Method and device for selecting target enterprise based on similarity

Abstract

The invention relates to the technical field of data processing, in particular to a method and device for selecting a target enterprise based on similarity, and the method comprises the steps: obtaining credit information of multiple benchmarking enterprises meeting a credit obtaining condition from a target data platform in advance; screening out candidate enterprises of which the credit investigation information does not meet a preset credit investigation condition from a plurality of first enterprises which are about to enter the credit audit node; for each candidate enterprise, obtaining second government affair data of the candidate enterprise, respectively calculating the similarity with the first government affair data of each benchmarking enterprise in the plurality of benchmarking enterprises, and determining a corresponding maximum similarity value as an evaluation value of the candidate enterprise for selecting a target enterprise based on the similarity; and according to the evaluation value of each candidate enterprise and the change condition of the government affair data of each candidate enterprise, selecting a second enterprise to enter a credit auditing node from the candidate enterprises. Therefore, the problem of low admission rate of small and micro enterprises under common rules is effectively solved, and the admission efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for selecting a target enterprise based on similarity. Background Art

[0002] Small and micro enterprises are the main force in solving the employment rate problem, but they have difficulty in financing and high financing costs. There is a set of entry rules for small and micro enterprises. Only when their corresponding financial indicators, tax payment indicators and other indicators reach a certain threshold can they enter the pre-loan review stage, or directly lend to enterprises with good credit in the past.

[0003] The small and micro enterprises that are allowed to enter the list are usually those with very good credit or have already obtained loans. For those with no credit record, it is difficult for them to enter the list due to insufficient credit-related data, and thus they cannot enjoy the support brought by inclusive finance. In addition, the current evaluation of whether small and micro enterprises can obtain loans usually adopts manual comparison, which is highly subjective and inefficient. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the present application provides a method and device for selecting target enterprises based on similarity. In the absence of financial data, based on the similarity calculation of government data, small and micro enterprises to be included in the credit review node can be automatically selected to improve access efficiency.

[0005] In a first aspect, the present application provides a method for selecting a target enterprise based on similarity, which is applied to a server or a terminal device, and the method comprises the following steps:

[0006] Accessing a target data platform in advance and obtaining credit information of a plurality of benchmarking enterprises that meet credit acquisition conditions from the target data platform, wherein the credit information includes first government affairs data;

[0007] Screening out candidate enterprises whose credit information does not meet preset credit conditions from multiple first enterprises to be entered into the credit review node;

[0008] For each candidate enterprise, obtain the second government affairs data of the candidate enterprise, and respectively calculate the similarity between the second government affairs data of the candidate enterprise and the first government affairs data of each of the multiple benchmarking enterprises, and determine the corresponding maximum similarity value as the evaluation value of the candidate enterprise for selecting the target enterprise based on the similarity;

[0009] According to the evaluation value of each candidate enterprise and the change of the government affairs data of each candidate enterprise, a second enterprise to enter the credit review node is selected from the candidate enterprises.

[0010] In a possible implementation, satisfying the credit acquisition conditions includes:

[0011] Having loan information and having designated loan information within the most recent preset time period;

[0012] No unusual events occurred within a specified time period after the loan period with specified loan information.

[0013] In a possible implementation, respectively calculating the similarity between the second government affairs data of the candidate enterprise and the first government affairs data of each of the multiple benchmarking enterprises, and determining the corresponding maximum similarity value as the evaluation value of the candidate enterprise for selecting the target enterprise based on the similarity, comprises the following steps:

[0014] By formula Calculate the similarity between the second government data of the candidate enterprise and the first government data of each benchmark enterprise in multiple benchmark enterprises respectively; where N represents the number of candidate enterprises, P represents the number of benchmark enterprises, and x i represents a benchmarking enterprise data set consisting of multiple benchmarking enterprises, where i ranges from 1 to P, representing the id of each benchmarking enterprise; x j represents a candidate enterprise data set consisting of multiple candidate enterprises, where j ranges from 1 to N, representing the id of each candidate enterprise;

[0015] The calculated similarity between the second government data of each candidate enterprise and the first government data of each benchmark enterprise in multiple benchmark enterprises is combined into a similarity set {Simi n,1 ,Simi n,2 …Simi n,P}, n∈{1,2,3…n};

[0016] Take the maximum value from the similarity set as the evaluation value Score of the corresponding candidate enterprise to select the target enterprise based on the similarity n,i =max{Simi n,1 ,Simi n,2 …Simi n,P}, n∈{1,2,3…n}.

[0017] In a possible implementation, the method further includes the following steps:

[0018] An evaluation threshold is set, and at least one type of target government data is selected from a plurality of types of government data according to the loan access index, and corresponding change rate thresholds are set for the selected at least one type of target government data.

[0019] In a possible implementation, the step of selecting a second enterprise to enter a credit review node from the candidate enterprises according to the evaluation value of each candidate enterprise and the change in the government affairs data of each candidate enterprise includes the following steps:

[0020] Compare the calculated evaluation value of each candidate enterprise with a preset evaluation threshold, and select the candidate enterprises whose evaluation values ​​are greater than the evaluation threshold as the preparatory candidate enterprises to enter the credit review node;

[0021] The target government affairs data and the corresponding change rate of the preliminary candidate enterprises are obtained, and the preliminary candidate enterprises whose change rates corresponding to each target government affairs data are greater than the corresponding set change rate threshold are selected as the second enterprises to enter the credit review node.

