Abnormal customer detection method and device

Abstract

The invention discloses an abnormal customer detection method and device, and relates to the technical field of big data. A specific embodiment of the method comprises the steps of obtaining to-be-detected customer data in response to an abnormal customer detection request, and performing feature extraction on the customer data based on a preset feature dimension to generate feature data of the customer data; inputting the feature data into a pre-generated data quality evaluation model to obtain a quality evaluation result of the customer data; and determining an abnormal detection result of the customer corresponding to the customer data according to a preset abnormal customer determination rule and the quality evaluation result. According to the embodiment, the feature data is obtained by performing feature extraction on the customer data, and the quality evaluation result is generated by using the data quality evaluation model to determine the anomaly detection result, so that the customer quality can be accurately judged, and the efficiency and accuracy of abnormal customer detection are improved.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to an abnormal customer detection method and device. Background Art

[0002] When sales staff are developing business with customers, they need to test the quality of customers to filter out abnormal customers and retain normal customers. Currently, the abnormal customer detection solution is to use manual inspection by inspectors or hard coding based on experience.

[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0004] The efficiency of abnormal customer detection is low and it is highly subjective, which makes it impossible to accurately judge customer quality, affecting the accuracy of abnormal customer detection. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides an abnormal customer detection method and device, which obtains feature data by extracting features from customer data, and generates quality assessment results using a data quality assessment model to determine the abnormal detection results. This can accurately judge customer quality and improve the efficiency and accuracy of abnormal customer detection.

[0006] To achieve the above objective, according to one aspect of an embodiment of the present invention, a method for detecting abnormal customers is provided.

[0007] A method for detecting abnormal customers comprises: in response to an abnormal customer detection request, obtaining customer data to be detected, and extracting features of the customer data based on preset feature dimensions to generate feature data of the customer data; inputting the feature data into a pre-generated data quality assessment model to obtain a quality assessment result of the customer data; and determining an abnormal detection result of a customer corresponding to the customer data according to a preset abnormal customer determination rule and the quality assessment result.

[0008] Optionally, the data quality assessment model is generated in the following manner: according to the abnormal customer determination rule, a plurality of preset customer types are obtained; for each of the customer types, a customer data sample corresponding to the customer type is obtained, and features are extracted from the customer data sample based on the feature dimension to generate feature data samples of the customer data samples; according to the customer type corresponding to the feature data sample, each feature data sample is labeled to obtain a corresponding assessment result sample; the feature data sample is used as input, and the corresponding assessment result sample is used as a training target to perform model training to generate the data quality assessment model.

[0009] Optionally, the feature data includes sub-feature data of each feature dimension; and labeling each feature data sample according to the customer type corresponding to the feature data sample to obtain a corresponding evaluation result sample includes: for each feature data sample, labeling the sub-feature data samples of each feature dimension included in the feature data sample according to the customer type corresponding to the feature data sample, obtaining a sub-evaluation result sample corresponding to each sub-feature data sample, so as to generate the evaluation result sample.

[0010] Optionally, the feature data includes sub-feature data of each of the feature dimensions, and the quality assessment result includes a sub-assessment result corresponding to each of the sub-feature data; determining the abnormal detection result of the customer corresponding to the customer data according to the quality assessment result in accordance with a preset abnormal customer determination rule includes: calculating the total assessment result of the customer data according to the sub-assessment result corresponding to each of the sub-feature data; determining the abnormal detection result of the customer corresponding to the customer data according to the total assessment result and the sub-assessment result corresponding to each of the sub-feature data in accordance with the abnormal customer determination rule.

[0011] Optionally, the anomaly detection result includes the customer type of the customer; after determining the anomaly detection result of the customer corresponding to the customer data, it also includes: processing the customer data according to the anomaly detection result in accordance with a preset anomaly processing rule.

