CN117151797A - Enterprise credit assessment system based on comprehensive data analysis - Google Patents
Enterprise credit assessment system based on comprehensive data analysis Download PDFInfo
- Publication number
- CN117151797A CN117151797A CN202311135624.XA CN202311135624A CN117151797A CN 117151797 A CN117151797 A CN 117151797A CN 202311135624 A CN202311135624 A CN 202311135624A CN 117151797 A CN117151797 A CN 117151797A
- Authority
- CN
- China
- Prior art keywords
- evaluation
- module
- credit
- unit
- enterprise
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007405 data analysis Methods 0.000 title claims abstract description 40
- 238000011156 evaluation Methods 0.000 claims abstract description 431
- 238000005065 mining Methods 0.000 claims abstract description 32
- 238000000605 extraction Methods 0.000 claims abstract description 29
- 238000004458 analytical method Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 3
- 238000013500 data storage Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 4
- 230000006872 improvement Effects 0.000 description 10
- 238000000034 method Methods 0.000 description 6
- 230000006399 behavior Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0282—Rating or review of business operators or products
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
- G06Q30/015—Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
- G06Q30/016—After-sales
Abstract
The invention relates to the technical field of enterprise assessment, in particular to an enterprise credit assessment system based on comprehensive data analysis. The system comprises a customer evaluation mining unit, an evaluation data analysis unit, a bad evaluation extraction unit, an after-sales analysis unit and a comprehensive credit evaluation unit. According to the invention, the customer evaluation is mined through the customer evaluation mining unit, the first evaluation and the subsequent evaluation are classified, the first evaluation type is conveniently judged through the evaluation data analysis unit, if the first evaluation type is the bad evaluation, the bad evaluation extraction unit is used for after-sales treatment, and the evaluation type of the customer evaluation after the treatment is analyzed through the after-sales analysis unit, so that the comprehensive credit evaluation unit is beneficial to evaluating the enterprise credit corresponding to the first evaluation and the enterprise credit corresponding to the subsequent evaluation respectively, and the customer can conveniently determine the initial marketing condition of the enterprise and the after-sales treatment condition when selecting the enterprise, and the accuracy and the comprehensiveness of the enterprise credit evaluation are beneficial.
Description
Technical Field
The invention relates to the technical field of enterprise assessment, in particular to an enterprise credit assessment system based on comprehensive data analysis.
Background
Along with the development and progress of society, the enterprise credit is more and more closely related to the economic benefit of the enterprise, a large amount of accumulated data in the enterprise can be found out from the data of the network by utilizing a data mining method, more data for evaluating the enterprise credit are provided with more scientific and reasonable basis for evaluating the enterprise credit, and the accuracy and the evaluation efficiency of the enterprise credit evaluation directly influence the selection of clients;
however, when mining the current enterprise data on the network to analyze enterprise credit, a manner of collecting customer evaluation is mostly adopted, for example: if the credit evaluation system only directly evaluates the credit of the enterprise according to the evaluation data of the customers after marketing, some evaluation data corresponding to the good evaluation are not necessarily good contents, the evaluation data corresponding to the bad evaluation are not uniform and poor contents, if the detailed data are not analyzed according to the good evaluation and the bad evaluation, the credit evaluation of the enterprise is incomplete, and the evaluation data after the after-sales processing of the good evaluation and the bad evaluation are inconvenient to analyze, so that the credit of the enterprise is not accurate, the credit of the enterprise cannot be evaluated according to the customer evaluation at the initial stage and the later stage, and the selection of the subsequent customers is affected.
Disclosure of Invention
The present invention is directed to an enterprise credit assessment system based on comprehensive data analysis, so as to solve the problems set forth in the background art.
