A consumer complaint credit supervision management system and method based on artificial intelligence
By screening and optimizing consumer complaint work orders, a sequence of processing needs and public information are generated, which solves the problems of fragmented complaint information and duplicate complaints, and improves processing efficiency and information retrieval.
Patent Information
- Application Number
- CN202411205638.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-08-30
AI Technical Summary
The existing AI-based consumer complaint credit supervision and management system suffers from problems such as duplicate complaints, false complaints, fragmented complaint information, and low processing efficiency, resulting in wasted resources and difficulty in retrieving publicized information.
By acquiring a set of pending work orders for consumer complaints, filtering out abnormal work orders and generating a sequence of optimized processing needs, and combining correlation relationships and correlation interference weights, the complaint information is integrated and optimized, generating a public disclosure sequence and credit supervision early warning information.
Effectively screen duplicate complaints, improve processing efficiency, reduce the difficulty of publicizing information, and achieve intelligent management of credit supervision of consumer complaints.
Smart Images

Figure CN118710141B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of consumer complaint credit supervision, and specifically to a consumer complaint credit supervision management system and method based on artificial intelligence. Background Art
[0002] In order to effectively handle these disputes and protect the legitimate rights and interests of consumers, the consumer complaint credit supervision and management system came into being. The system uses information technology to centrally manage and uniformly supervise consumer complaints to improve processing efficiency and data quality.
[0003] However, the current AI-based consumer complaint credit supervision and management system faces two difficulties during use. On the one hand, there are cases of repeated complaints and false complaints by complaining users, which increases the difficulty of the system's management of consumer complaints and wastes public resources to a certain extent. On the other hand, complaint information often shows that the same enterprise is complained by multiple users in a short period of time. The scattered and fragmented processing time of complaint information not only causes great distress to the corresponding enterprises, but also reduces the processing efficiency of complaint information. At the same time, the fragmentation of public information will also increase the difficulty for the public and enterprises to extract public information. Therefore, the existing technology has major defects. Summary of the invention
[0004] The purpose of the present invention is to provide a consumer complaint credit supervision management system and method based on artificial intelligence to solve the problems raised in the above background technology.
[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a consumer complaint credit supervision management system and method based on artificial intelligence, the method comprising the following steps:
[0006] S1. Obtain the pending work order information set corresponding to the consumer complaint, and filter the pending work order information set based on the work order information to obtain the abnormal work order information set and the optimized pending work order information set; extract information from the elements in the pending work order information set to obtain the feature extraction set of each pending work order information;
[0007] S2. Analyze the correlation between the feature extraction sets corresponding to each element in the to-be-done work order information optimization set, calculate the work order import priority feature index corresponding to each element in the to-be-done work order information optimization set, and generate an optimization processing requirement sequence for the to-be-done work orders;
[0008] S3. Real-time acquisition of interactive information of each element in the optimization processing requirement sequence of pending work orders during the processing process, combined with the historical work order information records processed in the database, to obtain the associated interference weights corresponding to each element in the optimization processing requirement sequence of pending work orders;
[0009] S4. Obtain the optimization processing requirement sequence of pending work orders, and generate a public announcement sequence of consumer complaint processing work orders based on the associated interference weights corresponding to each element in the optimization processing requirement sequence of pending work orders; generate credit supervision warning information corresponding to each public announcement object in the public announcement sequence based on the trend change chart of the industry proportion of consumer complaints, and feed it back to the corresponding public announcement object.
[0010] The artificial intelligence technology in the present invention is mainly reflected in the optimization of processing demand sequences by dynamically generating to-do work orders, and the dynamic acquisition of credit supervision warning information corresponding to each public announcement object in the public announcement sequence, thereby ensuring the intelligent management of consumer complaint credit supervision.
[0011] Further, when the abnormal work order information set and the optimized set of pending work order information are obtained in S1, the corresponding data in each work order information is obtained. When one or more data are missing in the work order information, the corresponding work order information is determined to be abnormal work order information. Otherwise, the work order information is determined to be normal work order information. The set consisting of the elements in the pending work order information set belonging to the abnormal work order information is recorded as the abnormal work order information set, and the set consisting of the elements in the pending work order information set belonging to the normal work order information is recorded as the optimized set of pending work order information.
[0012] When the feature extraction set of each pending work order information is obtained in S1, the feature extraction set includes the complainant, the complaint object, the complaint content and the complaint time.
[0013] The present invention determines whether the work order information is abnormal in order to pre-process the pending work order information set, avoid the subsequent work order processing process, and effectively resolve consumer complaint disputes and reduce the processing efficiency of consumer complaint handling personnel.
[0014] Further, when analyzing the association relationship between the feature extraction sets corresponding to the elements in the to-do work order information optimization set, the association relationship between the feature extraction sets corresponding to different elements includes a first association relationship, a second association relationship and a third association relationship.