[0012] Optionally, the customer type includes suspected risk customers; when the abnormal detection result indicates that the customer is a suspected risk customer, processing the customer data according to the abnormal detection result includes: setting the status of the customer data to a suspended status, and initiating customer data approval based on the customer data; in response to receiving the approval result of the customer data, processing the customer data based on the approval result.

[0013] Optionally, the customer type also includes risky customers; when the abnormal detection result indicates that the customer is a risky customer, or the approval result indicates that the customer is a risky customer, the processing of the customer data includes: setting the status of the customer data to a closed state, and adding the customer to a blacklist.

[0014] Optionally, before extracting features from the customer data based on preset feature dimensions and generating feature data of the customer data, it also includes: determining whether the customer exists in a pre-built blacklist; and if the customer exists in the blacklist, directly setting the status of the customer data to a closed state.

[0015] According to another aspect of an embodiment of the present invention, an abnormal customer detection device is provided.

[0016] An abnormal customer detection device includes: a feature data generation module, which is used to respond to an abnormal customer detection request, obtain customer data to be detected, and extract features of the customer data based on preset feature dimensions to generate feature data of the customer data; a quality assessment result generation module, which is used to input the feature data into a pre-generated data quality assessment model to obtain a quality assessment result of the customer data; and an abnormal detection result determination module, which is used to determine the abnormal detection result of the customer corresponding to the customer data according to the quality assessment result in accordance with preset abnormal customer determination rules.

[0017] Optionally, it also includes a data quality assessment model generation module, which is used to: obtain multiple preset customer types according to the abnormal customer determination rules; for each of the customer types, obtain a customer data sample corresponding to the customer type, and extract features from the customer data sample based on the feature dimension to generate a feature data sample of the customer data sample; label each feature data sample according to the customer type corresponding to the feature data sample to obtain a corresponding assessment result sample; use the feature data sample as input and the corresponding assessment result sample as a training target to perform model training to generate the data quality assessment model.

[0018] Optionally, the feature data includes sub-feature data of each feature dimension; the data quality assessment model generation module is also used to: for each feature data sample, according to the customer type corresponding to the feature data sample, respectively label the sub-feature data samples of each feature dimension included in the feature data sample, and obtain sub-assessment result samples corresponding to each sub-feature data sample to generate the assessment result sample.

[0019] Optionally, the feature data includes sub-feature data of each feature dimension, and the quality assessment result includes sub-assessment results corresponding to each sub-feature data; the abnormal detection result determination module is also used to: calculate the total assessment result of the customer data according to the sub-assessment results corresponding to each sub-feature data; and determine the abnormal detection result of the customer corresponding to the customer data according to the total assessment result and the sub-assessment results corresponding to each sub-feature data in accordance with the abnormal customer determination rule.

[0020] Optionally, the abnormality detection result includes the customer type of the customer; the device also includes an abnormality processing module, which is used to: process the customer data according to the abnormality detection result in accordance with preset abnormality processing rules.

[0021] Optionally, the customer type includes suspected risk customers; when the abnormal detection result indicates that the customer is a suspected risk customer, the abnormal handling module is also used to: set the status of the customer data to a suspended status, and initiate customer data approval based on the customer data; in response to receiving the approval result of the customer data, process the customer data based on the approval result.

[0022] Optionally, the customer type also includes risky customers; when the exception detection result indicates that the customer is a risky customer, or the approval result indicates that the customer is a risky customer, the exception handling module is also used to: set the status of the customer data to a closed state, and add the customer to a blacklist.

[0023] Optionally, the exception handling module is further used to: determine whether the customer exists in a pre-built blacklist; if the customer exists in the blacklist, directly set the status of the customer data to a closed state.

[0024] According to yet another aspect of the embodiments of the present invention, an electronic device is provided.

[0025] An electronic device includes: one or more processors; a memory for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the abnormal customer detection method provided by an embodiment of the present invention.

[0026] According to yet another aspect of an embodiment of the present invention, a computer readable medium is provided.

