Product pushing method and device based on intelligent outbound call technology and big data analysis
By combining intelligent outbound calling technology with big data analytics, a screening model and sentiment analysis were constructed, solving the problems of low efficiency and poor accuracy of manual outbound calling in the promotion of financial products. This achieved automated customer screening and marketing loop, improving accuracy and user experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- INDUSTRIAL AND COMMERCIAL BANK OF CHINA
- Filing Date
- 2022-09-05
- Publication Date
- 2026-07-31
AI Technical Summary
The promotion of existing financial products suffers from problems such as low efficiency, poor accuracy, and poor user experience in manual outbound calls. In particular, when the demand is unclear, conventional relationship graph recommendation methods lack flexibility and cannot be dynamically adjusted.
By combining intelligent outbound calling technology with big data analytics, and using feature engineering and the LightBGM algorithm to build a screening model, we can automatically filter target users and push products to them. By combining sentiment analysis and natural language processing, we can achieve end-to-end precision marketing.
It improved the accuracy and user-friendliness of product recommendations, reduced manual workload, and achieved automated customer screening and marketing loop.
Smart Images
Figure CN115422456B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence and can be applied to the financial and other fields, particularly to a product push method and device based on intelligent outbound calling technology and big data analysis. Background Technology
[0002] Currently, the promotion of financial products mainly relies on leveraging account managers' dedicated marketing sessions, combined with the advantages of human telephone marketing, to conduct one-on-one outbound marketing promotions. This method places a heavy burden on human resources, especially when the demand for the product in the target market is unclear and the target audience is uncertain. Large-scale human outbound calling will greatly reduce human efficiency.
[0003] At the same time, the user experience is also poor, which can easily cause unnecessary interference to users who are not interested. In order to improve the accuracy of recommendations, the industry mainly uses relationship graphs to improve the above-mentioned manual efficiency. However, conventional relationship graph recommendation methods are not flexible and cannot be dynamically adjusted after a single calculation, so the accuracy of subsequent recommendations cannot be effectively guaranteed.
[0004] Therefore, how to improve customer perception and identify customers with clear needs for targeted marketing is an urgent problem to be solved by front-line business staff. Summary of the Invention
[0005] The purpose of this application is to provide a product push method and device based on intelligent outbound calling technology and big data analysis, which combines big data analysis and intelligent outbound calling technology to achieve an end-to-end problem closed loop, forming a fully automated solution from target customer list screening to intelligent outbound calling marketing.
[0006] To achieve the above objectives, this application provides a product push method based on intelligent outbound calling technology and big data analysis. The method includes: acquiring users' product usage records and identity information; extracting corresponding feature data based on the product usage records and identity information through feature engineering; obtaining a first set of users through cluster analysis based on the feature data; constructing a filtering model for the corresponding first set by performing machine learning on the feature data of users in each first set using the LightBGM algorithm; obtaining target user information by analyzing the collected user feature data through the filtering model; and pushing products based on the target user information.
[0007] In the above-mentioned product push method based on intelligent outbound calling technology and big data analysis, optionally, pushing products based on the target user information further includes: obtaining the target user's emotional willingness based on the pushed product feedback; associating the emotional willingness with the target user information and providing it to a predetermined location.
[0008] In the above-mentioned product push method based on intelligent outbound calling technology and big data analysis, optionally, obtaining the target user's emotional willingness based on the pushed product feedback includes: obtaining the target user's interactive voice information based on the product push result; converting the interactive voice information into text data and then performing object separation to obtain the target user's communication text data; and identifying the target user's emotional willingness based on the communication text data using natural language analysis technology.
[0009] In the above-mentioned product push method based on intelligent outbound calling technology and big data analysis, optionally, the product usage record includes credit card installment records, credit card transaction records and financial behavior information within a preset period.
[0010] In the above-mentioned product push method based on intelligent outbound calling technology and big data analysis, optionally, obtaining target user information by analyzing the collected user feature data through the filtering model includes: obtaining the corresponding category based on the collected user feature data; determining the pre-stored filtering frequency based on the category of the feature data; and calling the filtering model to analyze the feature data to obtain target user information based on the filtering frequency.