[0022] In a possible implementation manner, the method further comprises the following steps:

[0023] Calculate the first pass rate of the preliminary candidate enterprises, and adjust the evaluation threshold based on the preset target first pass rate; and calculate the second pass rate of the second enterprise, and adjust the corresponding change rate threshold for at least one type of target government data selected based on the preset target second pass rate.

[0024] In a possible implementation, the government data includes one or more types of housing data, social security data, income tax payment data, and value-added tax payment data.

[0025] In a second aspect, the present application provides a device for selecting a target enterprise based on similarity, which is applied to a server or a terminal device, and the device includes:

[0026] An acquisition module, used to pre-access a target data platform and acquire credit information of a plurality of benchmarking enterprises that meet the credit acquisition conditions from the target data platform, wherein the credit information includes first government affairs data;

[0027] A screening module, used to screen out candidate enterprises whose credit information does not meet preset credit conditions from a plurality of first enterprises to be entered into a credit review node;

[0028] A determination module is used to obtain the second government affairs data of each candidate enterprise, and respectively calculate the similarity between the second government affairs data of the candidate enterprise and the first government affairs data of each benchmark enterprise in a plurality of benchmark enterprises, and determine the corresponding maximum similarity value as the evaluation value of the candidate enterprise for selecting a target enterprise based on the similarity;

[0029] The selection module is used to select a second enterprise to enter the credit review node from the candidate enterprises according to the evaluation value of each candidate enterprise and the change of the government affairs data of each candidate enterprise.

[0030] In a third aspect, the present application provides an electronic device comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method for selecting a target enterprise based on similarity as described in any one of the first aspects are performed.

[0031] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for selecting a target enterprise based on similarity as described in any one of the first aspects are executed.

[0032] The present embodiment provides a method and device for selecting a target enterprise based on similarity, which pre-accesses to a target data platform and obtains credit information of multiple benchmarking enterprises that meet credit acquisition conditions from the target data platform, wherein the credit information includes first government data; from multiple first enterprises to enter the credit review node, select candidate enterprises whose credit information does not meet the preset credit conditions; for each candidate enterprise, obtain the second government data of the candidate enterprise, and respectively calculate the similarity between the second government data of the candidate enterprise and the first government data of each benchmarking enterprise in multiple benchmarking enterprises, and determine the corresponding maximum similarity value as the evaluation value of the candidate enterprise for selecting the target enterprise based on similarity; according to the evaluation value of each candidate enterprise and the change of the government data of each candidate enterprise, select the second enterprise to enter the credit review node from the candidate enterprises. In this way, in the absence of financial data of small and micro enterprises, the government data can be used to calculate the similarity value between the candidate enterprise and the small and micro enterprises that have obtained loans and have good credit, and then determine the small and micro enterprises to enter the credit review node, effectively solving the problem of low access rate of small and micro enterprises under common rules, and improving access efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0034] Figure 1 A flowchart of a method for selecting a target enterprise based on similarity according to an embodiment of the present application is shown;

[0035] Figure 2 A flowchart of selecting a second enterprise to enter a credit review node according to an embodiment of the present application is shown;

[0036] Figure 3 A schematic diagram showing the structure of an apparatus for selecting a target enterprise based on similarity according to an embodiment of the present application is shown;

[0037] Figure 4 A structural block diagram of an electronic device described in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0038] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the present application, or remove one or more operations from the flowchart.

[0039] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0040] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0041] In view of the technical problems raised by the background technology, the present application provides a method and device for selecting target enterprises based on similarity, which can automatically select small and micro enterprises to enter the credit review node based on the similarity calculation of government data in the absence of financial data, thereby improving the access efficiency.

[0042] In one embodiment, see the attached specification Figure 1 The present application provides a method for selecting a target enterprise based on similarity, which is applied to a server or a terminal device. The method comprises the following steps:

[0043] S1. Access a target data platform in advance and obtain credit information of multiple benchmarking enterprises that meet credit acquisition conditions from the target data platform, wherein the credit information includes first government affairs data;

[0044] S2. Screening out candidate enterprises whose credit information does not meet preset credit conditions from the first enterprises to be entered into the credit review node;

[0045] S3. For each candidate enterprise, obtain the second government data of the candidate enterprise, and respectively calculate the similarity between the second government data of the candidate enterprise and the first government data of each of the multiple benchmarking enterprises, and determine the corresponding maximum similarity value as the evaluation value of the candidate enterprise for selecting the target enterprise based on the similarity;

[0046] S4. Select a second enterprise to enter the credit review node from the candidate enterprises based on the evaluation value of each candidate enterprise and the changes in the government affairs data of each candidate enterprise.

[0047] Specifically, in step S1, the target data platform may be a data platform of an institutional risk control system. In the application, the information of small and micro enterprises with financial data (with loan records) can be directly obtained by accessing the data platform of the institutional risk control system, and the credit of these small and micro enterprises obtained is guaranteed to be good, which can be used as subsequent benchmarking enterprises. Among them, satisfying the credit acquisition conditions includes: having loan information and having designated loan information within the most recent preset time period, and no abnormal events have occurred within a specified time period after the loan period with designated loan information.

[0048] For example, the credit acquisition conditions that benchmark companies need to meet are: (1) small and micro enterprises that have been loaned must have obtained inclusive small and micro enterprise loans in the past two years; (2) enterprises that have obtained inclusive small and micro enterprise loans must have undergone a performance period without any concerns, substandard, suspicious, or loss situations.

