are captured for the target to be processed; The output stage that best suits investment institutions.

2. According to the big data based precise matching algorithm of claim 1, it is characterized in that: The basic attributes of the entrepreneurial project are composed of a number of Chinese words, and the number of the Chinese words is no less than three, and the Chinese words are related to the content of the entrepreneurial project. The basic attributes of the entrepreneurial project are manually edited by the user. When storing the basic attributes of the entrepreneurial project, the Type I database simultaneously creates a data set based on the Chinese words contained in the basic attributes of the entrepreneurial project. The data set is simultaneously marked based on the name of the user who originated the entrepreneurial project and the amount of funds required to start the entrepreneurial project. The data sets corresponding to the basic attributes of each entrepreneurial project perform differentiated storage operations in the Type I database based on their marking information.

3. The accurate matching algorithm based on big data according to claim 1, characterized in that: The investment institution's intended investment target attributes are composed of a number of Chinese words, and the Chinese words are no less than three, and the Chinese words are related to the content of the entrepreneurial project. The investment institution's intended investment target attributes are manually edited by the user end. When storing the investment institution's intended investment target attributes, the Type II database simultaneously creates a data set based on the Chinese words contained in the intended investment target attributes. The data set is synchronously marked based on the source investment institution name of its intended investment target attributes and the acceptable entrepreneurial project start-up funding threshold. The corresponding data sets of each investment institution's intended investment target attributes perform differentiated storage operations in the Type II database based on its marking information.

4. The accurate matching algorithm based on big data according to claim 1, characterized in that: The operation of assigning weights to the basic attributes of each entrepreneurial project and the attributes of the intended investment target of each investment institution, and the operation of assigning weights to each Chinese word; The weights include: 0.1, 0.2, and 0.

3. The weight of each Chinese word is manually configured by the Chinese word source user during the Chinese word editing stage.

5. The accurate matching algorithm based on big data according to claim 1, characterized in that: The logic of traversing the Type II database based on the target to be processed and screening to obtain a group of investment institutions that basically match is expressed as follows: The amount of startup funds required for the entrepreneurial project in the tag information of the target to be processed is obtained. Based on the amount of startup funds required for the entrepreneurial project as the query target, the amount of startup funds required for the entrepreneurial project is compared with the threshold of acceptable startup funds required for the entrepreneurial project marked by the investment institution's intended investment target attribute, and the investment institution that meets the amount of startup funds required for the entrepreneurial project is queried in the type II database, and all the queried investment institutions are recorded as an investment institution group; Among them, the operation of querying the type II database for investment institutions that meet the amount of startup funds required for entrepreneurial projects is to query the investment institutions whose amount of startup funds required for entrepreneurial projects is less than or equal to the acceptable startup funds required threshold for entrepreneurial projects.

6. The accurate matching algorithm based on big data according to claim 1, characterized in that: The attribute similarity analysis logic of each investment institution in the investment institution group and the target to be processed is expressed as follows: Where: SIMM (A, B) is the attribute similarity between the target A to be processed and the investment institution B; is the average vector of the Chinese word set corresponding to the target A to be processed, and the average vector of the Chinese word set corresponding to the investment institution B; For vector Model; in, d is the dimension of the word vector; μ A,k , μ B,k For vector The value of the kth dimension, vector The value of the kth dimension, and is calculated in the same way.

7. The accurate matching algorithm based on big data according to claim 6 is characterized in that: Based on the logical calculation formula of SIMM (A, B), the attribute similarity between each investment institution and the target to be processed is requested, and the calculation results of each attribute similarity are corrected. The correction logic is expressed as: Get the configuration weight of each Chinese word in the Chinese word set corresponding to the target to be processed, select the Chinese word with the largest configuration weight, and calculate the similarity between the selected Chinese word and each Chinese word in the Chinese word set corresponding to the investment institution. The similarity calculation logic follows the calculation logic of SIMM (A, B), and all similarity calculation results are summed. The sum result is recorded as Q, and the correction result of the attribute similarity calculation result is: SIMM(A,B)′=SIMM(A,B)×Q A,B ; Where: SIMM(A,B)′ is the modified attribute similarity between the target A and the investment institution B; Q A,B Q corresponding to the target A to be processed and the investment institution B; Among them, based on the above formula, the attribute similarity between each investment institution in the investment institution group and the target to be processed is corrected, and the investment institution corresponding to the correction result with the largest value is used as the best-fit investment institution to perform the capture operation.

8. The accurate matching algorithm based on big data according to claim 1, characterized in that: In the output stage of the best-fit investment institution, the user terminal customizes the output number of the best-fit investment institution. When the number of the best-fit investment institution set by the user terminal is greater than one, the corrected attribute similarity calculation results are arranged in descending order, and the investment institutions corresponding to the number of results corresponding to the set output number at the front of the descending order queue are selected as output targets; Among them, the investment institutions that serve as output targets still maintain their relative positions in the descending queue.

9. The accurate matching algorithm based on big data according to claim 8, characterized in that: When the number of the best-matched investment institutions set on the user side is less than one, the basic attributes of the target to be processed are reset and then edited or deleted.

10. The accurate matching algorithm based on big data according to claim 1, characterized in that: Each time the target to be processed completes the output of the best-fit investment institution, it jumps to the selection stage of the target to be processed, and selects a new entrepreneurial project in the Type I database as the target to be processed, until all entrepreneurial projects corresponding to the basic attributes of all entrepreneurial projects in the Type I database are selected as the targets to be processed and the output of the best-fit investment institution is completed, and then the process ends.

Images