In order to achieve the above object, the present invention provides an enterprise credit evaluation system based on comprehensive data analysis, which comprises a customer evaluation mining unit, an evaluation data analysis unit, a bad evaluation extraction unit, an after-sales analysis unit and a comprehensive credit evaluation unit;
the client evaluation mining unit is used for mining evaluation data of the current enterprise, wherein the evaluation data comprises primary evaluation and subsequent evaluation;
the evaluation data analysis unit is used for extracting keywords of primary evaluation in the evaluation data of the client evaluation mining unit, sequentially inputting a preset evaluation data set, and outputting a primary evaluation type, wherein the primary evaluation type comprises a good evaluation, a medium evaluation and a poor evaluation;
the evaluation data analysis unit is used for outputting evaluation data corresponding to the evaluation in the primary evaluation type, and calling a marketing list and customer evaluation content corresponding to the evaluation data to remind enterprise workers, and performing after-sales processing on the evaluation aiming at the customer evaluation;
the after-sales analysis unit is used for tracking the after-sales processing result of the poor evaluation extraction unit, extracting keywords of the subsequent evaluation through the client evaluation mining unit, inputting the keywords into the evaluation data set, and outputting the subsequent evaluation types, wherein the subsequent evaluation types also comprise good evaluation, medium evaluation and poor evaluation;
the comprehensive credit evaluation unit is used for respectively analyzing the enterprise credit of the first evaluation in the evaluation data analysis unit and the enterprise credit of the subsequent evaluation in the after-sales analysis unit by adopting a point system, and summarizing the first evaluation and the subsequent evaluation to comprehensively analyze the enterprise credit.
As a further improvement of the technical scheme, the client evaluation mining unit adopts a web crawler technology to mine evaluation data of the current enterprise from the webpage.
As a further improvement of the technical scheme, the customer evaluation mining unit further comprises an evaluation data classification module, wherein the evaluation data classification module is used for marking the time stamp of each evaluation of the customer, classifying the first evaluation and the subsequent evaluation according to the time stamp, wherein the time stamp date is the first evaluation, and the time stamp date is the subsequent evaluation.
As a further improvement of the technical scheme, the evaluation data analysis unit comprises an evaluation data set, a keyword extraction module and a type output module;
the evaluation data set is used for establishing a set of keyword features corresponding to the evaluation types;
the keyword extraction module is used for extracting texts in the primary evaluation separated by the evaluation data classification module, and keyword characteristics in the texts are extracted by adopting a keyword extraction algorithm;
the type output module is used for inputting the keywords in the keyword extraction module into the evaluation data set and outputting the primary evaluation type.
As a further improvement of the technical scheme, the difference evaluation extraction unit comprises a signal receiving module, a data calling module and an after-sales reminding module;
the signal receiving module is connected with the output end of the type output module and is used for receiving the difference evaluation output by the type output module and sending out a difference evaluation signal;
the data retrieval module is used for associating text content of customer evaluation with the marketing list, and searching the data storage library for the marketing list corresponding to the evaluation signal in the signal receiving module;
the after-sale reminding module is used for sending marketing orders and customer evaluation contents corresponding to the difference evaluation to enterprise workers through the communication tool after receiving the difference evaluation signals sent by the signal receiving module.
As a further improvement of the technical scheme, the after-sales analysis unit comprises an after-sales tracking module and a subsequent evaluation and evaluation module;
the after-sale tracking module is used for tracking the signals of finishing the after-sale processing of the bad evaluation, and extracting keywords of a subsequent evaluation text from the subsequent evaluation mined by the client evaluation mining unit;
the subsequent evaluation module is used for calling out an evaluation data set, inputting the keywords of the subsequent evaluation text extracted by the after-sale tracking module into the evaluation data set, and outputting the evaluation type.
As a further improvement of the technical scheme, the comprehensive credit evaluation unit comprises an integral preset module, an independent evaluation module and a comprehensive evaluation module;
the score presetting module is used for setting a score rule, deducting scores on the basis of initial scores when bad scores appear, presetting a credit grade threshold value, comparing the credit grade threshold value with the credit scores, and outputting credit grades;
the independent evaluation module is used for respectively analyzing the quantity of the primary evaluation and the subsequent evaluation corresponding to the good evaluation, the medium evaluation and the poor evaluation, respectively inputting the evaluation types into the integral rule of the integral preset module, respectively outputting the credit score of the primary evaluation and the credit score of the subsequent evaluation, and determining the enterprise credit corresponding to the primary evaluation and the subsequent evaluation;
the comprehensive evaluation module is used for summarizing the first evaluation type in the type output module and the integral rule of the integral preset module of the subsequent evaluation type input integral in the subsequent evaluation module, outputting comprehensive credit score and determining comprehensive credit of enterprises.