[0049] After the target enterprise is determined, its government data is further extracted. The government data includes one or more of housing data, social security data, income tax payment data, and value-added tax payment data. In an embodiment of the present application, the government data includes the average amount of value-added tax payments in the past three months, the average amount of value-added tax payments in the past six months, the average amount of value-added tax payments in the past twelve months, the average year-on-year growth rate of the average amount of value-added tax payments in the past three months, the average year-on-year growth rate of the average amount of value-added tax payments in the past six months, the average year-on-year growth rate of the average amount of value-added tax payments in the past twelve months, the average income tax payment amount in the past three months, the average income tax payment amount in the past six months, the average income tax payment amount in the past twelve months, the average year-on-year growth rate of the average income tax payment amount in the past three months, the average year-on-year growth rate of the average income tax payment amount in the past six months, the average year-on-year growth rate of the average income tax payment amount in the past twelve months, the average amount of basic pension insurance (unit payment) payment for employees in the past three months, and the average amount of basic pension insurance (unit payment) payment for employees in the past six months. The variables include the average monthly basic pension insurance (paid by the employer) payment amount for employees, the average basic pension insurance (paid by the employer) payment amount for employees in the past twelve months, the year-on-year growth rate of the average basic pension insurance (paid by the employer) payment amount for employees in the past three months, the year-on-year growth rate of the average basic pension insurance (paid by the employer) payment amount for employees in the past six months, the year-on-year growth rate of the average basic pension insurance (paid by the employer) payment amount for employees in the past twelve months, the average housing provident fund contribution amount in the past three months, the average housing provident fund contribution amount in the past six months, the average housing provident fund contribution amount in the past twelve months, the year-on-year growth rate of the average housing provident fund contribution amount in the past three months, the year-on-year growth rate of the average housing provident fund contribution amount in the past six months, and the year-on-year growth rate of the average housing provident fund contribution amount in the past twelve months, for a total of 24 variables.

[0050] Step S2 is to select small and micro enterprises that have no loan records or limited loan records from the small and micro enterprises that meet the loan application conditions as candidate enterprises. These candidate enterprises are called credit-free households.

[0051] In one embodiment, P benchmarking enterprises are obtained from the data platform of the institutional risk control system to form a benchmarking enterprise data set; N candidate enterprises are screened out to form a candidate enterprise data set. For example, the candidate enterprise data set includes 20 candidate enterprises, the benchmarking enterprise data set includes 5 benchmarking enterprises, and the government affairs data of each candidate enterprise and each benchmarking enterprise (where V1-V24 represents 24 variables) are shown in Table 1, and the maximum and minimum values ​​of each variable are counted to normalize the corresponding variables based on the maximum and minimum values. The results after normalization are shown in Table 2.

[0052] v1 v2 v3 v4 v5 v6 v7 v8 v9 v10 v11 v12 v13 v14 v15 v16 v17 v18 v19 v20 v21 v22 v23 v24 Seed Enterprise P1 190 553 550 986 884 222 932 536 414 355 434 401 886 916 232 123 691 477 872 323 553 702 107 712 Seed Enterprise P2 518 865 142 729 58 774 88 965 588 181 167 189 941 495 777 668 193 396 824 655 840 715 813 175 Seed Enterprise P3 217 21 474 795 148 35 45 343 67 942 768 159 597 624 620 121 178 994 536 176 522 98 148 379 Seed Enterprise P4 691 339 617 444 911 468 117 14 418 208 27 187 436 119 911 791 802 776 132 671 488 623 349 813 Seed Enterprise P5 519 404 111 415 700 293 729 699 836 237 324 915 876 861 720 523 117 896 511 940 424 883 705 810 Credit white household N1 767 804 921 878 570 309 671 705 600 371 901 521 577 729 52 101 886 573 480 570 170 871 517 796 Credit white household N2 53 286 172 437 589 699 332 884 284 10 198 919 801 688 147 238 592 30 168 224 737 608 944 188 Credit white household N3 692 432 704 977 27 675 896 745 472 970 960 275 941 223 74 204 799 673 164 707 826 205 824 36 Credit white household N4 893 827 854 423 572 197 506 357 411 685 762 858 194 910 308 21 584 960 571 3 21 250 865 215 Credit white household N5 422 385 239 929 19 284 200 298 477 480 894 559 642 849 852 718 120 661 289 861 981 826 479 250 Credit white household N6 333 3 289 605 309 845 659 629 673 740 501 342 688 396 293 597 597 839 340 933 78 271 39 839 Credit white household N7 553 563 556 208 385 450 275 646 674 462 913 594 664 907 415 206 966 991 191 932 565 563 161 457 Credit white household N8 792 566 511 824 499 720 728 100 809 75 189 510 856 260 863 939 845 156 480 550 49 128 441 47 Credit white household N9 19 7 507 419 883 50 907 135 330 241 76 674 574 311 102 518 145 9 702 36 545 484 922 729 Credit white household N10 984 534 229 812 140 756 749 517 357 977 747 994 25 655 754 383 403 164 400 208 688 131 19 557 Credit white household N11 773 638 920 97 311 697 348 494 575 231 836 290 850 557 257 382 487 881 636 380 721 929 100 612 Credit white household N12 112 208 248 49 169 987 579 924 278 181 736 97 568 947 126 469 244 484 24 153 658 826 827 450 Credit white household N13 19 870 339 21 222 688 607 112 681 436 549 205 864 7 628 677 867 453 392 819 144 940 947 241 Credit white household N14 890 564 760 42 267 992 468 36 699 620 533 708 617 1 379 503 174 174 131 723 183 571 445 404 Credit white household N15 913 308 475 172 606 789 445 248 84 687 202 354 165 877 826 852 377 853 45 313 439 636 18 560 Credit white household N16 422 992 139 995 110 706 371 599 816 149 551 936 464 533 521 993 526 798 16 600 961 733 245 202 Credit white household N17 748 616 798 479 862 941 19 415 864 3 14 3 828 21 835 565 57 976 579 469 156 640 262 55 Credit white household N18 614 83 258 857 874 429 156 614 896 527 628 882 738 366 49 689 382 814 827 780 188 223 266 642 Credit white household N19 3 262 352 416 807 960 755 499 195 229 845 109 136 769 271 805 952 250 740 920 7 279 497 580 Credit white household N20 616 601 200 450 655 957 312 137 200 812 832 32 272 5 524 691 752 365 379 476 17 647 89 499 min 3 3 111 21 19 35 19 14 67 3 14 3 25 1 49 21 57 9 16 3 7 98 18 36 max 984 992 921 995 911 992 932 965 896 977 960 994 941 947 911 993 966 994 872 940 981 940 947 839

