Information processing method and device

Abstract

The present invention discloses an information processing method and device, which relate to the field of computer technology. A specific implementation of the method includes: receiving an after-sales information processing request; the after-sales information processing request includes user information; obtaining real-time after-sales data and historical after-sales data corresponding to the user information; processing the real-time after-sales data and historical after-sales data; inputting the processing results into one or more after-sales processing models to obtain risk values output by the one or more after-sales processing models; one or more after-sales processing models respectively correspond to processing types; determining a processing strategy based on the risk value and the processing type; processing the after-sales information processing request based on the processing strategy. This implementation facilitates accurate judgment of the processing type and corresponding after-sales processing language corresponding to the after-sales information processing request, thereby improving the processing efficiency of the after-sales information processing request and saving company costs.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an information processing method and device. Background Art

[0002] In the current e-commerce customer service after-sales system, customer service only uses a single user risk tag (e.g., whether the user is a risky customer) to assist customer service in reviewing user risk. This results in customer service being unable to accurately judge unreasonable user requests based on a single user risk tag, and thus unable to accurately handle after-sales information processing requests based on the risk tag, resulting in unnecessary cost waste. Summary of the Invention

[0003] In view of this, an embodiment of the present invention provides an information processing method and device to process real-time after-sales data and historical after-sales data corresponding to user information. A processing strategy containing one or more processing types can be obtained based on the real-time after-sales data, historical after-sales data and one or more after-sales processing models, and then the after-sales information processing request can be processed according to the processing strategy, thereby facilitating accurate judgment of the processing type and corresponding after-sales processing script corresponding to the after-sales information processing request, thereby improving the processing efficiency of the after-sales information processing request and saving company costs.

[0004] To achieve the above objective, according to a first aspect of an embodiment of the present invention, an information processing method is provided.

[0005] The information processing method of an embodiment of the present invention includes: receiving an after-sales information processing request; the after-sales information processing request includes user information; obtaining real-time after-sales data and historical after-sales data corresponding to the user information; processing the real-time after-sales data and historical after-sales data; inputting the processing results into one or more after-sales processing models to obtain risk values output by the one or more after-sales processing models; the one or more after-sales processing models respectively correspond to processing types; determining a processing strategy based on the risk value and the processing type; and processing the after-sales information processing request based on the processing strategy.

[0006] Optionally, the processing of the real-time after-sales data and the historical after-sales data and inputting the processing results into one or more after-sales processing models includes: performing feature processing on the real-time after-sales data and the historical after-sales data to generate feature processing results; and inputting the feature processing results into the after-sales processing model to obtain a risk value.

[0007] Optionally, the feature processing of the real-time after-sales data and the historical after-sales data to generate feature processing results includes: filtering the historical after-sales data according to one or more preset time periods to obtain one or more filtering results; and feature processing of the real-time after-sales data and one or more filtering results respectively to generate feature processing results.

[0008] Optionally, the method further includes: determining a user risk level corresponding to the user information based on a proportion of historical risk data included in the historical after-sales data; and determining a processing strategy corresponding to the user risk level.

[0009] Optionally, the historical risk data includes risk category data;

[0010] The method further includes: determining a processing type of the after-sales information processing request according to a matching result between the product category included in the real-time after-sales data and the risk category data.

[0011] Optionally, the processing strategy includes after-sales processing scripts corresponding to the processing type; determining the processing strategy based on the risk value and the processing type includes: determining a risk threshold based on the processing type; determining the terms for allowing or refusing to process the after-sales information processing request based on the relationship between the risk value and the risk threshold; generating the after-sales processing scripts based on the scenario terms corresponding to the processing type and the terms for allowing or refusing to process the after-sales information processing request.

[0012] Optionally, the method further includes: receiving feedback data about the processing strategy; and using the after-sales information processing request as a positive sample or a negative sample based on a label of whether the processing strategy is incorrect included in the feedback data to optimize one or more after-sales processing models.

[0013] Optionally, after receiving the feedback data after-sales information processing request about the processing strategy, the method includes: when the feedback data indicates that the after-sales information processing request is a risk request, monitoring and identifying user behavior corresponding to the user information.

[0014] To achieve the above objective, according to a second aspect of an embodiment of the present invention, an information processing device is provided.