As a further improvement of the technical scheme, the comprehensive credit evaluation unit further comprises an evaluation score module, wherein the evaluation score module is used for setting waiting time when the client finishes marketing, outputting score rewards to the client account when the client finishes evaluation within the waiting time, and not rewarding outside the waiting time.
As a further improvement of the technical scheme, the comprehensive credit evaluation unit further comprises a trend prediction module, wherein the trend prediction module is connected with the output end of the comprehensive evaluation module and is used for analyzing credit score change trend of the comprehensive evaluation module according to a time sequence, constructing a prediction model by taking credit data as a training set, outputting a prediction result of the credit score, inputting the prediction result of the credit score into the integral preset module and outputting a prediction result of the credit grade.
Compared with the prior art, the invention has the beneficial effects that:
in the enterprise credit evaluation system based on comprehensive data analysis, the client evaluation is mined through the client evaluation mining unit, the first evaluation and the subsequent evaluation are classified, the first evaluation type is conveniently judged through the first evaluation data analysis unit, if the first evaluation type is the bad evaluation, the bad evaluation extraction unit is enabled to carry out after-sales treatment, and after the evaluation type of the client evaluation is analyzed and treated through the after-sales analysis unit, the enterprise credit corresponding to the first evaluation and the enterprise credit corresponding to the subsequent evaluation are conveniently evaluated through the comprehensive credit evaluation unit, the initial marketing condition of the enterprise and the after-sales treatment condition are conveniently determined when the client selects the enterprise, and the accuracy and the comprehensiveness of the enterprise credit evaluation are facilitated.
Drawings
FIG. 1 is a schematic block diagram of the overall structure of the present invention;
fig. 2 is a schematic block diagram of the integrated credit assessment unit of the present invention.
The meaning of each reference sign in the figure is:
100. a customer evaluation mining unit; 200. an evaluation data analysis unit; 300. a difference evaluation extraction unit; 400. an after-market analysis unit; 500. and a comprehensive credit evaluation unit.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1-2, the present embodiment provides an enterprise credit assessment system based on comprehensive data analysis, which includes a customer evaluation mining unit 100, an evaluation data analysis unit 200, a bad evaluation extraction unit 300, an after-sales analysis unit 400, and a comprehensive credit assessment unit 500;
the client evaluation mining unit 100 is used for mining evaluation data of a current enterprise, wherein the evaluation data comprises a first evaluation and a subsequent evaluation;
specifically, the client evaluation mining unit 100 uses a web crawler technology to mine the evaluation data of the current enterprise from the web page, specifically, the web crawler technology automatically accesses the web page, parses the HTML content, and extracts the required information to collect the evaluation data, and the working principle is as follows: the web crawler first accesses the target webpage by sending an HTTP request, can use a request library in Python to send the request, and contains necessary header information and parameters, once the response is received, the crawler can acquire the HTML content of the webpage, can use a response.text method to acquire the HTML content in a plain text form, and can use an HTML parsing library such as Beautifluup to parse the HTML content, so that elements, tags and contents in the HTML content can be accessed and extracted in a more convenient manner, and the target evaluation data can be extracted according to the structure of the webpage, the tags where the evaluation data are located and other information by using a method and a selector grammar provided by the parsing library.
Since the customer evaluation mining unit 100 directly mines the evaluation data of the current enterprise, it cannot accurately determine whether the evaluation data belongs to the first evaluation or the subsequent evaluation, and thus the subsequent data analysis is inaccurate, so the customer evaluation mining unit 100 further includes an evaluation data classification module for marking the time stamp of each evaluation of the customer, classifying the first evaluation and the subsequent evaluation according to the time stamp, wherein the time stamp date is the first evaluation, the time stamp date is the subsequent evaluation, and the time stamp is marked when each evaluation is performed by the customer, the first evaluation is the first evaluation behavior of the customer after purchasing the commodity, and the subsequent evaluation is the follow-up evaluation behavior after a period of time or after-sales processing, therefore, the evaluation data can be classified according to the time stamp corresponding to the date, and the credit of the enterprise can be conveniently and subsequently analyzed respectively.