[0053] Table 1

[0054] v1 v2 v3 v4 v5 v6 v7 v8 v9 v10 v11 v12 v13 v14 v15 v16 v17 v18 v19 v20 v21 v22 v23 v24 Seed Enterprise P1 0.19 0.56 0.54 0.99 0.97 0.2 1 0.55 0.42 0.36 0.44 0.4 0.94 0.97 0.21 0.1 0.7 0.48 1 0.34 0.56 0.72 0.1 0.84 Seed Enterprise P2 0.52 0.87 0.04 0.73 0.04 0.77 0.08 1 0.63 0.18 0.16 0.19 1 0.52 0.84 0.67 0.15 0.39 0.94 0.7 0.86 0.73 0.86 0.17 Seed Enterprise P3 0.22 0.02 0.45 0.79 0.14 0 0.03 0.35 0 0.96 0.8 0.16 0.62 0.66 0.66 0.1 0.13 1 0.61 0.18 0.53 0 0.14 0.43 Seed Enterprise P4 0.7 0.34 0.62 0.43 1 0.45 0.11 0 0.42 0.21 0.01 0.19 0.45 0.12 1 0.79 0.82 0.78 0.14 0.71 0.49 0.62 0.36 0.97 Seed Enterprise P5 0.53 0.41 0 0.4 0.76 0.27 0.78 0.72 0.93 0.24 0.33 0.92 0.93 0.91 0.78 0.52 0.07 0.9 0.58 1 0.43 0.93 0.74 0.96 Credit white household N1 0.78 0.81 1 0.88 0.62 0.29 0.71 0.73 0.64 0.38 0.94 0.52 0.6 0.77 0 0.08 0.91 0.57 0.54 0.61 0.17 0.92 0.54 0.95 Credit white household N2 0.05 0.29 0.08 0.43 0.64 0.69 0.34 0.91 0.26 0.01 0.19 0.92 0.85 0.73 0.11 0.22 0.59 0.02 0.18 0.24 0.75 0.61 1 0.19 Credit white household N3 0.7 0.43 0.73 0.98 0.01 0.67 0.96 0.77 0.49 0.99 1 0.27 1 0.23 0.03 0.19 0.82 0.67 0.17 0.75 0.84 0.13 0.87 0 Credit white household N4 0.91 0.83 0.92 0.41 0.62 0.17 0.53 0.36 0.41 0.7 0.79 0.86 0.18 0.96 0.3 0 0.58 0.97 0.65 0 0.01 0.18 0.91 0.22 Credit white household N5 0.43 0.39 0.16 0.93 0 0.26 0.2 0.3 0.49 0.49 0.93 0.56 0.67 0.9 0.93 0.72 0.07 0.66 0.32 0.92 1 0.86 0.5 0.27 Credit white household N6 0.34 0 0.22 0.6 0.33 0.85 0.7 0.65 0.73 0.76 0.51 0.34 0.72 0.42 0.28 0.59 0.59 0.84 0.38 0.99 0.07 0.21 0.02 1 Credit white household N7 0.56 0.57 0.55 0.19 0.41 0.43 0.28 0.66 0.73 0.47 0.95 0.6 0.7 0.96 0.42 0.19 1 1 0.2 0.99 0.57 0.55 0.15 0.52 Credit white household N8 0.8 0.57 0.49 0.82 0.54 0.72 0.78 0.09 0.9 0.07 0.18 0.51 0.91 0.27 0.94 0.94 0.87 0.15 0.54 0.58 0.04 0.04 0.46 0.01 Credit white household N9 0.02 0 0.49 0.41 0.97 0.02 0.97 0.13 0.32 0.24 0.07 0.68 0.6 0.33 0.06 0.51 0.1 0 0.8 0.04 0.55 0.46 0.97 0.86 Credit white household N10 1 0.54 0.15 0.81 0.14 0.75 0.8 0.53 0.35 1 0.77 1 0 0.69 0.82 0.37 0.38 0.16 0.45 0.22 0.7 0.04 0 0.65 Credit white household N11 0.78 0.64 1 0.08 0.33 0.69 0.36 0.5 0.61 0.23 0.87 0.29 0.9 0.59 0.24 0.37 0.47 0.89 0.72 0.4 0.73 0.99 0.09 0.72 Credit white household N12 0.11 0.21 0.17 0.03 0.17 0.99 0.61 0.96 0.25 0.18 0.76 0.09 0.59 1 0.09 0.46 0.21 0.48 0.01 0.16 0.67 0.86 0.87 0.52 Credit white household N13 0.02 0.88 0.28 0 0.23 0.68 0.64 0.1 0.74 0.44 0.57 0.2 0.92 0.01 0.67 0.67 0.89 0.45 0.44 0.87 0.14 1 1 0.26 Credit white household N14 0.9 0.57 0.8 0.02 0.28 1 0.49 0.02 0.76 0.63 0.55 0.71 0.65 0 0.38 0.5 0.13 0.17 0.13 0.77 0.18 0.56 0.46 0.46 Credit white household N15 0.93 0.31 0.45 0.16 0.66 0.79 0.47 0.25 0.02 0.7 0.2 0.35 0.15 0.93 0.9 0.85 0.35 0.86 0.03 0.33 0.44 0.64 0 0.65 Credit white household N16 0.43 1 0.03 1 0.1 0.7 0.39 0.62 0.9 0.15 0.57 0.94 0.48 0.56 0.55 1 0.52 0.8 0 0.64 0.98 0.75 0.24 0.21 Credit white household N17 0.76 0.62 0.85 0.47 0.95 0.95 0 0.42 0.96 0 0 0 0.88 0.02 0.91 0.56 0 0.98 0.66 0.5 0.15 0.64 0.26 0.02 Credit white household N18 0.62 0.08 0.18 0.86 0.96 0.41 0.15 0.63 1 0.54 0.65 0.89 0.78 0.39 0 0.69 0.36 0.82 0.95 0.83 0.19 0.15 0.27 0.75 Credit white household N19 0 0.26 0.3 0.41 0.88 0.97 0.81 0.51 0.15 0.23 0.88 0.11 0.12 0.81 0.26 0.81 0.98 0.24 0.85 0.98 0 0.21 0.52 0.68 Credit white household N20 0.62 0.6 0.11 0.44 0.71 0.96 0.32 0.13 0.16 0.83 0.86 0.03 0.27 0 0.55 0.69 0.76 0.36 0.42 0.5 0.01 0.65 0.08 0.58