[0015] The first association relationship indicates that the complainants in the feature extraction set are the same and the rest of the extracted information is different; the second association relationship indicates that the complaint objects in the feature extraction set are the same and the rest of the extracted information is different; the third association relationship indicates that the corresponding complainants, complaint objects, and complaint contents in the feature extraction set are the same, but the complaint times are different;
[0016] The method for generating an optimized processing requirement sequence of pending work orders comprises the following steps:
[0017] S21, obtaining an optimized set of to-be-done work order information, and recording the jth element in the optimized set of to-be-done work order information as Bj;
[0018] S22, obtaining all the to-do work order information that has a correlation relationship between the feature extraction set corresponding to the to-do work order information optimization set and the feature extraction set corresponding to Bj;
[0019] S23. Obtain the work order import priority characteristic index in the to-do work order information optimization set, denoted as DYZ j ,
[0020] ,
[0021] ,
[0022] Wherein, r represents the conversion factor, and r is a constant preset in the database; PGj represents the correlation deviation value between the feature extraction set corresponding to the to-do work order information optimization set and the feature extraction set corresponding to Bj;
[0023] SUM1 j Indicates the number of to-do work order information that has a first association relationship between the corresponding feature extraction set in the to-do work order information optimization set and the feature extraction set corresponding to Bj;
[0024] SUM2 j Indicates the number of to-do work order information that has a second association relationship between the corresponding feature extraction set in the to-do work order information optimization set and the feature extraction set corresponding to Bj;
[0025] LS j Indicates the degree of membership between the complaint content and the business corresponding to the complaint object in the feature extraction set corresponding to Bj;
[0026] Get LS j When the value of is , extract the time and product involved in the complaint content, query the union of the business sets corresponding to the complaint objects at different time points during the period from the time involved to the current time, compare the obtained union with the product involved, and when the product involved belongs to the obtained union, determine LS j =1; otherwise, LS j =0;
[0027] SP j Indicates the time deviation coefficient of the complaint time in the feature extraction set corresponding to Bj based on the current time;
[0028] The SP j =TGj / TGmax,
[0029] TGmax represents the maximum value of the interval between the complaint time and the current time in the feature extraction set corresponding to each element in the to-do work order information optimization set;
[0030] TGj represents the maximum value of the interval between the complaint time and the current time in the feature extraction set corresponding to each element in the to-do work order information optimization set that has a third association relationship with Bj and Bj;
[0031] S24. Sort the elements in the to-be-done work order information optimization set in descending order according to the corresponding work order import priority characteristic index to obtain the optimization processing requirement sequence of the to-be-done work orders; bind the elements that have a third-party relationship with each other during the sorting process, and mark the remaining elements in the binding result except the element with the maximum interval between the complaint time and the current time; randomly assign the order of different elements with the same work order import priority characteristic index during the sorting process.
[0032] The acquisition of the optimization processing requirement sequence of the pending work orders in the present invention completes the integration optimization of the pending work order information, can effectively realize the screening of repeated complaint work orders, and improve the processing efficiency of consumer complaint handling personnel.
[0033] Furthermore, the interaction information includes the total interaction time and the work order processing result, and the work order processing result includes the complaint processing success, complaint failure and the two parties to be negotiated;
[0034] When obtaining the associated interference weights corresponding to the elements in the optimization processing requirement sequence of the to-be-done work orders, the associated interference weight corresponding to the i-th element in the optimization processing requirement sequence of the to-be-done work orders is recorded as GQi.
[0035] If the work order processing result corresponding to the i-th element in the optimization processing requirement sequence of pending work orders is complaint failure, then GQi=0; otherwise, GQi=[Ni·(1-Wi) / HYi+TCi / TPCi]·(1-W1);
[0036] Among them, TCi represents the sum of the total work order processing time corresponding to the i-th element and each element with a third association relationship in the optimization processing requirement sequence of the to-be-done work orders;
[0037] Ni represents the total number of complaints received by the complaint object in the i-th element of the optimization processing requirements sequence of pending work orders within a preset time period;
[0038] Wi represents the complaint object in the i-th element of the optimization processing requirement sequence of pending work orders. Among the work orders that were complained about within the preset time period, the number of times the corresponding work order processing result is complaint failure accounts for the total number of times;
[0039] HYi represents the average number of times the corresponding work order processing results of each complaint object in the preset time period are not complaint failures in the industry of the complaint object in the i-th element in the optimization processing requirements sequence of pending work orders; TPCi represents the average total interaction time corresponding to each interaction information in the industry to which the complaint object in the i-th element in the optimization processing requirements sequence of pending work orders belongs, and both HYi and TPCi are greater than 0.