[0011] In the above-mentioned product push method based on intelligent outbound calling technology and big data analysis, optionally, obtaining the first set of each user through cluster analysis based on the feature data includes: classifying and statistically analyzing the user information, account information and transaction information corresponding to the user according to the feature data; and dividing the user into multiple first sets with different priorities through cluster analysis based on the user information, the account information and the transaction information.
[0012] In the above-mentioned product push method based on intelligent outbound calling technology and big data analysis, optionally, the method further includes: collecting users' product usage records and identity information at a preset period; obtaining a second set of each user through cluster analysis based on the feature data corresponding to the product usage records and identity information; comparing the second set with the first set to obtain a difference ratio value; comparing the difference ratio value with a preset threshold; and adjusting the filtering model corresponding to each first set based on the comparison result.
[0013] This application also provides a product push device based on intelligent outbound calling technology and big data analysis. The device includes a feature extraction module, a training module, a filtering module, and a push module. The feature extraction module is used to acquire users' product usage records and identity information, and extract corresponding feature data through feature engineering based on the product usage records and identity information. The training module is used to obtain a first set of users through cluster analysis based on the feature data, and construct a filtering model corresponding to the first set by performing machine learning on the feature data of users in each first set using the LightBGM algorithm. The filtering module is used to obtain target user information by analyzing the collected user feature data through the filtering model. The push module is used to push products based on the target user information.
[0014] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method.
[0015] This application also provides a computer-readable storage medium storing a computer program that performs the above-described methods.
[0016] This application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.
[0017] The beneficial technical effects of this application are as follows: by combining and linking existing key technologies (big data analysis, intelligent outbound calling), an automated solution for the current installment product promotion and marketing can be formed, which can greatly alleviate the pressure of manual labor, while seizing the key marketing period and improving the accuracy of product promotion and user-friendliness. Attached Figure Description
[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, do not constitute a limitation thereof. In the drawings:
[0019] Figure 1 This is a flowchart illustrating a product push method based on intelligent outbound calling technology and big data analysis provided in an embodiment of this application.
[0020] Figure 2 This is a schematic diagram of a user feedback process provided in an embodiment of this application;
[0021] Figure 3 This is a schematic diagram illustrating the process of obtaining emotional intentions according to an embodiment of this application;
[0022] Figure 4This is a schematic diagram of the target user information filtering frequency adjustment process provided in an embodiment of the present application;
[0023] Figure 5 A schematic diagram illustrating the process of obtaining the first set according to an embodiment of this application;
[0024] Figure 6 This is a schematic diagram illustrating the process of adjusting the screening model provided in one embodiment of this application;
[0025] Figure 7 A schematic diagram of the structure of a product push device based on intelligent outbound calling technology and big data analysis provided in an embodiment of this application;
[0026] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation
[0027] The following will describe in detail the implementation methods of this application with reference to the accompanying drawings and embodiments, so as to fully understand how this application uses technical means to solve technical problems and achieve technical effects, and to implement it accordingly. It should be noted that, as long as there is no conflict, the various embodiments and features in each embodiment of this application can be combined with each other, and the resulting technical solutions are all within the protection scope of this application.
[0028] Furthermore, the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.
[0029] It should be noted that the customer information mining method and apparatus disclosed in this application can be used in the field of financial technology, or in any field other than financial technology. The application field of the customer information mining method and apparatus disclosed in this application is not limited.
[0030] Please refer to Figure 1 As shown, this application provides a product push method based on intelligent outbound calling technology and big data analysis, the method comprising:
[0031] S101 acquires the user's product usage records and identity information, and extracts corresponding feature data based on the product usage records and identity information through feature engineering;
[0032] S102 Based on the feature data, a first set of each user is obtained through cluster analysis. The LightBGM algorithm is used to perform machine learning on the feature data of users in each first set to construct a screening model for the corresponding first set.