The evaluation data analysis unit 200 is configured to extract keywords of the first evaluation in the evaluation data of the client evaluation mining unit 100, sequentially input a preset evaluation data set, and output a first evaluation type, where the first evaluation type includes a good evaluation, a medium evaluation and a poor evaluation;
the evaluation data analysis unit 200 includes an evaluation data set, a keyword extraction module, and a type output module;
the evaluation data set is used for establishing a set of keyword features and evaluation types, wherein the features of the realization keywords correspond to one evaluation type, and one evaluation type may correspond to a plurality of keyword features, for example: the key word is characterized in that words such as dislike, not suggested, not good are adopted, the corresponding evaluation type is poor evaluation, the evaluation type is conveniently determined according to text content evaluated by a client, and the accuracy of data analysis is improved;
the keyword extraction module is used for extracting texts in the primary evaluation separated by the evaluation data classification module, and extracting keyword features in the texts by adopting a keyword extraction algorithm, wherein the keyword extraction algorithm adopts a TextRank, is a graph-based sorting algorithm and is used for extracting keywords and abstracts from the texts, the importance of the keywords is determined by dividing the texts into sentences and words, constructing a graph according to the relation between the sentences and the words and running a Pagerank algorithm on the graph;
the type output module is used for inputting the keywords in the keyword extraction module into the evaluation data set, outputting the primary evaluation type, outputting the difference evaluation if the keyword features are dislike, outputting the difference evaluation signal, and similarly, outputting the good evaluation signal, outputting the medium evaluation signal, and outputting the medium evaluation signal.
The poor evaluation extraction unit 300 is configured to receive the evaluation data corresponding to the poor evaluation in the primary evaluation type and output the evaluation data by the evaluation data analysis unit 200, and call the marketing list and the customer evaluation content corresponding to the evaluation data to remind the enterprise worker, and perform after-sales processing on the poor evaluation for the customer evaluation;
further, the bad evaluation extraction unit 300 includes a signal receiving module, a data retrieving module and an after-sales reminding module;
the signal receiving module is connected with the output end of the type output module and is used for receiving the difference evaluation output by the type output module and sending out a difference evaluation signal;
the data retrieval module is used for associating the text content of the customer evaluation with the marketing list, and searching the data storage library for the marketing list corresponding to the poor evaluation signal in the signal receiving module;
after the reminding after-sales module is used for receiving the poor evaluation signal sent by the signal receiving module, the marketing list and the customer evaluation content corresponding to the poor evaluation are sent to enterprise workers through communication tools, such as mails, short messages and the like, so that the enterprise workers can timely receive the poor evaluation marketing list and the customer evaluation, follow-up poor evaluation aiming at the customer evaluation is facilitated, and the enterprise can take some measures to carry out after-sales treatment, for example, timely reply to customers, provide solutions, improve products or services and the like, and support and manual participation of relevant customer relation management CRM and after-sales service processes are needed.
The after-sales analysis unit 400 is configured to track the after-sales processing result of the bad evaluation extraction unit 300, extract the keywords of the subsequent evaluation again by the client evaluation mining unit 100, input the keywords into the evaluation data set, and output the subsequent evaluation types, where the subsequent evaluation types also include good evaluation, medium evaluation and bad evaluation;
in addition, the after-market analysis unit 400 includes an after-market tracking module and a subsequent evaluation module;
the after-sales tracking module is used for tracking the signal of finishing the after-sales processing of the bad evaluation, extracting keywords of a text of the subsequent evaluation from the subsequent evaluation mined by the customer evaluation mining unit 100, wherein the extracting keywords can be realized by text analysis and mining technologies, including word frequency statistics, N-gram language models, topic models and the like, and the technologies can help to identify keywords and phrases in the evaluation and provide input for the identification of the type of the subsequent evaluation;
the subsequent evaluation module is used for calling out an evaluation data set, inputting the keywords of the subsequent evaluation text extracted by the after-sale tracking module into the evaluation data set, and outputting the evaluation type.