[0040] Furthermore, the method for generating a public sequence of consumer complaint handling work orders includes the following steps:
[0041] S41, obtaining the association relationship between the optimization processing requirement sequence of the to-be-done work orders and the feature extraction sets corresponding to the elements thereof, and obtaining the association interference weight corresponding to each element in the optimization processing requirement sequence of the to-be-done work orders;
[0042] S42, extracting all elements whose corresponding associated interference weights are not 0 in the optimization processing requirement sequence of pending work orders, and arranging the extracted elements in descending order according to the corresponding associated interference weights, to obtain a candidate public announcement sequence for consumer complaint processing work orders;
[0043] S43, obtaining a set consisting of marked elements in the optimization processing requirement sequence of pending work orders, recorded as a marked analysis set; obtaining the intersection of the set consisting of elements in the alternative public announcement sequence of consumer complaint processing work orders and the marked analysis set, to obtain a public announcement interference information set;
[0044] S44. Eliminate the elements in the public interference information set from the alternative public sequence of the consumer complaint processing work order, and record the sequence corresponding to the remaining results as the public sequence of the consumer complaint processing work order.
[0045] Furthermore, when generating credit supervision warning information corresponding to each disclosure object in the disclosure sequence, the credit supervision warning information includes the disclosure sequence of consumer complaint handling work orders, a trend change chart of the proportion of consumer complaints by industry, and a fluctuation chart of the proportion of the number of complaint handling work orders disclosed by the corresponding disclosure object each time within a preset time period based on the proportion of the total number of processed work orders in the corresponding industry;
[0046] The trend change graph of the proportion of consumer complaints by industry is a line graph obtained by connecting the proportion of consumer complaints by the corresponding industry in each public announcement sequence within the corresponding time period with the data pairs constituted by the corresponding public announcement sequence. The proportion of consumer complaints by the corresponding industry in the public announcement sequence is the ratio between the number of work orders in the corresponding industry in the corresponding public announcement sequence and the industry to which the public announcement object corresponding to the corresponding credit supervision early warning information belongs and the total number of work orders in the corresponding public announcement sequence;
[0047] The fluctuation chart of the number of complaint handling work orders disclosed each time by the corresponding publicity object within the preset time period based on the proportion of the total number of processed work orders in the corresponding industry is a line chart obtained by connecting the first publicity quantity proportion of the publicity object corresponding to the corresponding credit supervision early warning information in each publicity sequence within the corresponding time period with the data pairs constituted by the corresponding publicity sequence. The first publicity quantity proportion of the publicity object corresponding to the corresponding credit supervision early warning information in the publicity sequence is the ratio between the number of work orders of the publicity object corresponding to the corresponding credit supervision early warning information in the corresponding publicity sequence and the number of work orders of the industry to which the publicity object corresponding to the corresponding credit supervision early warning information belongs.
[0048] A consumer complaint credit supervision management system and method based on artificial intelligence, the system includes the following modules:
[0049] A work order feature extraction module, wherein the work order feature extraction module obtains a pending work order information set corresponding to a consumer complaint, and filters the pending work order information set in combination with the work order information to obtain an abnormal work order information set and an optimized pending work order information set; extracts information from elements in the pending work order information set to obtain a feature extraction set of each pending work order information;
[0050] An import feature analysis module is used to analyze the correlation between the feature extraction sets corresponding to the elements in the to-be-done work order information optimization set, calculate the work order import priority feature index corresponding to the elements in the to-be-done work order information optimization set, and generate an optimization processing requirement sequence for the to-be-done work orders;
[0051] An interactive dynamic interference analysis module, which obtains the interactive information of each element in the optimization processing requirement sequence of the to-be-done work orders in real time during the processing process, and obtains the associated interference weight corresponding to each element in the optimization processing requirement sequence of the to-be-done work orders in combination with the historical work order information records processed in the database;
[0052] A regulatory feedback management module, wherein the regulatory feedback management module obtains an optimization processing requirement sequence of pending work orders, and generates a public announcement sequence of consumer complaint processing work orders in combination with the associated interference weights corresponding to each element in the optimization processing requirement sequence of pending work orders; and generates credit regulatory early warning information corresponding to each public announcement object in the public announcement sequence in combination with a trend change chart of the industry proportion of consumer complaints, and feeds back the credit regulatory early warning information to the corresponding public announcement object.
[0053] Furthermore, the import feature analysis module includes a correlation analysis unit, a feature index calculation unit and an optimization processing requirement sequence construction unit.
[0054] The correlation analysis unit analyzes the correlation between the feature extraction sets corresponding to each element in the to-be-done work order information optimization set;
[0055] The characteristic index calculation unit calculates the work order import priority characteristic index corresponding to each element in the to-be-done work order information optimization set in combination with the analysis result of the correlation analysis unit;
[0056] The optimization processing requirement sequence construction unit generates an optimization processing requirement sequence of the to-be-done work orders according to the analysis results of the association relationship analysis unit and the characteristic index calculation unit.