[0033] S103 obtains target user information by analyzing the collected user feature data through the filtering model;
[0034] S104 Pushes products based on the target user information.
[0035] The product usage records include credit card installment records, credit card transaction records, and financial behavior information within a preset period. Specifically, in practice, data features can be accumulated through feature engineering using internal bank user data and big data analytics. Then, corresponding predictive models are built based on machine learning algorithms to analyze these data features and screen target user groups, thereby reducing the complexity of manual screening while improving accuracy.
[0036] The acquisition, storage, use, and processing of data in the technical solution of this application all comply with the relevant provisions of national laws and regulations.
[0037] Please refer to Figure 2 As shown, in one embodiment of this application, pushing products based on the target user information further includes:
[0038] S201 Obtains the emotional intentions of target users based on the product feedback pushed to them;
[0039] S202 After associating the emotional intention with the target user information, it is provided to a predetermined location.
[0040] Please refer to the details. Figure 3 As shown, obtaining the target user's emotional willingness based on the product feedback pushed to them may include:
[0041] S301 obtains the target user's interactive voice information based on the product push results;
[0042] S302 After converting the interactive voice information into text data, object separation is performed to obtain the target user's communication text data;
[0043] S303 uses natural language processing technology to identify the target user's emotional intentions based on the communication text data.
[0044] In practical applications, the interactive voice information may include interaction rounds, interaction time, interaction end points, and voice files. The voice files are used to determine their semantics using natural language processing methods, and subsequently to assess the user's emotions. Interaction rounds, interaction time, and interaction end points are all used to assist in assessing the user's mood. Specific usage can be selected according to actual needs, and this application does not impose further limitations. The overall intelligent outbound calling process can be referenced as follows:
[0045] Customer call information acquisition: Make outbound calls to recommend installment products and collect information on interactions with customers, including interaction rounds, interaction time, interaction end points, and voice files.
[0046] Speech Analysis and Processing: Automatic Speech Recognition (ASR) technology is used to process the audio files from the previous steps into text, separating the agent's statements from the customer's statements. Then, Natural Language Processing (NLP) technology and a pre-set analysis model are used to determine the customer's level of willingness to purchase installment products. This model includes customer sentiment indicators and whether the response is affirmative, negative, or uncertain.
[0047] Outbound call results display: The above outbound call list and corresponding willingness levels are marked and displayed to business personnel through the system to facilitate their next steps.
[0048] The rules for determining the relationships between nodes are shown in Table 1 below:
[0049] Table 1
[0050]
[0051]
[0052] Please refer to Figure 4 As shown, in one embodiment of this application, obtaining target user information by analyzing the collected user feature data through the filtering model includes:
[0053] S401 obtains the corresponding category based on the collected user feature data;
[0054] S402 determines the pre-stored filtering frequency based on the category of the feature data;
[0055] S403 calls the filtering model to analyze the feature data and obtain target user information based on the filtering frequency.
[0056] In practical work, the corresponding category can be obtained based on the feature data using either numerical comparison or tabular comparison. Taking tabular comparison as an example, please refer to Table 2 below:
[0057] Table 2
[0058]
[0059]
[0060] Therefore, based on the frequency corresponding to these indicators, it can be determined which category and frequency the newly collected feature data belongs to. For example, if user A's credit card balance changes on a certain day, this feature data belongs to the daily change frequency. Therefore, at the end of the day, this feature data can be analyzed to confirm whether the user meets the target user criteria. The changes in information corresponding to static basic indicators and static bank indicators belong to long-term stable data. This data does not change easily, so a longer screening period can be set for analysis, such as screening analysis once every three months. Of course, those skilled in the art can also choose the corresponding screening frequency according to the actual situation, so as to reduce unnecessary consumption of computing resources while ensuring the timeliness of information.
[0061] Please refer to Figure 5 As shown, in one embodiment of this application, the first set of users obtained through cluster analysis based on the feature data may include:
[0062] S501 classifies and statistically analyzes the user information, account information, and transaction information corresponding to the user based on the feature data;
[0063] S502 uses cluster analysis to divide users into multiple first sets with different priorities based on the user information, account information, and transaction information.