[0055] Table 2

[0056] The cosine similarity between each candidate enterprise in the candidate enterprise data set and each benchmark enterprise in the benchmark enterprise data set is calculated by the following formula:

[0057]

[0058] Among them, X j ·X i represents the dot product, ||X j ||*||X i || represents the product of Euclidean norms; x i represents the benchmark enterprise data set, where i ranges from 1 to P, representing the id of the benchmark enterprise; x j Represents the candidate enterprise dataset, where j ranges from 1 to N and represents the id of the candidate enterprise.

[0059] The calculated cosine similarity between each candidate enterprise in the candidate enterprise data set and each benchmark enterprise in the benchmark enterprise data set is used to form a similarity set {Simi n,1 ,Simi n,2 …Simi n,P}, n∈{1,2,3…n}; take the maximum value from the similarity set as the evaluation value Score of the corresponding candidate enterprise to select the target enterprise based on similarity n,i =max{Simi n,1 ,Simi n,2 …Simi n,P}, n∈{1,2,3…n}. The calculated cosine similarity between each candidate enterprise in the candidate enterprise data set and each benchmark enterprise in the benchmark enterprise data set and the corresponding maximum value are shown in Table III.

[0060]

[0061] Table 3

[0062] See the instruction manual Figure 2 In step S4, the method of selecting a second enterprise to enter the credit review node from the candidate enterprises based on the value of each candidate selecting the target enterprise based on similarity and the change of the government affairs data of each candidate enterprise includes the following steps:

[0063] S401, comparing the calculated value of each candidate target enterprise selected based on similarity with a preset credit evaluation value threshold, and selecting the candidate enterprise whose credit evaluation value is greater than the credit evaluation value threshold as a preliminary candidate enterprise to enter the credit review node;

[0064] S402, obtaining the target government affairs data and the corresponding change rate of the preliminary candidate enterprises, and selecting the preliminary candidate enterprises whose corresponding change rate of each target government affairs data is greater than the corresponding set change rate threshold as the second enterprises to enter the credit review node.

[0065] It should be noted that before executing step S4, a credit assessment value threshold is set in advance, and at least one type of target government data is selected from multiple types of government data based on loan access indicators, and corresponding change rate thresholds are set for the selected at least one type of target government data.

[0066] For example, setting the evaluation threshold θ 1 , the evaluation values ​​of the 20 candidate enterprises N1-N20 representing credit-free households in the candidate enterprise data set calculated in step S401 are compared with the evaluation threshold θ 1 Compare, if the evaluation value is greater than the evaluation threshold θ 1 , the corresponding candidate enterprises are screened out from the candidate enterprise data set as preliminary candidate enterprises to enter the credit review node.

[0067] Among them, if the evaluation threshold θ 1 If the evaluation threshold θ is set to 0.843, it means that N7, N15, N5, N18, and N1 of the candidate enterprises with no credit records are selected as the candidate enterprises to be prepared for the credit review node, and the passing rate of the candidate enterprises to be prepared for the credit review node is 30%. 1 Set to 0.825, it means that N6, N16, N7, N15, N5, N18, and N1 of the candidate enterprises with no credit records are selected as the preliminary candidate enterprises to enter the credit review node, and the passing rate of the preliminary candidate enterprises to enter the credit review node is 40%. It can be seen that by setting the evaluation threshold θ 1 It can determine the pass rate of candidate enterprises to be reviewed at the credit review node. Therefore, if you want to initially relax the loan access conditions for credit-free households, you can adjust the evaluation threshold θ appropriately. 1 .