[0057] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention can monitor the repetitive behaviors of complaining users by analyzing the correlation relationships of pending work orders, and reduce the effective processing volume of complaint work orders by tag binding; at the same time, the analysis results of the correlation relationships of pending work orders, the work order import priority feature index and the associated interference weights corresponding to the work orders can, to a certain extent, realize the integration of scattered complaint information, which not only improves the processing efficiency and data quality of complaint work orders, but also reduces the difficulty of the public and enterprises to extract public information, and effectively realizes the supervision and management of consumer complaint credit. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0059] Figure 1 It is a structural schematic diagram of a consumer complaint credit supervision management system based on artificial intelligence of the present invention;
[0060] Figure 2 It is a flow chart of a consumer complaint credit supervision management method based on artificial intelligence in the present invention. DETAILED DESCRIPTION
[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0062] See also Figure 1 The present invention provides a technical solution: a consumer complaint credit supervision management system and method based on artificial intelligence, the system includes the following modules:
[0063] A work order feature extraction module, wherein the work order feature extraction module obtains a pending work order information set corresponding to a consumer complaint, and filters the pending work order information set in combination with the work order information to obtain an abnormal work order information set and an optimized pending work order information set; extracts information from elements in the pending work order information set to obtain a feature extraction set of each pending work order information;
[0064] An import feature analysis module is used to analyze the correlation between the feature extraction sets corresponding to the elements in the to-be-done work order information optimization set, calculate the work order import priority feature index corresponding to the elements in the to-be-done work order information optimization set, and generate an optimization processing requirement sequence for the to-be-done work orders;
[0065] An interactive dynamic interference analysis module, which obtains the interactive information of each element in the optimization processing requirement sequence of the to-be-done work orders in real time during the processing process, and obtains the associated interference weight corresponding to each element in the optimization processing requirement sequence of the to-be-done work orders in combination with the historical work order information records processed in the database;
[0066] A regulatory feedback management module, wherein the regulatory feedback management module obtains an optimization processing requirement sequence of pending work orders, and generates a public announcement sequence of consumer complaint processing work orders in combination with the associated interference weights corresponding to each element in the optimization processing requirement sequence of pending work orders; and generates credit regulatory early warning information corresponding to each public announcement object in the public announcement sequence in combination with a trend change chart of the industry proportion of consumer complaints, and feeds back the credit regulatory early warning information to the corresponding public announcement object.
[0067] The import feature analysis module includes a correlation analysis unit, a feature index calculation unit and an optimization processing requirement sequence construction unit.
[0068] The correlation analysis unit analyzes the correlation between the feature extraction sets corresponding to each element in the to-be-done work order information optimization set;
[0069] The characteristic index calculation unit calculates the work order import priority characteristic index corresponding to each element in the to-be-done work order information optimization set in combination with the analysis result of the correlation analysis unit;
[0070] The optimization processing requirement sequence construction unit generates an optimization processing requirement sequence of the to-be-done work orders according to the analysis results of the association relationship analysis unit and the characteristic index calculation unit.
[0071] like Figure 2 As shown, a consumer complaint credit supervision management system and method based on artificial intelligence, the method comprises the following steps:
[0072] S1. Obtain the pending work order information set corresponding to the consumer complaint, and filter the pending work order information set based on the work order information to obtain the abnormal work order information set and the optimized pending work order information set; extract information from the elements in the pending work order information set to obtain the feature extraction set of each pending work order information;
[0073] When the abnormal work order information set and the optimized set of pending work order information are obtained in S1, the corresponding data in each work order information is obtained. When one or more data are missing in the work order information, the corresponding work order information is determined to be abnormal work order information. Otherwise, the work order information is determined to be normal work order information. The set consisting of the elements of the pending work order information set belonging to the abnormal work order information is recorded as the abnormal work order information set, and the set consisting of the elements of the pending work order information set belonging to the normal work order information is recorded as the optimized set of pending work order information;
[0074] The work order information in this embodiment includes the work order number, data source, complainant, complainee, complaint content, complaint time, credit code of the complainee, company name of the complainee, region of the complainee, etc.;
[0075] When the feature extraction set of each pending work order information is obtained in S1, the feature extraction set includes the complainant, the complaint object, the complaint content and the complaint time.
[0076] S2. Analyze the correlation between the feature extraction sets corresponding to each element in the to-be-done work order information optimization set, calculate the work order import priority feature index corresponding to each element in the to-be-done work order information optimization set, and generate an optimization processing requirement sequence for the to-be-done work orders;
[0077] When analyzing the association relationship between the feature extraction sets corresponding to the elements in the to-do work order information optimization set, the association relationship between the feature extraction sets corresponding to different elements includes a first association relationship, a second association relationship and a third association relationship.