[0064] Specifically, in actual work, different users have different preferences. Therefore, in the above embodiment, users are divided into multiple clusters according to user value or consumption preferences by defining rules based on relevant user information such as user information, account information and transaction information. Each cluster corresponds to a different priority. Then, LightBGM modeling is performed for each cluster, and the corresponding target users are effectively filtered through the filtering model generated by the modeling.
[0065] Please refer to Figure 6 As shown, in one embodiment of this application, the method further includes:
[0066] S601 collects users' product usage records and identity information according to a preset cycle, and obtains a second set of users through cluster analysis based on the feature data corresponding to the product usage records and identity information.
[0067] S602 compares the second set with the first set to obtain a difference ratio value, and compares the difference ratio value with a preset threshold;
[0068] S603 adjusts the screening model corresponding to each first set based on the comparison results.
[0069] Specifically, in practical work, the purpose of this embodiment is to update the screening model according to a preset period to improve the accuracy of each model and achieve the purpose of dynamic adjustment. In this overload, the comparison of the first set and the second set can be done by similarity comparison method or association comparison method, etc., and this application does not further limit it here. The method of generating the second set can refer to the method of generating the first set. Both are completed by using feature extraction combined with aggregation classification, which will not be described in detail here.
[0070] Please refer to Figure 7 As shown, this application also provides a product push device based on intelligent outbound calling technology and big data analysis. The device includes a feature extraction module, a training module, a filtering module, and a push module. The feature extraction module is used to acquire users' product usage records and identity information, and extract corresponding feature data through feature engineering based on the product usage records and identity information. The training module is used to obtain a first set of users through cluster analysis based on the feature data, and construct a filtering model corresponding to the first set by performing machine learning on the feature data of users in each first set using the LightBGM algorithm. The filtering module is used to analyze the collected user feature data through the filtering model to obtain target user information. The push module is used to push products based on the target user information. In actual work, the specific implementation logic of each component has been described in the foregoing embodiments, and will not be detailed here.
[0071] The beneficial technical effects of this application are as follows: by combining and linking existing key technologies (big data analysis, intelligent outbound calling), an automated solution for the current installment product promotion and marketing can be formed, which can greatly alleviate the pressure of manual labor, while seizing the key marketing period and improving the accuracy of product promotion and user-friendliness.
[0072] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method.
[0073] This application also provides a computer-readable storage medium storing a computer program that performs the above-described methods.
[0074] This application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.
[0075] like Figure 8 As shown, the electronic device 600 may also include: a communication module 110, an input unit 120, an audio processing unit 130, a display 160, and a power supply 170. It is worth noting that the electronic device 600 does not necessarily need to include these components. Figure 8 All components shown; in addition, the electronic device 600 may also include Figure 8 For components not shown, please refer to existing technologies.
[0076] like Figure 8 As shown, the central processing unit 100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device. The central processing unit 100 receives inputs and controls the operation of various components of the electronic device 600.
[0077] The memory 140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 100 may execute the program stored in the memory 140 to perform information storage or processing, etc.
[0078] Input unit 120 provides input to central processing unit 100. Input unit 120 may be, for example, a keypad or touch input device. Power supply 170 provides power to electronic device 600. Display 160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.
[0079] The memory 140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 140 can also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application / function storage unit 142 for storing application programs and function programs or processes for executing the operation of the electronic device 600 via the central processing unit 100.
[0080] The memory 140 may also include a data storage unit 143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 144 of the memory 140 may include various drivers for the electronic device's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).
[0081] The communication module 110 is a transmitter / receiver 110 that transmits and receives signals via antenna 111. The communication module (transmitter / receiver) 110 is coupled to the central processing unit 100 to provide input signals and receive output signals, which can be the same as in a conventional mobile communication terminal.