[0027] A computer-readable medium stores a computer program, which, when executed by a processor, implements the abnormal customer detection method provided by an embodiment of the present invention.

[0028] An embodiment of the above invention has the following advantages or beneficial effects: by responding to an abnormal customer detection request, obtaining customer data to be detected, and extracting features from the customer data based on preset feature dimensions to generate feature data of the customer data; inputting the feature data into a pre-generated data quality assessment model to obtain a quality assessment result of the customer data; according to preset abnormal customer determination rules and based on the quality assessment result, determining the technical solution of the abnormal detection result of the customer corresponding to the customer data, extracting features from the customer data to obtain feature data, and using the data quality assessment model to generate a quality assessment result to determine the abnormal detection result, which can accurately judge the customer quality and improve the efficiency and accuracy of abnormal customer detection.

[0029] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0031] Figure 1 is a schematic diagram of the main steps of an abnormal customer detection method according to an embodiment of the present invention;

[0032] Figure 2 is a flow chart of an abnormal customer detection method according to an embodiment of the present invention;

[0033] Figure 3 is a schematic diagram of an abnormal customer detection process according to an embodiment of the present invention;

[0034] Figure 4 is a schematic diagram of the data storage principle of an embodiment of the present invention;

[0035] Figure 5 is a schematic diagram of main modules of an abnormal customer detection device according to an embodiment of the present invention;

[0036] Figure 6 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0037] Figure 7 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0039] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, storage and other aspects of user personal information involved in the technical solution of the present invention are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security and national security.

[0040] Figure 1 1 is a schematic diagram of the main steps of an abnormal customer detection method according to an embodiment of the present invention.

[0041] like Figure 1As shown, an abnormal customer detection method according to an embodiment of the present invention mainly includes the following steps S101 to S103.

[0042] Step S101: in response to an abnormal customer detection request, obtain customer data to be detected, and extract features from the customer data based on preset feature dimensions to generate feature data of the customer data, wherein the feature data may include sub-feature data of each feature dimension.

[0043] Specifically, feature extraction can be performed on the customer data to be detected through the field processing plug-in to obtain sub-feature data of each feature dimension. The sub-feature data of each feature dimension may include the customer's unique ID, contact person's mobile phone number, the customer's order quantity in the most recent preset time period, the customer's shipment quantity in the most recent preset time period, the status of the customer data (i.e., the account status of the customer data), whether the customer has made illegal quotations, whether the customer has privately changed promotional activities, etc.

[0044] Step S102: Input the feature data into a pre-generated data quality assessment model to obtain a quality assessment result of the customer data, wherein the quality assessment result may include a sub-assessment result corresponding to each sub-feature data.

[0045] In one embodiment, a data quality assessment model may be generated in the following manner: based on abnormal customer determination rules, a plurality of preset customer types are obtained; for each customer type, a customer data sample corresponding to the customer type is obtained, and features are extracted from the customer data sample based on feature dimensions to generate feature data samples of the customer data sample; for each feature data sample, sub-feature data samples of each feature dimension included in the feature data sample are respectively labeled according to the customer type corresponding to the feature data sample to obtain sub-assessment result samples corresponding to each sub-feature data sample to generate an assessment result sample; the feature data sample is used as input and the corresponding assessment result sample is used as a training target to perform model training to generate a data quality assessment model.