[0068] The obtained candidate enterprises to be entered into the credit review node do not mean that they can obtain loan access qualifications. This is because there are hard rules set in terms of government data. Therefore, it is necessary to continue to execute step S402. Among them, as mentioned above, the government data includes multiple types. It is necessary to select at least one type of target government data from the government data according to the loan access index to set a corresponding change rate threshold for it. For example, in the loan access index, social security data, the housing data, and the value-added tax payment data are relatively important, so corresponding change rate thresholds can be set for them. In one embodiment, the growth rate of the cumulative deposit amount of employees' basic pension insurance (unit payment) in the past six months compared with the same period last year, the growth rate of the cumulative deposit amount of provident fund in the past six months compared with the same period last year, and the growth rate of the cumulative amount of value-added tax of the enterprise in the past six months compared with the same period last year are respectively set as thresholds for social security data threshold data θ 2 , Housing data threshold θ 3 , VAT payment data threshold θ 4 .

[0069] When making the comparison, it is necessary to first obtain the government data of the candidate enterprises to be included in the credit review node, including the growth rate of the accumulated amount of basic pension insurance (paid by the employer) for employees in the past six months compared with the same period last year, the growth rate of the accumulated amount of provident fund in the past six months compared with the same period last year, and the growth rate of the accumulated amount of value-added tax in the past six months compared with the same period last year; then compare them with the set social security threshold data θ 2 , Housing data threshold θ 3 , VAT payment data threshold θ 4 For comparison, if the growth rate of the accumulated amount of basic pension insurance (paid by the employer) for employees in the past six months compared with the same period of the previous year, the growth rate of the accumulated amount of provident fund in the past six months compared with the same period of the previous year, and the growth rate of the accumulated amount of value-added tax of the enterprise in the past six months compared with the same period of the previous year are all greater than the set social security category threshold value θ 2 , Housing data threshold θ 3 , VAT payment data threshold θ 4 , the corresponding preliminary candidate enterprise to enter the credit review node will be used as the second enterprise to enter the final credit review node.

[0070] Among them, the evaluation threshold θ 10.825, and the candidate enterprises are N6, N16, N7, N15, N5, N18, and N1. First, the government data of N6, N16, N7, N15, N5, N18, and N1 are obtained, including the growth rate of the accumulated amount of basic pension insurance (paid by the unit) for employees in the past six months compared with the same period last year, the growth rate of the accumulated amount of provident fund in the past six months compared with the same period last year, and the growth rate of the accumulated amount of value-added tax in the past six months compared with the same period last year as shown in Table 4. If the threshold value of social security data is set as θ 2 is 0.3, the threshold value of housing data is θ 3 is 0.28, the VAT payment data threshold θ 4 is 0.188, then the second enterprises to be entered into the credit review node are N6, N7, N11, and N18, and the final passing rate of the second enterprises to be entered into the credit review node is 20%. Similarly, the change rate threshold of the corresponding target government data can be set according to the preset passing rate of the second enterprises to be entered into the credit review node.

[0071]

[0072] Table 4

[0073] It can be seen that the method for selecting target enterprises based on similarity provided in this application can effectively and deeply utilize government data and previous loan enterprise information based on the similarity algorithm to conduct credit assessment on small and micro enterprises with no credit records, thereby increasing more small and micro enterprises that meet the requirements to enter the credit review stage, improving the situation where small and micro enterprises that are really in difficulty and need financial support cannot obtain loans, and while improving the access efficiency, optimizing the cold start problem of credit investigation for small and micro enterprises.

[0074] Based on the same inventive concept, an apparatus for selecting a target enterprise based on similarity is also provided in an embodiment of the present application. Since the principle of solving the problem by the apparatus in the embodiment of the present application is similar to the method for selecting a target enterprise based on similarity in the embodiment of the present application, the implementation of the apparatus can refer to the implementation of the method, and the repeated parts will not be repeated.

[0075] As the instruction manual Figure 3 As shown, an embodiment of the present application provides a device for selecting a target enterprise based on similarity, which is applied to a server or a terminal device, and the device includes:

[0076] The acquisition module 301 is used to pre-access the target data platform and acquire the credit information of multiple benchmark enterprises that meet the credit acquisition conditions from the target data platform, wherein the credit information includes the first government data; the credit acquisition conditions include having loan information and having designated loan information within the latest preset time period; and no abnormal event occurs within a specified time period after the loan period with the designated loan information;

[0077] A screening module 302 is used to screen out candidate enterprises whose credit information does not meet the preset credit conditions from the first enterprises to be entered into the credit review node;

[0078] The determination module 303 is used to obtain the second government affairs data of each candidate enterprise, and respectively calculate the similarity between the second government affairs data of the candidate enterprise and the first government affairs data of each benchmark enterprise in a plurality of benchmark enterprises, and determine the corresponding maximum similarity value as the value for selecting the target enterprise based on the similarity of the candidate enterprise;

[0079] The selection module 304 is used to select a second enterprise to enter the credit review node from the candidate enterprises according to the value of selecting the target enterprise based on the similarity of each candidate and the change of the government affairs data of each candidate enterprise.