[0078] The first association relationship indicates that the complainants in the feature extraction set are the same and the rest of the extracted information is different; the second association relationship indicates that the complaint objects in the feature extraction set are the same and the rest of the extracted information is different; the third association relationship indicates that the corresponding complainants, complaint objects, and complaint contents in the feature extraction set are the same, but the complaint times are different;
[0079] The method for generating an optimized processing requirement sequence of pending work orders comprises the following steps:
[0080] S21, obtaining an optimized set of to-be-done work order information, and recording the jth element in the optimized set of to-be-done work order information as Bj;
[0081] S22, obtaining all the to-do work order information that has a correlation relationship between the feature extraction set corresponding to the to-do work order information optimization set and the feature extraction set corresponding to Bj;
[0082] S23. Obtain the work order import priority characteristic index in the to-do work order information optimization set, denoted as DYZ j ,
[0083] ,
[0084] ,
[0085] Wherein, r represents the conversion factor, and r is a constant preset in the database; PGj represents the correlation deviation value between the feature extraction set corresponding to the to-do work order information optimization set and the feature extraction set corresponding to Bj;
[0086] SUM1 j Indicates the number of to-do work order information that has a first association relationship between the corresponding feature extraction set in the to-do work order information optimization set and the feature extraction set corresponding to Bj;
[0087] SUM2 j Indicates the number of to-do work order information that has a second association relationship between the corresponding feature extraction set in the to-do work order information optimization set and the feature extraction set corresponding to Bj;
[0088] LS j Indicates the degree of membership between the complaint content and the business corresponding to the complaint object in the feature extraction set corresponding to Bj;
[0089] Get LS j When the value of is , extract the time and product involved in the complaint content, query the union of the business sets corresponding to the complaint objects at different time points during the period from the time involved to the current time, compare the obtained union with the product involved, and when the product involved belongs to the obtained union, determine LS j =1; otherwise, LS j =0;
[0090] SP j Indicates the time deviation coefficient of the complaint time in the feature extraction set corresponding to Bj based on the current time;
[0091] The SP j =TGj / TGmax,
[0092] TGmax represents the maximum value of the interval between the complaint time and the current time in the feature extraction set corresponding to each element in the to-do work order information optimization set;
[0093] TGj represents the maximum value of the interval between the complaint time and the current time in the feature extraction set corresponding to each element in the to-do work order information optimization set that has a third association relationship with Bj and Bj;
[0094] S24. Sort the elements in the to-be-done work order information optimization set in descending order according to the corresponding work order import priority characteristic index to obtain the optimization processing requirement sequence of the to-be-done work orders; bind the elements that have a third-party relationship with each other during the sorting process, and mark the remaining elements in the binding result except the element with the maximum interval between the complaint time and the current time; randomly assign the order of different elements with the same work order import priority characteristic index during the sorting process.
[0095] In this embodiment, when processing the optimization processing requirement sequence of pending work orders, manual screening and importing can also be performed. When manually screening and importing, the platform can query the import batch data through fields such as import batch, import time, and matching rules. Click "Batch Data" in the corresponding module to view the details of the imported work orders in each batch.
[0096] S3. Real-time acquisition of interactive information of each element in the optimization processing requirement sequence of pending work orders during the processing process, combined with the historical work order information records processed in the database, to obtain the associated interference weights corresponding to each element in the optimization processing requirement sequence of pending work orders;
[0097] The interaction information includes the total interaction time and the work order processing result, and the work order processing result includes successful complaint processing, complaint failure, and pending negotiation between the two parties;
[0098] When obtaining the associated interference weights corresponding to the elements in the optimization processing requirement sequence of the to-be-done work orders, the associated interference weight corresponding to the i-th element in the optimization processing requirement sequence of the to-be-done work orders is recorded as GQi.
[0099] If the work order processing result corresponding to the i-th element in the optimization processing requirement sequence of pending work orders is complaint failure, then GQi=0; otherwise, GQi=[Ni·(1-Wi) / HYi+TCi / TPCi]·(1-W1);
[0100] Among them, TCi represents the sum of the total work order processing time corresponding to the i-th element and each element with a third association relationship in the optimization processing requirement sequence of the to-be-done work orders;
[0101] Ni represents the total number of complaints received by the complaint object in the i-th element of the optimization processing requirements sequence of pending work orders within a preset time period;
[0102] Wi represents the complaint object in the i-th element of the optimization processing requirement sequence of pending work orders. Among the work orders that were complained about within the preset time period, the number of times the corresponding work order processing result is complaint failure accounts for the total number of times;
[0103] HYi represents the average number of times the corresponding work order processing results of each complaint object in the preset time period are not complaint failures in the industry of the complaint object in the i-th element in the optimization processing requirements sequence of pending work orders; TPCi represents the average total interaction time corresponding to each interaction information in the industry to which the complaint object in the i-th element in the optimization processing requirements sequence of pending work orders belongs.
[0104] S4. Obtain the optimization processing requirement sequence of pending work orders, and generate a public announcement sequence of consumer complaint processing work orders based on the associated interference weights corresponding to each element in the optimization processing requirement sequence of pending work orders; generate credit supervision warning information corresponding to each public announcement object in the public announcement sequence based on the trend change chart of the industry proportion of consumer complaints, and feed it back to the corresponding public announcement object.
[0105] The method for generating a public sequence of consumer complaint handling work orders comprises the following steps:
[0106] S41, obtaining the association relationship between the optimization processing requirement sequence of the to-be-done work orders and the feature extraction sets corresponding to the elements thereof, and obtaining the association interference weight corresponding to each element in the optimization processing requirement sequence of the to-be-done work orders;
[0107] S42, extracting all elements whose corresponding associated interference weights are not 0 in the optimization processing requirement sequence of pending work orders, and arranging the extracted elements in descending order according to the corresponding associated interference weights, to obtain a candidate public announcement sequence for consumer complaint processing work orders;
[0108] S43, obtaining a set consisting of marked elements in the optimization processing requirement sequence of pending work orders, recorded as a marked analysis set; obtaining the intersection of the set consisting of elements in the alternative public announcement sequence of consumer complaint processing work orders and the marked analysis set, to obtain a public announcement interference information set;
[0109] S44. Eliminate the elements in the public interference information set from the alternative public sequence of the consumer complaint processing work order, and record the sequence corresponding to the remaining results as the public sequence of the consumer complaint processing work order.
[0110] When generating the credit supervision warning information corresponding to each disclosure object in the disclosure sequence, the credit supervision warning information includes the disclosure sequence of consumer complaint handling work orders, a trend change chart of the proportion of consumer complaints by industry, and a fluctuation chart of the proportion of the number of complaint handling work orders disclosed by the corresponding disclosure object each time within a preset time period based on the proportion of the total number of processed work orders in the corresponding industry;
[0111] The trend change graph of the proportion of consumer complaints by industry is a line graph obtained by connecting the proportion of consumer complaints by the corresponding industry in each public announcement sequence within the corresponding time period with the data pairs constituted by the corresponding public announcement sequence. The proportion of consumer complaints by the corresponding industry in the public announcement sequence is the ratio between the number of work orders in the corresponding industry in the corresponding public announcement sequence and the industry to which the public announcement object corresponding to the corresponding credit supervision early warning information belongs and the total number of work orders in the corresponding public announcement sequence;
[0112] In this embodiment, the horizontal axis of the trend change graph obtained is the public announcement sequence number corresponding to the public announcement sequence in the corresponding time period in descending order of acquisition time; the vertical axis of the trend change graph obtained is the proportion of consumer complaints of the corresponding industry in the public announcement sequence;
[0113] The fluctuation chart of the number of complaint handling work orders disclosed each time by the corresponding publicity object within the preset time period based on the proportion of the total number of processed work orders in the corresponding industry is a line chart obtained by connecting the first publicity quantity proportion of the publicity object corresponding to the corresponding credit supervision early warning information in each publicity sequence within the corresponding time period with the data pairs constituted by the corresponding publicity sequence. The first publicity quantity proportion of the publicity object corresponding to the corresponding credit supervision early warning information in the publicity sequence is the ratio between the number of work orders of the publicity object corresponding to the corresponding credit supervision early warning information in the corresponding publicity sequence and the number of work orders of the industry to which the publicity object corresponding to the corresponding credit supervision early warning information belongs.
[0114] The horizontal axis of the obtained fluctuation graph in this embodiment is the publicity sequence number corresponding to the publicity sequence in the corresponding time period in descending order according to the acquisition time; the vertical axis of the obtained fluctuation graph is the proportion of the first publicity quantity of the publicity object corresponding to the corresponding credit supervision early warning information in the publicity sequence.
[0115] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0116] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A consumer complaint credit supervision and management method based on artificial intelligence, characterized in that: The method comprises the following steps: S1. Obtain the pending work order information set corresponding to the consumer complaint, and filter the pending work order information set based on the work order information to obtain the abnormal work order information set and the optimized pending work order information set; extract information from the elements in the pending work order information set to obtain the feature extraction set of each pending work order information; S2. Analyze the correlation between the feature extraction sets corresponding to each element in the to-be-done work order information optimization set, calculate the work order import priority feature index corresponding to each element in the to-be-done work order information optimization set, and generate an optimization processing requirement sequence for the to-be-done work orders; S3. Real-time acquisition of interactive information of each element in the optimization processing requirement sequence of pending work orders during the processing process, combined with the historical work order information records processed in the database, to obtain the associated interference weights corresponding to each element in the optimization processing requirement sequence of pending work orders; S4. Obtain the optimization processing requirement sequence of pending work orders, combine the associated interference weights corresponding to each element in the optimization processing requirement sequence of pending work orders, and generate a public announcement sequence of consumer complaint processing work orders; combine the trend change chart of the proportion of consumer complaints in the industry, generate credit supervision warning information corresponding to each public announcement object in the public announcement sequence, and feed it back to the corresponding public announcement object; When the abnormal work order information set and the optimized set of pending work order information are obtained in S1, the corresponding data in each work order information is obtained. When one or more data are missing in the work order information, the corresponding work order information is determined to be abnormal work order information. Otherwise, the work order information is determined to be normal work order information. The set consisting of the elements of the pending work order information set belonging to the abnormal work order information is recorded as the abnormal work order information set, and the set consisting of the elements of the pending work order information set belonging to the normal work order information is recorded as the optimized set of pending work order information; When the feature extraction set of each pending work order information is obtained in S1, the feature extraction set includes the complainant, the object of complaint, the complaint content and the complaint time; When analyzing the association relationship between the feature extraction sets corresponding to the elements in the to-do work order information optimization set, the association relationship between the feature extraction sets corresponding to different elements includes a first association relationship, a second association relationship and a third association relationship. The first association relationship indicates that the complainants in the feature extraction set are the same and the rest of the extracted information is different; the second association relationship indicates that the complaint objects in the feature extraction set are the same and the rest of the extracted information is different; the third association relationship indicates that the corresponding complainants, complaint objects, and complaint contents in the feature extraction set are the same, but the complaint times are different; The method for generating an optimized processing requirement sequence of pending work orders comprises the following steps: S21, obtaining an optimized set of to-be-done work order information, and recording the jth element in the optimized set of to-be-done work order information as Bj; S22, obtaining all the to-do work order information that has a correlation relationship between the corresponding feature extraction set in the to-do work order information optimization set and the feature extraction set corresponding to Bj; S23. Obtain the work order import priority characteristic index in the to-do work order information optimization set, recorded as DYZ j , DYZ j =e PGj +r·LS j ·SP j , PGj=SUM2 j -SUM1 j , Wherein, r represents the conversion factor, and r is a constant preset in the database; PGj represents the correlation deviation value between the feature extraction set corresponding to the to-do work order information optimization set and the feature extraction set corresponding to Bj; SUM1 j Indicates the number of to-do work order information that has a first association relationship between the corresponding feature extraction set in the to-do work order information optimization set and the feature extraction set corresponding to Bj; SUM2 j Indicates the number of to-do work order information that has a second association relationship between the corresponding feature extraction set in the to-do work order information optimization set and the feature extraction set corresponding to Bj; LS j Indicates the degree of membership between the complaint content and the business corresponding to the complaint object in the feature extraction set corresponding to Bj; Get LS j When the value of is , extract the time and product involved in the complaint content, query the union of the business sets corresponding to the complaint objects at different time points during the period from the time involved to the current time, compare the obtained union with the product involved, and when the product involved belongs to the obtained union, determine LS j =1; otherwise, LS j =0; SP j Indicates the time deviation coefficient of the complaint time in the feature extraction set corresponding to Bj based on the current time; The SP j =TGj / TGmax, TGmax represents the maximum value of the interval between the complaint time and the current time in the feature extraction set corresponding to each element in the to-do work order information optimization set; TGj represents the maximum value of the interval between the complaint time and the current time in the feature extraction set corresponding to each element in the to-do work order information optimization set that has a third association relationship with Bj and Bj; S24, sorting the elements in the to-be-done work order information optimization set in descending order of the corresponding work order import priority characteristic index to obtain an optimization processing requirement sequence for the to-be-done work orders; in the sorting process, binding the elements that have a third association relationship with each other, and marking the remaining elements in the binding result except the element with the maximum interval between the complaint time and the current time; in the sorting process, randomly assigning the order of different elements with the same work order import priority characteristic index; The interaction information includes the total interaction time and the work order processing result, and the work order processing result includes successful complaint processing, complaint failure, and pending negotiation between the two parties; When obtaining the associated interference weights corresponding to the elements in the optimization processing requirement sequence of the to-be-done work orders, the associated interference weight corresponding to the i-th element in the optimization processing requirement sequence of the to-be-done work orders is recorded as GQi. If the work order processing result corresponding to the i-th element in the optimization processing requirement sequence of pending work orders is complaint failure, then GQi=0; otherwise, GQi=[Ni·(1-Wi) / HYi+TCi / TPCi]·(1-W1); Among them, TCi represents the sum of the total work order processing time corresponding to the i-th element and each element with a third association relationship in the optimization processing requirement sequence of the to-be-done work orders; Ni represents the total number of complaints received by the complaint object in the i-th element of the optimization processing requirements sequence of pending work orders within a preset time period; Wi represents the complaint object in the i-th element of the optimization processing requirement sequence of pending work orders. Among the work orders that were complained about within the preset time period, the number of times the corresponding work order processing result is complaint failure accounts for the total number of times; HYi represents the number of times that the corresponding work order processing result of each complaint object in the industry of the i-th element in the optimization processing demand sequence of pending work orders is not a complaint failure in the preset time period; TPCi represents the average total interaction time corresponding to each interaction information in the industry to which the complaint object in the i-th element in the optimization processing requirements sequence of pending work orders belongs; The method for generating a public sequence of consumer complaint handling work orders comprises the following steps: S41, obtaining the association relationship between the optimization processing requirement sequence of the to-be-done work orders and the feature extraction sets corresponding to the elements thereof, and obtaining the association interference weight corresponding to each element in the optimization processing requirement sequence of the to-be-done work orders; S42, extracting all