The integrated credit assessment unit 500 is used for analyzing the enterprise credit of the first evaluation in the evaluation data analysis unit 200 and the enterprise credit of the subsequent evaluation in the after-sales analysis unit 400, respectively, using the point system, and integrating the first evaluation and the subsequent evaluation to analyze the enterprise credit.
The comprehensive credit evaluation unit 500 includes an integral preset module, an independent evaluation module, and a comprehensive evaluation module;
the score presetting module is used for setting a score rule, deducting on the basis of initial scores to record credit scores when bad scores appear, presetting a credit level threshold value, comparing the credit scores with the credit scores, outputting credit levels, for example, setting the credit scores of each enterprise to be 100 scores initially, deducting 2 scores when bad scores appear, if bad scores appear more, corresponding credit score data are reduced, and indicating enterprise credit reduction, wherein the credit levels can be a first level, a second level and a third level, respectively presetting the credit level threshold values, such as: the first-level threshold value is 90 to represent the credit preference of the enterprise, the second-level threshold value is 70 to represent the credit preference of the enterprise, the third-level threshold value is 65 to represent the credit preference of the enterprise, the credit score is compared with the threshold value, the grade of the credit score is output, and the credit of the enterprise is determined;
the independent evaluation module is used for respectively analyzing the quantity of the primary evaluation and the subsequent evaluation corresponding to the good evaluation, the medium evaluation and the bad evaluation, respectively inputting the evaluation types into the integral rule of the integral preset module, respectively outputting the credit score of the primary evaluation and the credit score of the subsequent evaluation, determining the enterprise credit corresponding to the primary evaluation and the subsequent evaluation, and conveniently evaluating the primary marketing customer feedback condition of the enterprise according to the enterprise credit corresponding to the primary evaluation and evaluating the after-sales treatment customer feedback condition of the enterprise according to the enterprise credit corresponding to the subsequent evaluation;
the comprehensive evaluation module is used for summarizing the integral rule of the integral preset module input by the initial evaluation type in the type output module and the subsequent evaluation type in the subsequent evaluation module, outputting comprehensive credit score, determining enterprise comprehensive credit, and being beneficial to comprehensively analyzing the influence of the initial evaluation and the subsequent evaluation on enterprise credit.
In summary, when the present enterprise data analysis enterprise credit is mined on the network, the method for collecting the customer evaluation is mostly adopted, for example: the present credit evaluation system directly evaluates the credit of the enterprise only aiming at the evaluation data of the customers after marketing, but if the evaluation data corresponding to some good evaluation is not necessarily good content and the evaluation data corresponding to poor evaluation is not bad content, if the evaluation data corresponding to poor evaluation is not analyzed only by virtue of good evaluation and poor evaluation, the evaluation data after the after-sales treatment of the good evaluation and poor evaluation is inconvenient to analyze, the credit of the enterprise is not accurate, the credit of the enterprise cannot be evaluated by integrating the customer evaluation of the initial stage and the later stage, and the selection of the customers after-sales is affected, therefore, the customer evaluation is mined by the customer evaluation mining unit 100, the first evaluation type is conveniently judged by the evaluation data analysis unit 200, if the first evaluation type is poor evaluation, the poor evaluation extraction unit 300 is enabled to analyze the evaluation type of the after-sales evaluation by the after-sales analysis unit 400, the credit evaluation condition of the customers corresponding to the first evaluation and the customers after-sales evaluation is convenient to determine the credit condition of the customers after-sales, and the credit condition of the customers corresponding to the customers after-sales evaluation can be conveniently and comprehensively evaluated by the comprehensive evaluation unit 500.