[0082] Based on different communication technologies, multiple communication modules 110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module (transmitter / receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132, thereby enabling typical telecommunications functions. The audio processor 130 may include any suitable buffer, decoder, amplifier, etc. Additionally, the audio processor 130 is coupled to a central processing unit 100, enabling on-device recording via the microphone 132 and on-device playback of stored audio via the speaker 131.
[0083] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0084] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0085] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.
[0086] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0087] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.
Claims
1. A product pushing method based on intelligent outbound technology and big data analysis, characterized in that, The method includes: Obtain the user's product usage records and identity information, and extract corresponding feature data based on the product usage records and identity information through feature engineering; Based on the feature data, a first set of each user is obtained through cluster analysis. The LightGBM algorithm is then used to perform machine learning on the feature data of users in each first set to construct a screening model for the corresponding first set. The target user information is obtained by analyzing the collected user feature data through the filtering model. Product recommendations are made based on the target user information. The process of obtaining target user information by analyzing the collected user feature data through the filtering model includes: The corresponding category is obtained based on the collected user feature data; The pre-stored filtering frequency is determined based on the category of the feature data; The filtering model is invoked according to the filtering frequency to analyze the feature data and obtain target user information; wherein, the method for obtaining the corresponding category according to the feature data is either numerical comparison method or tabular comparison method.
2. The product push method based on intelligent outbound calling technology and big data analysis according to claim 1, characterized in that, Product push based on the target user information also includes: Obtain the emotional intentions of target users based on the product feedback pushed to them; The emotional intentions and the target user information are associated and then provided to a predetermined location.
3. The product push method based on intelligent outbound calling technology and big data analysis according to claim 2, characterized in that, Obtaining the emotional intentions of target users based on the product feedback pushed to them includes: Obtain interactive voice information from target users based on product push results; After converting the interactive voice information into text data, object separation is performed to obtain the target user's communication text data. The target user's emotional intentions are identified using natural language processing techniques based on the communication text data.
4. The product push method based on intelligent outbound calling technology and big data analysis according to claim 1, characterized in that, The product usage records include credit card installment records, credit card transaction records, and financial behavior information within a preset period.
5. The product push method based on intelligent outbound calling technology and big data analysis according to claim 1, characterized in that, Based on the aforementioned feature data, cluster analysis is used to obtain a first set of users, which includes: Based on the aforementioned feature data, user information, account information, and transaction information corresponding to each user are categorized and statistically analyzed. Based on the user information, account information, and transaction information, users are divided into multiple first sets with different priorities through cluster analysis.
6. The product push method based on intelligent outbound calling technology and big data analysis according to claim 1, characterized in that, The method further includes: Collect users’ product usage records and identity information at a preset period, and obtain a second set of each user through cluster analysis based on the feature data corresponding to the product usage records and identity information. The difference ratio is obtained by comparing the second set with the first set, and the difference ratio is compared with a preset threshold. The screening model corresponding to each first set is adjusted based on the comparison results.
7. A product push device based on intelligent outbound calling technology and big data analysis, characterized in that, The device includes a feature extraction module, a training module, a filtering module, and a push module; The feature extraction module is used to obtain the user's product usage records and identity information, and extract corresponding feature data through feature engineering based on the product usage records and identity information. The training module is used to obtain a first set of users based on the feature data through cluster analysis, and to construct a screening model for the corresponding first set by performing machine learning on the feature data of users in each first set using the LightGBM algorithm. The filtering module is used to analyze the collected user feature data through the filtering model to obtain target user information; The push module is used to push products based on the target user information; The screening model is specifically used for: The corresponding category is obtained based on the collected user feature data; The pre-stored filtering frequency is determined based on the category of the feature data; The filtering model is invoked according to the filtering frequency to analyze the feature data and obtain target user information; wherein, the method for obtaining the corresponding category according to the feature data is either numerical comparison method or tabular comparison method.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that enables a computer to execute the method of any one of claims 1 to 6.
10. A computer program product comprising computer program instructions, characterized in that, When the computer program instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 6.