[0015] An information processing device according to an embodiment of the present invention includes:

[0016] A receiving module, configured to receive an after-sales information processing request, wherein the after-sales information processing request includes user information;

[0017] An acquisition module, configured to acquire real-time after-sales data and historical after-sales data corresponding to the user information;

[0018] A data module, used for processing the real-time after-sales data and historical after-sales data;

[0019] A prediction module, configured to input the processing result into one or more after-sales processing models to obtain a risk value output by the one or more after-sales processing models; the one or more after-sales processing models respectively correspond to a processing type;

[0020] A policy module is used to determine a processing policy according to the risk value and the processing type; and a processing module is used to process the after-sales information processing request according to the processing policy.

[0021] To achieve the above objective, according to a third aspect of an embodiment of the present invention, a device for processing a request is provided.

[0022] A device for processing requests in an embodiment of the present invention includes: one or more processors; a storage system 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 information processing method in the after-sales scenario of an embodiment of the present invention.

[0023] To achieve the above objective, according to a fourth aspect of an embodiment of the present invention, a computer-readable medium is provided.

[0024] A computer program is stored on a computer-readable medium according to an embodiment of the present invention. When the program is executed by a processor, the information processing method according to the embodiment of the present invention is implemented.

[0025] An embodiment of the above invention has the following advantages or beneficial effects: In an embodiment of the present invention, the real-time after-sales data and historical after-sales data corresponding to the user information are processed, and a processing strategy containing one or more processing types can be obtained based on the real-time after-sales data, historical after-sales data and one or more after-sales processing models, and then the after-sales information processing request can be processed according to the processing strategy, thereby facilitating accurate judgment of the processing type and corresponding after-sales processing script corresponding to the after-sales information processing request, thereby improving the processing efficiency of the after-sales information processing request, and saving company costs.

[0026] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings are provided for a better understanding of the present invention and are not intended to limit the present invention.

[0028] Figure 1 This is a schematic diagram of the main process of an information processing method according to an embodiment of the present invention;

[0029] Figure 2 is a schematic diagram of a main process of determining inputs of one or more after-sales processing models according to an embodiment of the present invention;

[0030] Figure 3 Schematic diagram of the main process of generating feature processing results according to an embodiment of the present invention;

[0031] Figure 4 is a schematic diagram of a main process of determining a processing strategy according to an embodiment of the present invention;

[0032] Figure 5 This is a schematic diagram of the main process of determining a processing strategy according to a risk value and a processing type according to an embodiment of the present invention;

[0033] Figure 6 It is a schematic diagram of the main process after manual review of the processing strategy in an embodiment of the present invention;

[0034] Figure 7 This is a schematic diagram of a risk control system for prompting e-commerce customer service to conduct review based on risk identification according to an embodiment of the present invention;

[0035] Figure 8 is a schematic diagram of the internal system of the risk control system according to an embodiment of the present invention;

[0036] Figure 9 A schematic diagram of the main modules of a risk identification device according to an embodiment of the present invention;

[0037] Figure 10 An exemplary system architecture diagram in which embodiments of the present invention may be applied;

[0038] Figure 11 A schematic diagram of the structure of a computer system of a terminal device or server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following description of exemplary embodiments of the present invention is made in conjunction with the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding. These details should be considered as merely exemplary. Therefore, it should be appreciated by those skilled 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 the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0040] According to a first aspect of an embodiment of the present invention, an information processing method is provided.

[0041] Figure 1 This is a schematic diagram of the main process of an information processing method according to an embodiment of the present invention. Figure 1 As shown, the method mainly includes:

[0042] Step S101: receiving an after-sales information processing request; the after-sales information processing request includes user information;

[0043] Step S102: acquiring real-time after-sales data and historical after-sales data corresponding to the user information according to the user information;

[0044] Step S103: Processing the real-time after-sales data and historical after-sales data;

[0045] Step S104: inputting the processing result into one or more after-sales processing models to obtain the risk value output by the one or more after-sales processing models; the one or more after-sales processing models respectively correspond to a processing type;

[0046] Step S105: Determine a processing strategy based on the risk value and the processing type; Step S106: Process the after-sales information processing request based on the processing strategy.

[0047] On existing e-commerce platforms, after-sales information processing requests refer to user-initiated requests in after-sales scenarios, primarily including requests for returns with refunds, exchanges, and compensation (refunds without returns). Merchants generally accept normal customer requests for returns and exchanges, but with the rapid development of information technology, malicious refund and return requests are on the rise. For example, counterfeit-for-old product swapping involves requesting an after-sales exchange, replacing genuine goods with counterfeit goods, then returning them and requesting a refund. This is done to exchange counterfeit goods for genuine goods. Therefore, risk identification for malicious return and exchange behaviors is necessary during the after-sales phase to mitigate losses for merchants and platforms.

[0048] In an optional embodiment, the user information included in the after-sales information processing request includes: user device code, user mobile phone number, delivery address, recipient, and the IP address of the current login, etc. Based on the different information in the user information, the historical data corresponding to a certain information can be retrieved from the database of the after-sales system. Among them, user information can only be used with the permission of the user. The present invention does not limit the method of user permission, which can be fingerprint, bone pattern, agreement permission, etc. For example, for obtaining historical after-sales data based on user information, the user's historical order data can be obtained based on the user's mobile phone number, or the user's device code, delivery address, recipient, and the IP address of the current login can be used to retrieve the associated historical order data of the same device code, the same delivery address, the same recipient, and the same IP address from the database. Although these associated historical order data may not be for the same user (for example, if different users log in to the same device, the historical order data of multiple users associated with the device can be obtained based on the device code), for malicious thieves, this associated information may be frequently stolen. Therefore, obtaining this associated historical order data can provide more accurate identification results for subsequent risk identification.

[0049] Real-time after-sales data includes relevant data of the user's after-sales information processing request, such as return address, user mobile phone number, return reason, return time, order time, etc., to determine whether there is any risk in the user's after-sales information processing request.

[0050] Historical after-sales data usually includes the number of historical return orders, return amount, number of related orders with the same mobile phone number and device, number of purchases, data on coupon usage, etc. Historical after-sales data can be used to determine whether there are risks in the user's historical consumption process.

[0051] The types of after-sales processing models are usually summarized based on common and frequently occurring processing types. In an optional embodiment, different after-sales processing models correspond to different processing types. For example, the after-sales processing model for the processing type of address cheating risk is an address anti-cheating model.

[0052] Regarding the input of determining one or more after-sales processing models in step S103, in an optional embodiment, as shown in FIG. Figure 2 As shown, specifically including:

[0053] Step S201: performing feature processing on real-time after-sales data and historical after-sales data to generate feature processing results;

[0054] Step S202: Input the feature processing results into one or more after-sales processing models to obtain a risk value.

[0055] Because real-time and historical after-sales data contain a wide variety of data types, they cannot be directly used as input for different after-sales processing models. Therefore, feature processing is required for real-time and historical after-sales data. Specifically, in an optional embodiment, the feature processing algorithm includes any one or more of the following: data summation, averaging, variance calculation, normalization algorithm, discretization algorithm, missing value processing, and exponential conversion. Using these algorithms, multiple different feature processing results can be obtained based on real-time and historical after-sales data.

[0056] Feature processing is the process of calculating real-time and historical after-sales data based on the aforementioned algorithm. Multiple relevant values for different fields can be obtained from these data. For example, if one of the fields is the address field, then after feature processing, multiple relevant values related to the address field can be obtained: the number of different delivery provinces in the past day, the rate of change of the number of delivery addresses in the past 30 days / the number of delivery addresses in the past 180 days, the proportion of provinces among the top 1 delivery addresses in the past 30 days, and so on. Therefore, after feature processing, the complex real-time and historical after-sales data can be organized into multiple inputs related to multiple different fields. Then, according to the input fields required for each after-sales processing model, different processing types and risk values can be output.

[0057] In an optional embodiment, as Figure 3 As shown, step S201 may include:

[0058] Step S301: Filter historical after-sales data according to one or more preset time periods to obtain one or more filtering results;

[0059] Step S302: performing feature processing on the real-time after-sales data and one or more screening results respectively to generate feature processing results.

[0060] The purpose of filtering is to select historical after-sales data with greater timeliness as model input to achieve more accurate prediction results. For example, the address field for the quantity of goods delivered to different provinces can be derived from historical after-sales data from the past 10 days, 30 days, or even 180 days. In actual applications, some user accounts may have been stolen for a period of time, but after discovering the theft, they retrieved the account. Therefore, historical data that is too old may mistakenly identify these accounts as risky accounts, increasing the difficulty of identification. Therefore, filtering the data before feature processing and then processing the filtered results can achieve more accurate prediction results.

[0061] Regarding step S202, in an optional embodiment, a first threshold may be manually set to classify the risk value. When the risk value output by the after-sales processing model is greater than or equal to the first threshold, the risk value is high, and it can be considered that the current after-sales information processing request has a processing type corresponding to the after-sales processing model. When the risk value output by the after-sales processing model is less than the first threshold, the risk value is low, and it can be considered that the current after-sales information processing request does not have a processing type corresponding to the after-sales processing model.

[0062] However, since the samples used in the training process of the after-sales processing model cannot satisfy an infinite number of possible processing types, even if the risk value output by the after-sales processing model is less than the first threshold, the current after-sales information processing request may still be risky. In an optional embodiment, if the risk value output by one or more after-sales processing models is less than the first threshold, the following steps are performed:

[0063] Step 1: Determine whether the feature processing result corresponding to the risk value less than the first threshold satisfies the risk strategy;

[0064] Step 2: Based on the judgment result, determine the risk result corresponding to the risk value less than the first threshold.

[0065] In an embodiment of the present application, the risk strategy can be a processing type and rule that cannot be accurately predicted by the risk prediction model based on the summary of historical risk data. For example, it is usually only possible to identify the four-level address (province, city, district, street), but it is not possible to identify the specific community and house number. When encountering a situation where some addresses are followed by special symbols, the address anti-cheating model can be judged as a false address, but when the community and house number are wrong, the address anti-cheating model cannot make an accurate judgment. The risk user takes advantage of this shortcoming of the model and enters a false address in the details of the address. For example, there are 8 buildings in Community A, and each building has only 5 units. The address entered in the after-sales information processing request is Unit 10, Building 12, Community A. At this time, the risk strategy can be judged and the processing type corresponding to this after-sales application is output as a false address risk.

[0066] In addition to determining the processing type based on the current after-sales application, in an optional embodiment, the user's risk level can also be determined, such as Figure 4 As shown, specifically including:

[0067] Step S401: determining the user risk level corresponding to the user information based on the proportion of historical risk data included in the historical after-sales data;

[0068] Step S401: Determine a processing strategy corresponding to the user risk level.

[0069] Exemplarily, historical risk data is historical after-sales data with a processing type label. For example, user A has a total of 100 historical after-sales data, including 30 return orders. Among the return orders, 10 orders have the risk of sudden return, and the remaining 20 return orders are considered to be returns due to normal reasons. Then the proportion of historical risk data is 10%.

[0070] In a further optional embodiment, the user risk level can be divided differently according to the proportion of different historical risk data. For example, when the proportion of historical risk data is 0-10%, the user risk level is level 1 low risk; when the proportion of historical risk data is 10%-30%, the user risk level is level 2 low risk; when the proportion of historical risk data is 30%-50%, the user risk level is level 3 medium risk; when the proportion of historical risk data is 50%-80%, the user risk level is level 4 high risk, and so on.

[0071] For step S401, when the user risk level is different, even if the after-sales processing model is the same and the output risk value is the same, the determined processing strategy is different. In an optional embodiment, for different user risk levels, the setting value of the first threshold is different, and the higher the risk level, the lower the setting value of the first threshold. For example, when the user risk level is level 1 low risk, the first threshold is set to 0.8, that is, when the output risk value is greater than or equal to 0.8, the user is considered to be at risk under the processing type corresponding to the after-sales processing model; when the user risk level is level 4 high risk, the first threshold is set to 0.4, that is, the output risk value only needs to be greater than or equal to 0.4 to be considered to be at risk under the processing type corresponding to the after-sales processing model.

[0072] In an optional embodiment, the historical risk data includes risk category data, and the method further includes: determining a processing type of the after-sales information processing request based on a matching result between the categories included in the real-time after-sales data and the risk category data.

[0073] In an optional embodiment, the processing type includes category risk. Risk category data refers to historical risk data for a category that reaches a second threshold in the historical risk data. For example, if there are 10 historical risk data items in the after-sales historical data, and 8 of them were generated when a user purchased home appliances, then the home appliances category is a risk category for that user.

[0074] In another optional embodiment, the processing type includes the risk of counterfeit and old package substitution. The counterfeit and old package substitution risk data can be data in historical risk data indicating that the after-sales application time and order time are too close. When the proportion of counterfeit and old package substitution risk data in the historical risk data reaches a third threshold, the current user is considered to be at risk of counterfeit and old package substitution.

[0075] In an optional embodiment, the processing strategy includes after-sales processing words corresponding to the processing type. For the specific process of determining the processing strategy according to the risk value and the processing type in step S105, as shown in FIG. Figure 5 Shown, including:

[0076] Step S501: determining a risk threshold according to the processing type;

[0077] Step S502: determining the terms for allowing or rejecting the after-sales information processing request based on the relationship between the risk value and the risk threshold;

[0078] Step S503: Generate after-sales processing scripts based on the scenario terms corresponding to the processing type and the terms for allowing or rejecting the processing of the after-sales information processing request.

[0079] Among them, the after-sales processing script will be generated according to the characteristics of each processing type, saving the time required for subsequent customer service to organize language when feedback the review results to the user. The risk handling script can be directly fed back to the user, or simple changes can be made, thereby improving the efficiency of the entire manual review process.

[0080] For example, different processing types correspond to different risk thresholds, which are mainly determined by the degree of loss to merchants caused by the processing type. For example, the category risk in the processing type has a lower degree of loss to merchants than the counterfeit and old product substitution risk, that is, the risk threshold for category risk can be set to 0.7, and the risk threshold for counterfeit and old product substitution risk can be set to 0.3. Assume that after-sales information processing request a is processed, and after-sales information processing request a is a replacement request, by processing real-time after-sales data and historical after-sales data, the final risk value output by the after-sales processing model for the category risk processing type is 0.6, and the risk value output by the after-sales processing model for the counterfeit and old product substitution risk type is 0.4. It is considered that the after-sales information processing request has the counterfeit and old product substitution risk, but does not have the category risk. Since after-sales information processing request a is a replacement request, which is highly consistent with the risk of counterfeit and old product substitution, the after-sales information processing request will be rejected. The corresponding term for rejecting the after-sales information processing request can be "does not meet the replacement requirements", and the scenario term corresponding to the processing type can be "the system determines that the order has the risk of counterfeit and old product substitution". By combining the above two terms, the generated after-sales processing language is "Hello customer, the system determines that the order has the risk of counterfeit and old product substitution, and does not meet the replacement requirements. Please contact customer service for return processing. If you have any questions, please call xxxx."

[0081] Similarly, if the after-sales information processing request does not have the risk of counterfeit and old product substitution, but does have category risk, it means that the exchange request is most likely not a malicious request, and the after-sales information processing request will be allowed to be processed. The corresponding term for allowing the after-sales information processing request to be processed can be "meeting the exchange requirements", and the scenario term corresponding to the processing type can be "the system determines that the order has category risk". By combining the above two terms, the generated after-sales processing language is "Hello customer, the system determines that the order may have category risk, but because of your good overall reputation, we can complete the exchange service for you. If you have any questions, please call xxxx."

[0082] In an optional embodiment, after manual review of the processing strategy, such as Figure 6 As shown, the method further includes:

[0083] Step S601: receiving feedback data on the processing strategy;

[0084] Step S602: Based on the label of whether the processing strategy included in the feedback data is wrong, the after-sales information processing request is used as a positive sample or a negative sample to optimize one or more after-sales processing models.

[0085] Among them, manual review can be the review feedback on risk results sent by the customer service system, which is obtained by customer service personnel through manual judgment based on the risk results received. In an optional embodiment, the content of the feedback generally includes: whether the processing strategy obtained by the risk control system is wrong, whether the current after-sales information processing request is a risk request, and the review result (whether the user's after-sales information processing request is agreed).

[0086] If the review feedback indicates that the processing strategy obtained by the risk control system is correct, the current after-sales information processing request is used as a positive sample. If the review feedback indicates that the processing strategy obtained by the risk control system is incorrect, the processing strategy obtained by the risk control system is incorrect, which may be due to a low sample size in the after-sales processing model, resulting in a prediction error. The current after-sales information processing request is used as a negative sample. Both positive and negative samples can be used to subsequently optimize one or more after-sales processing models to obtain a more accurate processing strategy.

[0087] In an optional embodiment, Figure 7 This paper presents a design for a risk control system that helps e-commerce companies overcome audit challenges based on risk identification. The design includes an after-sales system, a risk control system, an audit system, and a user stratification system. The information processing method and device provided by the embodiments of the present invention can be configured and applied within the risk control system. The specific steps are as follows:

[0088] Step 1: The after-sales system receives the user application (after-sales service application) and sends the user information to the risk control system;

[0089] Step 2: The risk control system obtains the user's real-time behavior data (real-time after-sales data), user historical data (historical after-sales data), and user basic attribute data corresponding to the user information from the data center based on the user information;

[0090] Step 3: Perform feature processing on the acquired data and obtain feature processing results;

[0091] Step 4: Input the feature processing results into different post-sales processing models (such as complex network cluster model, address anti-fraud model, marketing arbitrage model), and output the risk values corresponding to different post-sales processing models;

[0092] Step 5: When the risk value output by one or more post-sales processing models is less than the first threshold, the policy center determines whether the feature processing results corresponding to the risk value less than the first threshold satisfy the risk strategy (rule set, policy set, decision set, etc.) and generates a judgment result.

[0093] Step 6: The decision center determines the risk result of the current user's after-sales application based on the risk values and judgment results output by different after-sales processing models;

[0094] Step 7: The decision center sends the obtained processing strategy to the user stratification system. Based on the user risk scenarios obtained by the user stratification system, the risk and processing strategy are prompted. The review system then obtains the service order scenario-based risk prompts.

[0095] Step 8: Manual customer service reviews the service order based on the risk prompts of each scenario to determine whether the current user's application is approved. If not, the review is terminated and the approval indication is sent to the after-sales system; otherwise, the after-sales application is stored in the sample library of the risk control system as a predicted negative sample for subsequent model training.

[0096] In an optional embodiment, after receiving the after-sales information processing request of the feedback data about the processing strategy, the method includes: when the feedback data indicates that the after-sales information processing request is a risk request, monitoring and identifying the user behavior corresponding to the user information.

[0097] For example, Figure 8As shown, in the risk control system, risk identification is performed jointly by the data system and the identification system. The data system is primarily used for risk identification, while the identification system assists the data system by acquiring user information for the data system through various identification operations. Specifically, the identification system may include: real-name verification systems, SMS verification systems, voice verification systems, early warning systems, and rights restriction systems. User behaviors may include: user registration, user login, coupon redemption, red envelope redemption, order placement, order acceptance / rejection, after-sales returns and exchanges, after-sales pickup, and after-sales service rights. If a request for after-sales information processing is deemed risky, all user behavior after the request is initiated is monitored and identified to enhance the accuracy of user identity throughout each operation. If a legitimate user's account is compromised, subsequent monitoring and identification will hinder further account theft, ensuring the security of the current account.

[0098] Furthermore, when the feedback data indicates that the after-sales information processing request is a risky request, account theft warning information and risk warning information can be sent to the currently logged-in user account, and the user can be prompted to change the account password in time by sending a reminder text message to the user's mobile phone number to ensure the subsequent normal use of the account.

[0099] The information processing method of an embodiment of the present invention processes the real-time after-sales data and historical after-sales data corresponding to the user information. Based on the real-time after-sales data, historical after-sales data and one or more after-sales processing models, a processing strategy containing one or more processing types can be obtained, and then the after-sales information processing request can be processed according to the processing strategy. A processing strategy containing processing types can be provided for the subsequent customer service system to facilitate customer service to make accurate judgments on unreasonable demands of users, thereby saving company costs.

[0100] According to a second aspect of an embodiment of the present invention, an information processing apparatus is provided.

[0101] Figure 9 Schematic diagram of the main modules of an information processing device 900 provided according to an embodiment of the present invention. Figure 9 Shown, including:

[0102] Receiving module 901, configured to receive an after-sales information processing request; the after-sales information processing request includes user information;

[0103] An acquisition module 902 is configured to acquire real-time after-sales data and historical after-sales data corresponding to the user information based on the user information;

[0104] Prediction module 903, configured to input the processing result into one or more post-sales processing models to obtain risk values output by the one or more post-sales processing models; the one or more post-sales processing models respectively correspond to processing types;

[0105] A strategy module 904 is configured to determine a processing strategy based on the risk value and the processing type;

[0106] The processing module 905 is configured to process the after-sales information processing request according to the processing strategy.

[0107] In one embodiment of the present invention, the prediction module 903 is further used to perform feature processing on the real-time after-sales data and the historical after-sales data to generate feature processing results; and use the feature processing results as input to the one or more risk prediction models.

[0108] In one embodiment of the present invention, the prediction module 903 is further used to filter the historical after-sales data according to one or more preset time periods to obtain one or more filtering results; and perform feature processing on the real-time after-sales data and one or more filtering results respectively to generate feature processing results.

[0109] In one embodiment of the present invention, the policy module 904 is further configured to determine the user risk level corresponding to the user information based on the proportion of historical risk data included in the historical after-sales data; and determine a processing strategy corresponding to the user risk level.

[0110] In one embodiment of the present invention, the historical risk data includes risk category data; the prediction module 903 is further used to determine the processing type of the after-sales information processing request based on the matching result between the category included in the real-time after-sales data and the risk category data.

[0111] In one embodiment of the present invention, the processing strategy includes after-sales processing words corresponding to the processing type; the strategy module 904 is further used to determine the risk threshold according to the processing type; determine the terms for allowing or refusing to process the after-sales information processing request according to the relationship between the risk value and the risk threshold; generate the after-sales processing words according to the scenario terms corresponding to the processing type and the terms for allowing or refusing to process the after-sales information processing request

[0112] In one embodiment of the present invention, the receiving module 901 is also used to receive feedback data about the processing strategy; the device also includes a storage module for using the after-sales information processing request as a positive sample or a negative sample based on a label of whether the processing strategy is wrong included in the feedback data, so as to optimize one or more after-sales processing models.

[0113] In one embodiment of the present invention, the device also includes a monitoring and identification module for monitoring and identifying user behavior corresponding to the user information after receiving the feedback data after-sales information processing request regarding the processing strategy, when the feedback data characterizes that the after-sales information processing request is a risk request.

[0114] Figure 10 An exemplary system architecture 1000 of an information processing system to which embodiments of the present invention may be applied is shown.

[0115] like Figure 10 As shown, system architecture 1000 may include terminal devices 1001, 1002, and 1003, a network 1004, and a server 1005. Network 1004 is used to provide a medium for communication links between terminal devices 1001, 1002, and 1003 and server 1005. Network 1004 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0116] Users can use terminal devices 1001, 1002, and 1003 to interact with server 1005 via network 1004 to send task execution requests or receive response information to the requests. Various communication client applications can be installed on terminal devices 1001, 1002, and 1003, such as online service applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0117] The terminal devices 1001 , 1002 , and 1003 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.

[0118] Server 1005 may be a server that provides various services, such as a backend management server that supports online service requests sent by users using terminal devices 1001, 1002, and 1003, or a server that processes after-sales information processing requests. The backend management server may analyze and process received data such as after-sales information processing requests, and feedback the processing results (e.g., whether the after-sales information processing request is approved) to the terminal device.

[0119] It should be noted that the information processing method provided by the embodiment of the present invention is generally executed by the server 1005 , and accordingly, the information processing device provided by the embodiment of the present invention is generally set in the server 1005 .

[0120] It should be understood that Figure 10 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0121] Reference below Figure 11 , which shows a schematic structural diagram of a computer system 1100 of a terminal device suitable for implementing an embodiment of the present invention. Figure 11 The terminal device 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.

[0122] like Figure 11 As shown, the computer system 1100 includes a central processing unit (CPU) 1101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1102 or a program loaded from a storage unit 1108 into a random access memory (RAM) 1103. Various programs and data required for the operation of the system 1100 are also stored in the RAM 1103. The CPU 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0123] The following components are connected to the I / O interface 1105: an input section 1106 including a keyboard, a mouse, and the like; an output section 1105 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and speakers; a storage section 1108 including a hard disk; and a communication section 1109 including a network interface card such as a LAN card or a modem. The communication section 1109 performs communication processing via a network such as the Internet. A drive 1110 is also connected to the I / O interface 1105 as needed. Removable media 1111, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 1110 as needed, so that computer programs read therefrom can be installed in the storage section 1108 as needed.

[0124] In particular, according to the embodiments disclosed in the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 1109, and / or installed from a removable medium 1111. When the computer program is executed by the central processing unit (CPU) 1101, the above-mentioned functions defined in the system of the present invention are performed.

[0125] It should be noted that the computer-readable medium described in the present invention may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. 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 conjunction with an instruction execution system, system, or device. In the present invention, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying 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 thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, system, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wireline, optical cable, RF, or any suitable combination thereof.

[0126] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that 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 flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0127] The modules described in the embodiments of the present invention may be implemented in software or hardware. The modules described may also be provided within a processor. For example, they may be described as comprising a receiving module, an acquisition module, a prediction module, and a processing module. The names of these modules do not, in some cases, limit the modules themselves. For example, the receiving module may also be described as a "module for receiving after-sales information processing requests."

[0128] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiments, or may exist independently without being incorporated into the device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by the device, the device includes:

[0129] Receive an after-sales information processing request; the after-sales information processing request includes user information; obtain real-time after-sales data and historical after-sales data corresponding to the user information; determine the input of one or more after-sales processing models based on the real-time after-sales data and historical after-sales data; the one or more after-sales processing models respectively correspond to processing types; determine a processing strategy based on the risk value output by the one or more after-sales processing models and the processing type; and process the after-sales information processing request according to the processing strategy.

[0130] An information processing method and device according to an embodiment of the present invention processes real-time after-sales data and historical after-sales data corresponding to user information. A processing strategy including one or more processing types can be obtained based on the real-time after-sales data, historical after-sales data, and one or more after-sales processing models. After-sales information processing requests can then be processed according to the processing strategy. A processing strategy including processing types can be provided for subsequent customer service systems to facilitate customer service to make accurate judgments on unreasonable user demands, thereby saving company costs.

[0131] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. An information processing method, characterized in that: include: Receive after-sales information processing requests; The after-sales information processing request includes user information; Acquire real-time after-sales data and historical after-sales data corresponding to the user information; Processing the real-time after-sales data and historical after-sales data; Inputting the processing results into one or more after-sales processing models to obtain risk values output by the one or more after-sales processing models; the one or more after-sales processing models respectively correspond to processing types; determining a treatment strategy according to the risk value and the treatment type; Processing the after-sales information processing request according to the processing strategy; It also includes: determining the user risk level corresponding to the user information based on the proportion of historical risk data included in the historical after-sales data; wherein the historical risk data is historical after-sales data with a processing type label; and determining a processing strategy corresponding to the user risk level.

2. The method according to claim 1, characterized in that The processing of the real-time after-sales data and the historical after-sales data and inputting the processing results into one or more after-sales processing models includes: Performing feature processing on the real-time after-sales data and the historical after-sales data to generate feature processing results; The feature processing result is input into the after-sales processing model to obtain the risk value.

3. The method according to claim 2, characterized in that The performing feature processing on the real-time after-sales data and the historical after-sales data to generate feature processing results includes: Filtering the historical after-sales data according to one or more preset time periods to obtain one or more filtering results; Feature processing is performed on the real-time after-sales data and one or more screening results to generate feature processing results.

4. The method according to claim 1, wherein The historical risk data includes risk category data; and also includes: The processing type of the after-sales information processing request is determined according to a matching result between the product category included in the real-time after-sales data and the risk category data.

5. The method according to claim 1, wherein The processing strategy includes after-sales processing words corresponding to the processing type; The determining of a processing strategy according to the risk value and the processing type includes: determining a risk threshold based on the type of treatment; determining, based on a magnitude relationship between the risk value and the risk threshold, terms for allowing or refusing to process the after-sales information processing request; The after-sales processing script is generated based on the scenario terms corresponding to the processing type and the terms for allowing or rejecting the processing of the after-sales information processing request.

6. The method according to claim 1, characterized in that Also includes: receiving feedback data regarding the processing strategy; According to a label of whether the processing strategy is wrong included in the feedback data, the after-sales information processing request is used as a positive sample or a negative sample to optimize one or more after-sales processing models.

7. The method according to claim 6, characterized in that After receiving the request for after-sales information processing of the feedback data regarding the processing strategy, the method further includes: When the feedback data indicates that the after-sales information processing request is a risky request, user behavior corresponding to the user information is monitored and identified.

8. An information processing device, characterized in that include: A receiving module, used for receiving after-sales information processing requests; The after-sales information processing request includes user information; An acquisition module, configured to acquire real-time after-sales data and historical after-sales data corresponding to the user information; A data module, used for processing the real-time after-sales data and historical after-sales data; A prediction module, configured to input the processing result into one or more after-sales processing models to obtain a risk value output by the one or more after-sales processing models; the one or more after-sales processing models respectively correspond to a processing type; A strategy module, configured to determine a processing strategy according to the risk value and the processing type; Determining a user risk level corresponding to the user information based on a proportion of historical risk data included in the historical after-sales data; wherein the historical risk data is historical after-sales data with a processing type tag; and determining a processing strategy corresponding to the user risk level; A processing module is used to process the after-sales information processing request according to the processing strategy.

9. A device for processing information, characterized in that include: one or more processors; a storage system 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 7.

10. A computer-readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images