Example 2:
because the customers cannot evaluate the enterprise in time after the marketing event in the enterprise is completed, and the poor evaluation extraction unit 300 cannot evaluate the enterprise in time when the after-sales process is performed for the poor evaluation, the credit evaluation efficiency of the enterprise is affected, so that in order to encourage the customers to evaluate in time, the implementation is different from embodiment 1 in that: the comprehensive credit evaluation unit 500 further comprises an evaluation point module, wherein the evaluation point module is used for setting waiting time when the client finishes marketing, outputting point rewards to the client account when the client finishes evaluation within the waiting time, and outputting no rewards outside the waiting time, so that timely evaluation on the client after finishing marketing orders is improved, the point rewards can be used for changing services such as preferential price, commodity and the like through preset point rules, the enthusiasm of the client is improved, and the efficiency of evaluating enterprise credit according to the client evaluation is facilitated.
Example 3:
in the case of evaluating enterprise credit by static data of good, medium and bad evaluation, there is also a limitation in the structure of analysis, for example, some of the early marketing strategies of enterprises have problems, leading to a large early evaluation, and after improvement, the change of the credit of the enterprises after improvement cannot be judged, so that this embodiment is different from embodiment 1 and embodiment 2 in that: the comprehensive credit assessment unit 500 further includes a trend prediction module, which is connected to an output end of the comprehensive assessment module, and is configured to analyze credit score variation trend of the credit score in the comprehensive assessment module according to a time sequence, and use credit data as a training set to construct a prediction model, output a prediction result of the credit score, input the prediction result of the credit score to a score preset module, and output a prediction result of the credit grade, where, when analyzing the time sequence data of the credit score, a traditional time sequence model such as an ARIMA model may be used, or a more complex model such as a recurrent neural network RNN, a long-term memory network LSTM or a gate-controlled circulation unit GRU may be used, and these models may capture long-term dependency relationships in the time sequence data, so as to more accurately predict the variation trend of the credit score in the time sequence, and then use a supervised learning algorithm such as decision tree, random forest, logic regression, etc. to predict the trend of the credit score in a later period, so as to facilitate outputting the variation of the credit grade of the enterprise, and determine whether the enterprise is in an ascending state or a descending state, so that it is inconvenient for a user to select to make a modification policy for the enterprise to predict the enterprise.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the above-described embodiments, and that the above-described embodiments and descriptions are only preferred embodiments of the present invention, and are not intended to limit the invention, and that various changes and modifications may be made therein without departing from the spirit and scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (9)
1. The enterprise credit assessment system based on comprehensive data analysis is characterized in that: the system comprises a client evaluation mining unit (100), an evaluation data analysis unit (200), a difference evaluation extraction unit (300), an after-sales analysis unit (400) and a comprehensive credit evaluation unit (500);
the client evaluation mining unit (100) is used for mining evaluation data of a current enterprise, wherein the evaluation data comprises a primary evaluation and a subsequent evaluation;
the evaluation data analysis unit (200) is used for extracting keywords of primary evaluation in the evaluation data of the client evaluation mining unit (100), sequentially inputting a preset evaluation data set, and outputting a primary evaluation type, wherein the primary evaluation type comprises a good evaluation, a medium evaluation and a poor evaluation;
the evaluation data analysis unit (200) is used for outputting evaluation data corresponding to the evaluation in the primary evaluation type, and calling a marketing list and customer evaluation content corresponding to the evaluation data to remind enterprise workers, and performing after-sales processing on the evaluation for the customer;
the after-sales analysis unit (400) is used for tracking the after-sales processing result of the bad evaluation extraction unit (300), extracting keywords of the subsequent evaluation through the client evaluation mining unit (100) again, inputting the keywords into the evaluation data set, and outputting the subsequent evaluation type, wherein the subsequent evaluation type also comprises good evaluation, medium evaluation and bad evaluation;
the comprehensive credit evaluation unit (500) is used for respectively analyzing the enterprise credit of the first evaluation in the evaluation data analysis unit (200) and the enterprise credit of the subsequent evaluation in the after-sales analysis unit (400) by adopting an integral system, and summarizing the first evaluation and the subsequent evaluation to comprehensively analyze the enterprise credit.
2. The integrated data analysis-based enterprise credit assessment system of claim 1, wherein: the client evaluation mining unit (100) adopts a web crawler technology to mine evaluation data of a current enterprise from a web page.