elements whose corresponding associated interference weights are not 0 in the optimization processing requirement sequence of pending work orders, and arranging the extracted elements in descending order according to the corresponding associated interference weights, to obtain a candidate public announcement sequence for consumer complaint processing work orders; S43, obtaining a set consisting of marked elements in the optimization processing requirement sequence of pending work orders, recorded as a marked analysis set; obtaining the intersection of the set consisting of elements in the alternative public announcement sequence of consumer complaint processing work orders and the marked analysis set, to obtain a public announcement interference information set; S44, removing the elements in the public interference information set from the alternative public sequence of the consumer complaint processing work order, and recording the sequence corresponding to the remaining results as the public sequence of the consumer complaint processing work order; When generating the credit supervision warning information corresponding to each disclosure object in the disclosure sequence, the credit supervision warning information includes the disclosure sequence of consumer complaint handling work orders, a trend change chart of the proportion of consumer complaints by industry, and a fluctuation chart of the proportion of the number of complaint handling work orders disclosed by the corresponding disclosure object each time within a preset time period based on the proportion of the total number of processed work orders in the corresponding industry; The trend change graph of the proportion of consumer complaints by industry is a line graph obtained by connecting the proportion of consumer complaints by the corresponding industry in each public announcement sequence within the corresponding time period with the data pairs constituted by the corresponding public announcement sequence. The proportion of consumer complaints by the corresponding industry in the public announcement sequence is the ratio between the number of work orders in the corresponding industry in the corresponding public announcement sequence and the industry to which the public announcement object corresponding to the corresponding credit supervision early warning information belongs and the total number of work orders in the corresponding public announcement sequence; The fluctuation chart of the number of complaint handling work orders disclosed each time by the corresponding publicity object within the preset time period based on the proportion of the total number of processed work orders in the corresponding industry is a line chart obtained by connecting the first publicity quantity proportion of the publicity object corresponding to the corresponding credit supervision early warning information in each publicity sequence within the corresponding time period with the data pairs constituted by the corresponding publicity sequence. The first publicity quantity proportion of the publicity object corresponding to the corresponding credit supervision early warning information in the publicity sequence is the ratio between the number of work orders of the publicity object corresponding to the corresponding credit supervision early warning information in the corresponding publicity sequence and the number of work orders of the industry to which the publicity object corresponding to the corresponding credit supervision early warning information belongs.
2. A consumer complaint credit supervision management system based on artificial intelligence, applying the consumer complaint credit supervision management method based on artificial intelligence described in claim 1, characterized in that: The system includes the following modules: A work order feature extraction module, wherein the work order feature extraction module obtains a pending work order information set corresponding to a consumer complaint, and filters the pending work order information set in combination with the work order information to obtain an abnormal work order information set and an optimized pending work order information set; extracts information from elements in the pending work order information set to obtain a feature extraction set of each pending work order information; An import feature analysis module is used to analyze the correlation between the feature extraction sets corresponding to the elements in the to-be-done work order information optimization set, calculate the work order import priority feature index corresponding to the elements in the to-be-done work order information optimization set, and generate an optimization processing requirement sequence for the to-be-done work orders; An interactive dynamic interference analysis module, which obtains the interactive information of each element in the optimization processing requirement sequence of the to-be-done work orders in real time during the processing process, and obtains the associated interference weight corresponding to each element in the optimization processing requirement sequence of the to-be-done work orders in combination with the historical work order information records processed in the database; A regulatory feedback management module, wherein the regulatory feedback management module obtains an optimization processing requirement sequence of pending work orders, and generates a public announcement sequence of consumer complaint processing work orders in combination with the associated interference weights corresponding to each element in the optimization processing requirement sequence of pending work orders; and generates credit regulatory early warning information corresponding to each public announcement object in the public announcement sequence in combination with a trend change chart of the industry proportion of consumer complaints, and feeds back the credit regulatory early warning information to the corresponding public announcement object.
3. According to claim 2, a consumer complaint credit supervision management system based on artificial intelligence is characterized by: The import feature analysis module includes a correlation analysis unit, a feature index calculation unit and an optimization processing requirement sequence construction unit. The correlation analysis unit analyzes the correlation between the feature extraction sets corresponding to the elements in the to-be-done work order information optimization set; The characteristic index calculation unit calculates the work order import priority characteristic index corresponding to each element in the to-be-done work order information optimization set in combination with the analysis result of the correlation analysis unit; The optimization processing requirement sequence construction unit generates an optimization processing requirement sequence of the to-be-done work orders according to the analysis results of the association relationship analysis unit and the characteristic index calculation unit.
Citation Information
Patent Citations
Method and device for monitoring customer complaints
CN101730118A
Work order fusion processing method, electronic equipment and storage medium
CN116205444A
User complaint processing method and device and readable storage medium
CN118396638A
Technical entrepreneurship education service quality evaluation system based on fahp
KR101734144B1
Regulatory compliance assessment and business risk prediction system
US20200273046A1