[0080] In one embodiment, the determination module 303 respectively calculates the similarity between the second government data of the candidate enterprise and the first government data of each of the multiple benchmarking enterprises, and determines the corresponding maximum similarity as the evaluation value of the candidate enterprise for selecting the target enterprise based on the similarity, including: Calculate the similarity between the second government data of the candidate enterprise and the first government data of each benchmark enterprise in multiple benchmark enterprises respectively; where N represents the number of candidate enterprises, P represents the number of benchmark enterprises, and x i represents a benchmarking enterprise data set consisting of multiple benchmarking enterprises, where i ranges from 1 to P, representing the id of each benchmarking enterprise; x j represents a candidate enterprise data set composed of multiple candidate enterprises, where j ranges from 1 to N, representing the id of each candidate enterprise; the calculated similarity between the second government data of each candidate enterprise and the first government data of each benchmark enterprise in multiple benchmark enterprises is formed into a similarity set {Simi n,1 ,Simi n,2 …Simi n,P}, n∈{1,2,3…n}; take the maximum value from the similarity set as the evaluation value Score of the corresponding candidate enterprise to select the target enterprise based on similarity n,i =max{Simi n,1 ,Simi n,2 …Simi n,P}, n∈{1,2,3…n}.

[0081] In one embodiment, the device further comprises:

[0082] The setting module is used to set an evaluation threshold, and select at least one type of target government data from multiple types of government data according to the loan access index, and set corresponding change rate thresholds for the selected at least one type of target government data.

[0083] In one embodiment, the selection module 304 selects a second enterprise to enter the credit review node from the candidate enterprises based on the evaluation value of each candidate enterprise and the change of the government data of each candidate enterprise, including: comparing the calculated evaluation value of each candidate enterprise with a preset evaluation threshold, and selecting the candidate enterprises with evaluation values ​​greater than the evaluation threshold as preliminary candidate enterprises to enter the credit review node; obtaining the target government data and the corresponding change rate of the preliminary candidate enterprises, and selecting the preliminary candidate enterprises with each target government data corresponding change rate greater than the corresponding set change rate threshold as the second enterprise to enter the credit review node.

[0084] In one embodiment, the device further comprises:

[0085] An adjustment module is used to calculate the first pass rate of the preliminary candidate enterprises and adjust the evaluation threshold based on the preset target first pass rate; and to calculate the second pass rate of the second enterprise and adjust the corresponding change rate threshold for at least one type of target government data selected based on the preset target second pass rate.

[0086] In one embodiment, the government data includes one or more types of housing data, social security data, income tax payment data, and value-added tax payment data.

[0087] The present application provides a device for selecting a target enterprise based on similarity, which pre-accesses a target data platform through an acquisition module and obtains credit information of multiple benchmarking enterprises that meet credit acquisition conditions from the target data platform, wherein the credit information includes first government data; through a screening module, a candidate enterprise whose credit information does not meet the preset credit conditions is screened out from multiple first enterprises to enter the credit review node; through a determination module, the second government data of the candidate enterprise is obtained for each candidate enterprise, and the similarity between the second government data of the candidate enterprise and the first government data of each benchmarking enterprise in multiple benchmarking enterprises is calculated respectively, and the corresponding maximum similarity value is determined as the evaluation value of the candidate enterprise to select the target enterprise based on similarity; through a selection module, according to the evaluation value of each candidate enterprise and the change of the government data of each candidate enterprise, the second enterprise to enter the credit review node is selected from the candidate enterprises. In this way, in the case of the lack of financial data of small and micro enterprises, the government data can be used to calculate the similarity value between the candidate enterprise and the small and micro enterprises that have obtained loans and have good credit, and then the small and micro enterprises to enter the credit review node can be determined, which effectively solves the problem of low access rate of small and micro enterprises under common rules and improves access efficiency.

[0088] Based on the same concept of the present invention, the specification is attached Figure 4 As shown, the structure of an electronic device 400 provided in an embodiment of the present application includes: at least one processor 401, at least one network interface 404 or other user interface 403, a memory 405, and at least one communication bus 402. The communication bus 402 is used to realize the connection and communication between these components. The electronic device 400 optionally includes a user interface 403, including a display (for example, a touch screen, LCD, CRT, holographic imaging (Holographic) or projection (Projector), etc.), a keyboard or a pointing device (for example, a mouse, a trackball (trackball), a touch pad or a touch screen, etc.).

[0089] The memory 405 may include a read-only memory and a random access memory, and provides instructions and data to the processor 401. A portion of the memory 405 may also include a non-volatile random access memory (NVRAM).

[0090] In some implementations, the memory 405 stores the following elements, executable modules or data structures, or a subset thereof, or an extended set thereof:

[0091] Operating system 4051, including various system programs for implementing various basic services and processing hardware-based tasks;

[0092] The application module 4052 includes various application programs, such as a launcher, a media player, a browser, etc., which are used to implement various application services.

[0093] In an embodiment of the present application, by calling the program or instructions stored in the memory 405, the processor 401 is used to execute steps in a method for selecting target enterprises based on similarity. In the absence of financial data, it can automatically select small and micro enterprises to enter the credit review node based on the similarity calculation of government data, thereby improving the access efficiency.

[0094] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the method for selecting a target enterprise based on similarity are executed.

[0095] Specifically, the storage medium can be a general storage medium, such as a mobile disk, a hard disk, etc. When the computer program on the storage medium is run, the above-mentioned method of selecting a target enterprise based on similarity can be executed.

[0096] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0097] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0098] In addition, each functional unit in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0099] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0100] Finally, it should be noted that the above embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the aforementioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the aforementioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application. They should all be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for selecting a target enterprise based on similarity, characterized in that: Applied to a server or a terminal device, the method comprises the following steps: Accessing a target data platform in advance and obtaining credit information of a plurality of benchmarking enterprises that meet credit acquisition conditions from the target data platform, wherein the credit information includes first government affairs data; Screening out candidate enterprises whose credit information does not meet preset credit conditions from multiple first enterprises to be entered into the credit review node; For each candidate enterprise, obtain the second government affairs data of the candidate enterprise, and respectively calculate the similarity between the second government affairs data of the candidate enterprise and the first government affairs data of each of the multiple benchmarking enterprises, and determine the corresponding maximum similarity value as the evaluation value of the candidate enterprise for selecting the target enterprise based on the similarity; According to the evaluation value of each candidate enterprise and the change of the government affairs data of each candidate enterprise, a second enterprise to enter the credit review node is selected from the candidate enterprises.

2. A method for selecting a target enterprise based on similarity according to claim 1, characterized in that: The credit conditions include: Having loan information and having designated loan information within the most recent preset time period; No unusual events occurred within a specified time period after the loan period with specified loan information.

3. A method for selecting a target enterprise based on similarity according to claim 2, characterized in that: The method of respectively calculating the similarity between the second government affairs data of the candidate enterprise and the first government affairs data of each of the plurality of benchmarking enterprises, and determining the corresponding maximum similarity value as the evaluation value of the candidate enterprise for selecting the target enterprise based on the similarity, comprises the following steps: By formula Calculate the similarity between the second government data of the candidate enterprise and the first government data of each benchmark enterprise in multiple benchmark enterprises respectively; where N represents the number of candidate enterprises, P represents the number of benchmark enterprises, and x i represents a benchmarking enterprise data set consisting of multiple benchmarking enterprises, where i ranges from 1 to P, representing the id of each benchmarking enterprise; x j represents a candidate enterprise data set consisting of multiple candidate enterprises, where j ranges from 1 to N, representing the id of each candidate enterprise; The calculated similarity between the second government data of each candidate enterprise and the first government data of each benchmark enterprise in multiple benchmark enterprises is combined into a similarity set {Simi n,1 ,Simi n,2 …Simi n,P }, n∈{1,2,3…n}; Take the maximum value from the similarity set as the evaluation value Score of the corresponding candidate enterprise to select the target enterprise based on the similarity n,i =max{Simi n,1 ,Simi n,2 …Simi n,P }, n∈{1,2,3…n}.

4. A method for selecting a target enterprise based on similarity according to claim 3, characterized in that: The method further comprises the following steps: An evaluation threshold is set, and at least one type of target government data is selected from a plurality of types of government data according to the loan access index, and corresponding change rate thresholds are set for the selected at least one type of target government data.

5. The method for selecting a target enterprise based on similarity according to claim 4, characterized in that: The step of selecting a second enterprise to enter the credit review node from the candidate enterprises according to the evaluation value of each candidate enterprise and the change of the government affairs data of each candidate enterprise comprises the following steps: Compare the calculated evaluation value of each candidate enterprise with a preset evaluation threshold, and select the candidate enterprises whose evaluation values ​​are greater than the evaluation threshold as the preliminary candidate enterprises to enter the credit review node; The target government affairs data and the corresponding change rate of the preliminary candidate enterprises are obtained, and the preliminary candidate enterprises whose change rates corresponding to each target government affairs data are greater than the corresponding set change rate threshold are selected as the second enterprises to enter the credit review node.

6. A method for selecting a target enterprise based on similarity according to claim 5, characterized in that: The method further comprises the following steps: Calculate the first pass rate of the preliminary candidate enterprises, and adjust the evaluation threshold based on the preset target first pass rate; and calculate the second pass rate of the second enterprise, and adjust the corresponding change rate threshold for at least one type of target government data selected based on the preset target second pass rate.

7. The method for selecting a target enterprise based on similarity according to claim 6, characterized in that: The government affairs data includes one or more types of housing data, social security data, income tax payment data, and value-added tax payment data.

8. A device for selecting a target enterprise based on similarity, characterized in that: Applied to a server or a terminal device, the device comprises: An acquisition module, used to pre-access a target data platform and acquire credit information of a plurality of benchmarking enterprises that meet the credit acquisition conditions from the target data platform, wherein the credit information includes first government affairs data; A screening module, used to screen out candidate enterprises whose credit information does not meet preset credit conditions from a plurality of first enterprises to be entered into a credit review node; A determination module is used to obtain the second government affairs data of each candidate enterprise, and respectively calculate the similarity between the second government affairs data of the candidate enterprise and the first government affairs data of each benchmark enterprise in a plurality of benchmark enterprises, and determine the corresponding maximum similarity value as the value for selecting the target enterprise based on the similarity of the candidate enterprise; The selection module is used to select a second enterprise to enter the credit review node from the candidate enterprises according to the value of selecting the target enterprise based on similarity for each candidate and the change of the government affairs data of each candidate enterprise.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method for selecting a target enterprise based on similarity as described in any one of claims 1 to 7 are performed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for selecting a target enterprise based on similarity as claimed in any one of claims 1 to 7 are executed.

Images