3. The integrated data analysis-based enterprise credit assessment system of claim 2, wherein: the customer evaluation mining unit (100) further comprises an evaluation data classification module, wherein the evaluation data classification module is used for marking the time stamp of each evaluation of the customer, classifying the first evaluation and the subsequent evaluation according to the time stamp, wherein the time stamp date is the first evaluation, and the time stamp date is the subsequent evaluation.
4. The integrated data analysis-based enterprise credit assessment system of claim 3, wherein: the evaluation data analysis unit (200) comprises an evaluation data set, a keyword extraction module and a type output module;
the evaluation data set is used for establishing a set of keyword features corresponding to the evaluation types;
the keyword extraction module is used for extracting texts in the primary evaluation separated by the evaluation data classification module, and keyword characteristics in the texts are extracted by adopting a keyword extraction algorithm;
the type output module is used for inputting the keywords in the keyword extraction module into the evaluation data set and outputting the primary evaluation type.
5. The integrated data analysis based enterprise credit assessment system of claim 4, wherein: the difference evaluation extraction unit (300) comprises a signal receiving module, a data retrieving module and an after-sales reminding module;
the signal receiving module is connected with the output end of the type output module and is used for receiving the difference evaluation output by the type output module and sending out a difference evaluation signal;
the data retrieval module is used for associating text content of customer evaluation with the marketing list, and searching the data storage library for the marketing list corresponding to the evaluation signal in the signal receiving module;
the after-sale reminding module is used for sending marketing orders and customer evaluation contents corresponding to the difference evaluation to enterprise workers through the communication tool after receiving the difference evaluation signals sent by the signal receiving module.
6. The integrated data analysis based enterprise credit assessment system of claim 5, wherein: the after-sales analysis unit (400) comprises an after-sales tracking module and a subsequent evaluation and assessment module;
the after-sale tracking module is used for tracking the signal of finishing the after-sale processing of the bad evaluation, and extracting keywords of a subsequent evaluation text from the subsequent evaluation mined by the client evaluation mining unit (100);
the subsequent evaluation module is used for calling out an evaluation data set, inputting the keywords of the subsequent evaluation text extracted by the after-sale tracking module into the evaluation data set, and outputting the evaluation type.
7. The integrated data analysis based enterprise credit assessment system of claim 6, wherein: the comprehensive credit evaluation unit (500) comprises an integral preset module, an independent evaluation module and a comprehensive evaluation module;
the score presetting module is used for setting a score rule, deducting scores on the basis of initial scores when bad scores appear, presetting a credit grade threshold value, comparing the credit grade threshold value with the credit scores, and outputting credit grades;
the independent evaluation module is used for respectively analyzing the quantity of the primary evaluation and the subsequent evaluation corresponding to the good evaluation, the medium evaluation and the poor evaluation, respectively inputting the evaluation types into the integral rule of the integral preset module, respectively outputting the credit score of the primary evaluation and the credit score of the subsequent evaluation, and determining the enterprise credit corresponding to the primary evaluation and the subsequent evaluation;
the comprehensive evaluation module is used for summarizing the first evaluation type in the type output module and the integral rule of the integral preset module of the subsequent evaluation type input integral in the subsequent evaluation module, outputting comprehensive credit score and determining comprehensive credit of enterprises.
8. The integrated data analysis based enterprise credit assessment system of claim 7, wherein: the comprehensive credit evaluation unit (500) further comprises an evaluation point module, wherein the evaluation point module is used for setting waiting time when the client finishes marketing, outputting point rewards to the client account when the client finishes evaluation within the waiting time, and not rewarding outside the waiting time.