[0046] Specifically, the preset multiple customer types may include normal customers, suspected risk customers, and risk customers. Customer data samples corresponding to normal customers, suspected risk customers, and risk customers are obtained respectively. Then, feature extraction is performed on the customer data samples based on the feature dimensions to generate feature data samples of the customer data samples. Each feature data sample may include sub-feature data samples of each feature dimension, such as the order quantity of the customer in the most recent preset time period, the shipment quantity of the customer in the most recent preset time period, the status of the customer data, etc. According to the customer type corresponding to the feature data sample, the sub-feature data samples of each feature dimension included in the feature data sample are respectively labeled to obtain the sub-evaluation result samples corresponding to each sub-feature data sample. For example, the sub-evaluation result samples can be represented by evaluation scores. The sub-evaluation result samples of the order quantity of the normal customer in the most recent preset time period can be labeled as 80 points, the sub-evaluation result samples of the order quantity of the suspected risk customer in the most recent preset time period can be labeled as 50 points, and the sub-evaluation result samples of the order quantity of the risk customer in the most recent preset time period can be labeled as 20 points. After the labeling is completed, the feature data sample can be used as input, and the corresponding evaluation result sample can be used as the training target for model training to generate a data quality evaluation model.

[0047] In another embodiment, it is also possible to directly mark the sub-feature data sample according to its quality. For example, if the order quantity in the most recent preset time period reaches 200 pieces, the corresponding sub-evaluation result sample can be marked as 80 points. If the order quantity in the most recent preset time period does not reach 200 pieces but reaches 50 pieces, the corresponding sub-evaluation result sample can be marked as 50 points. If the order quantity in the most recent preset time period does not reach 50 pieces, the corresponding sub-evaluation result sample can be marked as 20 points. The contact's mobile phone number can be a normal mobile phone number or an abnormal mobile phone number (such as the mobile phone length does not meet the requirements, the mobile phone number has too many abnormal marks, etc.), and the evaluation score of the normal mobile phone number is higher than the evaluation score of the abnormal mobile phone number. The status of the customer data can include an activated state, a suspended state, a closed state, and an initial state, and the evaluation scores of each state decrease in the order of the activated state, the suspended state, the closed state, and the initial state.

[0048] In one embodiment, before extracting features from customer data based on preset feature dimensions and generating feature data of the customer data, it may also include: determining whether the customer exists in a pre-built blacklist; if the customer exists in the blacklist, directly setting the status of the customer data to a closed state.

[0049] Specifically, the pre-built blacklist includes multiple predetermined risky customers. The blacklist can be used to determine whether the customer to be detected already exists in the blacklist. If the customer to be detected already exists in the blacklist, it means that the customer to be detected is a risky customer, and the status of the corresponding customer data can be directly set to a closed state.

[0050] Step S103: According to the preset abnormal customer determination rules and the quality assessment results, determine the abnormal detection results of the customers corresponding to the customer data. The abnormal detection results may include the customer type of the customer, and the customer type may include normal customers, suspected risk customers, and risk customers.

[0051] According to preset abnormal customer determination rules, based on the quality assessment results, determining the abnormal detection results of the customers corresponding to the customer data can include: calculating the total assessment results of the customer data based on the sub-assessment results corresponding to each sub-feature data; according to the abnormal customer determination rules, determining the abnormal detection results of the customers corresponding to the customer data based on the total assessment results and the sub-assessment results corresponding to each sub-feature data.

[0052] Specifically, the sub-assessment results corresponding to each sub-feature data may be weighted averaged to obtain a total assessment result of the customer data, which may be a total assessment score. Then, according to the abnormal customer determination rule, the abnormal detection result of the customer corresponding to the customer data is determined based on the total assessment result and the sub-assessment results corresponding to each sub-feature data. For example, when the total assessment score is greater than 60 points, the abnormal detection result of the corresponding customer may be a normal customer. When the total assessment score is less than 30 points and the assessment score of the customer data status is lower than the score corresponding to the suspended status, the abnormal detection result of the corresponding customer may be a risk customer. Otherwise, the abnormal detection result of the corresponding customer may be a suspected risk customer.

[0053] In one embodiment, after determining the abnormality detection result of the customer corresponding to the customer data, the method may further include: processing the customer data according to the abnormality detection result according to a preset abnormality processing rule.

[0054] In one embodiment, when the anomaly detection result indicates that the customer is a suspected risk customer, processing the customer data according to the anomaly detection result may include: setting the status of the customer data to a suspended state, and initiating customer data approval based on the customer data; in response to receiving the approval result of the customer data, processing the customer data based on the approval result.

[0055] Specifically, for suspected risk customers, the suspected risk customers can be frozen by setting the status of the customer data corresponding to the suspected risk customers to a suspended status, and then an approval is initiated based on the customer data of the suspected risk customers, and manual inspection is performed manually based on the customer data and preset approval rules. The approval rules include whether the cumulative number of approvals for the suspected risk customer reaches a preset number threshold (such as 5 times), etc. The approval result includes whether the suspected risk customer is a normal customer or a risky customer. When the approval result indicates that the customer is a normal customer, the corresponding status is set to an activated status to unfreeze the suspected risk customer.

[0056] In one embodiment, when the abnormal detection result indicates that the customer is a risky customer, or the approval result indicates that the customer is a risky customer, processing the customer data may include: setting the status of the customer data to a closed state, and adding the customer to a blacklist.

[0057] Figure 2 It is a flowchart of an abnormal customer detection method according to an embodiment of the present invention.

[0058] like Figure 2 As shown, in one embodiment, first, for each customer type, a feature data sample is obtained, and an evaluation result sample is marked and generated. The feature data sample is used as input, and the corresponding evaluation result sample is used as a training target for model training to generate a data quality evaluation model. Then, the customer data to be detected is obtained, and feature data of the customer data is generated by feature extraction. The feature data is input into the data quality evaluation model to obtain the quality evaluation result of the customer data. The quality evaluation result includes the sub-evaluation result corresponding to each sub-feature data. According to each sub-evaluation result, the total evaluation result of the customer data is calculated, and according to the abnormal customer determination rule, the abnormal detection result of the corresponding customer is determined according to the total evaluation result and each sub-evaluation result. The abnormal detection result includes the customer type of the customer, and the customer type includes a normal customer, a suspected risk customer, and a risk customer. In the case where the abnormal detection result indicates that the customer is a suspected risk customer, the status of the customer data is set to a suspended state, and the customer data approval is initiated according to the customer data, and the customer data is processed based on the approval result. In the case where the abnormal detection result indicates that the customer is a risk customer, or the approval result indicates that the customer is a risk customer, the status of the customer data is set to a closed state, and the customer is added to the blacklist.

[0059] Figure 3 FIG. 1 is a schematic diagram of an abnormal customer detection process according to an embodiment of the present invention. Figure 3As shown, in an embodiment of the present invention, the input parameters of the abnormal customer detection process are customer data and the model number of the data quality assessment model preset by the current applicable system. The customer data mainly includes the contact phone number, user account, order quantity, performance account status (which may include activation status, suspension status, closed status, initial status, and quality decreases in sequence), the last delivery time, whether the price of the item is illegal, whether the promotion activity is privately changed, and other data. First, query the configured feature factors according to the model number, and perform data conversion matching to extract features. After that, query the configured feature factors according to the model number. If the features configured in the model do not exist in the customer data, it ends directly and treats the customer as a whitelist customer. Then, query the configured black and white lists according to the model number for secondary filtering. During the filtering process, the configured rule code can be executed for logical processing. If it is determined that the feature corresponding to the customer data is in the blacklist, the quality customer data is directly generated, and the index aggregation of the customer data is performed after processing. The system provides all supported function calculation interfaces of Elasticsearch (distributed search and analysis engine) by default. The data after index aggregation will be used to train the data quality assessment model for customer early warning. After that, the model file generated by the training is saved in the Redis cache.

[0060] When the system starts, the data quality assessment model automatically loads all configured data quality assessment models of the system, using the model number as the key name Key and the model file object as the key value Value, and saves them in the system memory in the form of a Map for subsequent use.

[0061] After using the data quality assessment model to score customer data, the corresponding policy processor configured according to the model automatically generates a violation letter processing pool, and business personnel will subsequently confirm the violation letters and take penalty measures.

[0062] Figure 4 FIG. 1 is a schematic diagram of the data storage principle of an embodiment of the present invention. Figure 4 As shown, in an embodiment of the present invention, in order to improve the response efficiency of the interface, the maintained features, policies, rule scripts, blacklists and other configurations can be maintained in the cache. After the system administrator or developer performs operations such as model definition, feature field definition, feature processing rule configuration, policy configuration, rule script maintenance, preprocessing field rule setting, blacklist maintenance, etc., these configuration information can be submitted to the database and broadcast uniformly through the subscription and publishing method of the message queue MQ, so that each monitoring unit can cache its own data, mainly including feature cache, feature processing cache, policy cache, rule script cache, blacklist cache, etc.

[0063] The embodiment of the present invention generates a data quality assessment model by training a model through feature data samples, extracts features from customer data to obtain feature data, uses the data quality assessment model to generate a quality assessment result, and then determines the anomaly detection result. Different anomaly processing is performed according to different anomaly detection results. It can accurately judge customer quality, improve the efficiency and accuracy of abnormal customer detection, filter out low-quality customers, and avoid the loss of high-quality customers.

[0064] Figure 5 4 is a schematic diagram of main modules of an abnormal customer detection device according to an embodiment of the present invention.

[0065] like Figure 5 As shown, an abnormal customer detection device 500 according to an embodiment of the present invention mainly includes: a feature data generation module 501, a quality assessment result generation module 502, and an abnormal detection result determination module 503.

[0066] The feature data generating module 501 is used to respond to an abnormal customer detection request, obtain customer data to be detected, and extract features from the customer data based on a preset feature dimension to generate feature data of the customer data.

[0067] The quality assessment result generating module 502 is used to input the feature data into a pre-generated data quality assessment model to obtain the quality assessment result of the customer data.

[0068] The abnormal detection result determination module 503 is used to determine the abnormal detection result of the customer corresponding to the customer data according to the preset abnormal customer determination rule and the quality assessment result.

[0069] In one embodiment, a data quality assessment model generation module (not shown in the figure) may also be included, which is used to: obtain multiple preset customer types according to abnormal customer determination rules; for each customer type, obtain a customer data sample corresponding to the customer type, and extract features from the customer data sample based on the feature dimension to generate a feature data sample of the customer data sample; label each feature data sample according to the customer type corresponding to the feature data sample to obtain a corresponding assessment result sample; use the feature data sample as input and the corresponding assessment result sample as a training target to perform model training to generate a data quality assessment model.

[0070] In one embodiment, the feature data may include sub-feature data of each feature dimension; the data quality assessment model generation module is specifically used to: for each feature data sample, according to the customer type corresponding to the feature data sample, respectively label the sub-feature data samples of each feature dimension included in the feature data sample, obtain the sub-assessment result samples corresponding to each sub-feature data sample, so as to generate an assessment result sample.

[0071] In one embodiment, the feature data may include sub-feature data of each feature dimension, and the quality assessment result may include sub-assessment results corresponding to each sub-feature data; the abnormal detection result determination module 503 is specifically used to: calculate the total assessment result of the customer data according to the sub-assessment results corresponding to each sub-feature data; and determine the abnormal detection result of the customer corresponding to the customer data according to the abnormal customer determination rule based on the total assessment result and the sub-assessment results corresponding to each sub-feature data.

[0072] In one embodiment, the anomaly detection result may include the customer type of the customer; the device may also include an anomaly processing module (not shown in the figure) for processing the customer data according to the anomaly detection result in accordance with preset anomaly processing rules.

[0073] In one embodiment, the customer type may include suspected risk customers; when the abnormal detection result indicates that the customer is a suspected risk customer, the abnormal handling module is specifically used to: set the status of the customer data to a suspended state, and initiate customer data approval based on the customer data; in response to receiving the approval result of the customer data, process the customer data based on the approval result.

[0074] In one embodiment, the customer type may also include risky customers; when the exception detection result indicates that the customer is a risky customer, or the approval result indicates that the customer is a risky customer, the exception handling module is specifically used to: set the status of the customer data to a closed state, and add the customer to a blacklist.

[0075] In one embodiment, the exception handling module is specifically used to: determine whether the customer exists in the pre-built blacklist; if the customer exists in the blacklist, directly set the status of the customer data to a closed state.

[0076] In addition, the specific implementation content of the abnormal customer detection device in the embodiment of the present invention has been described in detail in the above abnormal customer detection method, so the repeated content will not be described here.

[0077] Figure 6 An exemplary system architecture 600 to which the abnormal customer detection method or abnormal customer detection apparatus according to the embodiment of the present invention can be applied is shown.

[0078] like Figure 6 As shown, system architecture 600 may include terminal devices 601, 602, 603, network 604 and server 605. Network 604 is used to provide a medium for communication links between terminal devices 601, 602, 603 and server 605. Network 604 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0079] The user can use the terminal devices 601, 602, 603 to interact with the server 605 through the network 604 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 601, 602, 603, such as abnormal customer detection applications, business processing applications, data quality detection applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0080] The terminal devices 601 , 602 , and 603 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc.

[0081] Server 605 may be a server that provides various services, such as a background management server (only an example) that provides support for abnormal customer detection websites browsed by users using terminal devices 601, 602, and 603. The background management server may respond to abnormal customer detection requests and other data received, obtain customer data to be detected, and extract features of customer data based on preset feature dimensions to generate feature data of customer data; input the feature data into a pre-generated data quality assessment model to obtain a quality assessment result of the customer data; determine the abnormal detection result of the customer corresponding to the customer data according to the preset abnormal customer determination rules and the quality assessment result, and feed back the processing result (such as abnormal customer detection result - only an example) to the terminal device.

[0082] It should be noted that the abnormal client detection method provided in the embodiment of the present invention is generally executed by the server 605 , and accordingly, the abnormal client detection device is generally arranged in the server 605 .

[0083] It should be understood that Figure 6 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0084] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system 700 of a terminal device or a server suitable for implementing an embodiment of the present invention. Figure 7 The terminal device or server shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0085] like Figure 7As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0086] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read therefrom is installed into the storage section 708 as needed.

[0087] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-mentioned functions defined in the system of the present invention are executed.

[0088] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0089] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0090] The modules involved in the embodiments of the present invention may be implemented in software or hardware. The modules described may also be set in a processor. For example, they may be described as: a processor includes a feature data generation module, a quality assessment result generation module, and an abnormality detection result determination module. The names of these modules do not constitute a limitation on the modules themselves in some cases. For example, the feature data generation module may also be described as "a module for responding to abnormal customer detection requests, obtaining customer data to be detected, and extracting features from customer data based on preset feature dimensions to generate feature data of customer data".

[0091] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: in response to an abnormal customer detection request, obtaining customer data to be detected, and extracting features of the customer data based on a preset feature dimension to generate feature data of the customer data; inputting the feature data into a pre-generated data quality assessment model to obtain a quality assessment result of the customer data; according to the preset abnormal customer determination rule, and based on the quality assessment result, determining the abnormal detection result of the customer corresponding to the customer data.

[0092] According to the technical solution of the embodiment of the present invention, in response to an abnormal customer detection request, customer data to be detected is obtained, and feature extraction is performed on the customer data based on a preset feature dimension to generate feature data of the customer data; the feature data is input into a pre-generated data quality assessment model to obtain a quality assessment result of the customer data; and according to the preset abnormal customer determination rule and the quality assessment result, an abnormal detection result of the customer corresponding to the customer data is determined. By extracting features from customer data to obtain feature data, and using a data quality assessment model to generate a quality assessment result to determine an abnormal detection result, the customer quality can be accurately judged, and the efficiency and accuracy of abnormal customer detection can be improved.

[0093] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for detecting abnormal customers, characterized in that: include: In response to an abnormal customer detection request, obtain customer data to be detected, and extract features from the customer data based on a preset feature dimension to generate feature data of the customer data; Inputting the characteristic data into a pre-generated data quality assessment model to obtain a quality assessment result of the customer data; According to the preset abnormal customer determination rule and based on the quality assessment result, the abnormal detection result of the customer corresponding to the customer data is determined.

2. The method according to claim 1, characterized in that The data quality assessment model is generated in the following way: According to the abnormal customer determination rule, obtaining a plurality of preset customer types; For each of the customer types, obtain a customer data sample corresponding to the customer type, and perform feature extraction on the customer data sample based on the feature dimension to generate a feature data sample of the customer data sample; According to the customer type corresponding to the characteristic data sample, each characteristic data sample is labeled to obtain a corresponding evaluation result sample; The characteristic data samples are taken as input, and the corresponding evaluation result samples are taken as training targets for model training to generate the data quality evaluation model.

3. The method according to claim 2, characterized in that The feature data includes sub-feature data of each feature dimension; The step of labeling each of the characteristic data samples according to the customer type corresponding to the characteristic data sample to obtain a corresponding evaluation result sample includes: For each of the feature data samples, the sub-feature data samples of each feature dimension included in the feature data sample are labeled respectively according to the customer type corresponding to the feature data sample, and the sub-evaluation result samples corresponding to each sub-feature data sample are obtained to generate the evaluation result sample.

4. The method according to claim 1, characterized in that: The feature data includes sub-feature data of each feature dimension, and the quality assessment result includes a sub-assessment result corresponding to each sub-feature data; The determining, according to the preset abnormal customer determination rule and the quality assessment result, the abnormal customer detection result corresponding to the customer data includes: Calculate the overall evaluation result of the customer data according to the sub-evaluation results corresponding to each of the sub-feature data; According to the abnormal customer determination rule, the abnormal detection result of the customer corresponding to the customer data is determined based on the overall evaluation result and the sub-evaluation results corresponding to each of the sub-feature data.

5. The method according to claim 1, characterized in that The abnormal detection result includes the customer type of the customer; After determining the abnormality detection result of the customer corresponding to the customer data, the method further includes: According to the preset exception handling rules, the customer data is processed according to the exception detection result.

6. The method according to claim 5, characterized in that The customer types include suspected risk customers; When the abnormality detection result indicates that the customer is a suspected risk customer, the processing of the customer data according to the abnormality detection result includes: Setting the status of the customer data to a suspended status, and initiating customer data approval based on the customer data; In response to receiving the approval result of the customer data, the customer data is processed based on the approval result.

7. The method according to claim 5 or 6, characterized in that: The customer types also include risky customers; When the abnormality detection result indicates that the customer is a risky customer, or the approval result indicates that the customer is a risky customer, the processing of the customer data includes: The status of the customer data is set to closed, and the customer is added to the blacklist.

8. The method according to claim 1, characterized in that Before extracting features from the customer data based on a preset feature dimension to generate feature data of the customer data, the method further includes: Determining whether the customer exists in a pre-built blacklist; In the case that the customer exists in the blacklist, the status of the customer data is directly set to a closed state.

9. An abnormal customer detection device, characterized in that: include: A feature data generation module, used to respond to an abnormal customer detection request, obtain customer data to be detected, and extract features from the customer data based on a preset feature dimension to generate feature data of the customer data; A quality assessment result generating module, used for inputting the feature data into a pre-generated data quality assessment model to obtain a quality assessment result of the customer data; The abnormal detection result determination module is used to determine the abnormal detection result of the customer corresponding to the customer data according to the preset abnormal customer determination rule and the quality assessment result.

10. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 8.

11. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images