9. The integrated data analysis based enterprise credit assessment system of claim 7, wherein: the comprehensive credit evaluation unit (500) further comprises a trend prediction module, wherein the trend prediction module is connected with the output end of the comprehensive evaluation module and is used for analyzing credit score change trend of the credit score in the comprehensive evaluation module according to a time sequence, constructing a prediction model by taking credit data as a training set, outputting a prediction result of the credit score, inputting the prediction result of the credit score into the integral preset module and outputting a prediction result of the credit grade.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311135624.XA CN117151797A (en) | 2023-09-05 | 2023-09-05 | Enterprise credit assessment system based on comprehensive data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311135624.XA CN117151797A (en) | 2023-09-05 | 2023-09-05 | Enterprise credit assessment system based on comprehensive data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117151797A true CN117151797A (en) | 2023-12-01 |
Family
ID=88905767
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311135624.XA Pending CN117151797A (en) | 2023-09-05 | 2023-09-05 | Enterprise credit assessment system based on comprehensive data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117151797A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005134938A (en) * | 2002-10-15 | 2005-05-26 | Kagoshima Bank Ltd | Enterprise credit rating system and enterprise credit rating program |
| CN113011966A (en) * | 2021-03-18 | 2021-06-22 | 中国光大银行股份有限公司 | Credit scoring method and device based on deep learning |
| CN113793169A (en) * | 2021-08-12 | 2021-12-14 | 惠州Tcl云创科技有限公司 | User comment data processing method, device, equipment and storage medium |
| CN115187065A (en) * | 2022-07-11 | 2022-10-14 | 湖北帅锐服装贸易有限公司 | Garment commodity quality analysis system based on electronic commerce platform |
-
2023
- 2023-09-05 CN CN202311135624.XA patent/CN117151797A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005134938A (en) * | 2002-10-15 | 2005-05-26 | Kagoshima Bank Ltd | Enterprise credit rating system and enterprise credit rating program |
| CN113011966A (en) * | 2021-03-18 | 2021-06-22 | 中国光大银行股份有限公司 | Credit scoring method and device based on deep learning |
| CN113793169A (en) * | 2021-08-12 | 2021-12-14 | 惠州Tcl云创科技有限公司 | User comment data processing method, device, equipment and storage medium |
| CN115187065A (en) * | 2022-07-11 | 2022-10-14 | 湖北帅锐服装贸易有限公司 | Garment commodity quality analysis system based on electronic commerce platform |
Non-Patent Citations (1)
| Title |
|---|
| 令狐公子: "2021年淘宝网评价规则全新改版", pages 1 - 3, Retrieved from the Internet <URL:https://www.zhihu.com/question/455980375/answer/1849257672?utm_id=0> * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101741509B1 (en) | Device and method for analyzing corporate reputation by data mining of news, recording medium for performing the method | |
| Chung | BizPro: Extracting and categorizing business intelligence factors from textual news articles | |
| US8781989B2 (en) | Method and system to predict a data value | |
| Sehgal et al. | Sops: stock prediction using web sentiment | |
| US20190213488A1 (en) | Systems and methods for semantic analysis based on knowledge graph | |
| CN111125343A (en) | Text analysis method and device suitable for human-sentry matching recommendation system | |
| CN103154991A (en) | Credit risk mining | |
| CN104702492A (en) | Garbage message model training method, garbage message identifying method and device thereof | |
| CN103399891A (en) | Method, device and system for automatic recommendation of network content | |
| CN110888990A (en) | Text recommendation method, device, equipment and medium | |
| CN102402717A (en) | Data analysis facility and method | |
| CN105138577A (en) | Big data based event evolution analysis method | |
| CN111782793A (en) | Intelligent customer service processing method, system and equipment | |
| CN116109373A (en) | Recommendation method and device for financial products, electronic equipment and medium | |
| CN114202243A (en) | Engineering project management risk early warning method and system based on random forest | |
| US11755677B2 (en) | Data mining method, data mining apparatus, electronic device and storage medium | |
| CN115577172A (en) | Article recommendation method, device, equipment and medium | |
| Darena et al. | Machine learning-based analysis of the association between online texts and stock price movements | |
| CN117251570A (en) | Bidding purchasing demand analysis method and system based on text analysis | |
| CN117151797A (en) | Enterprise credit assessment system based on comprehensive data analysis | |
| CN115994531A (en) | Multi-dimensional text comprehensive identification method | |
| Kelly | News, sentiment and financial markets: A computational system to evaluate the influence of text sentiment on financial assets | |
| Deshpande et al. | BI and sentiment analysis | |
| CN113127465A (en) | Data fusion method and system | |
| CN113408263A (en) | Criminal period prediction method and device, storage medium and electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |