Service recommendation method, device and equipment and storage medium thereof

Abstract

The embodiment of the invention belongs to the technical field of artificial intelligence, is applied to a service management and control recommendation image scene, and relates to a service recommendation method, device and equipment and a storage medium thereof. According to the service recommendation method and device, service recommendation is carried out on the target customer through the service seat according to the reaching strategy, it is guaranteed that the seat strictly follows the reaching strategy when carrying out service recommendation on the customer, excessive marketing on the customer can be avoided in the preset time period, the customer is prevented from generating the dislike emotion of being disturbed by business marketing as much as possible, and the customer stickiness is kept.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and is applied to service management and recommendation image scenarios, and in particular to a service recommendation method, device, equipment and storage medium thereof. Background Art

[0002] In the current market environment, companies are facing severe competition challenges and evolving customer needs, and need to adopt more sophisticated customer management strategies to meet increasingly personalized and complex customer needs. Among them, communicating with customers and providing services through social media platforms has become an emerging and effective means. The convenience, efficiency and broad user base of social media platforms enable companies to establish more direct and personalized connections with customers and provide customized services and products.

[0003] However, with the popularity of this new communication method, business companies need to meet customer needs in terms of business services, and must also face the problem of how to prevent excessive marketing from causing customer disgust. Therefore, there is an urgent need for a service recommendation method that allows agents to avoid excessive marketing to customers within a preset time period when serving customers, prevent customers from having negative emotions caused by business marketing harassment, and maintain customer stickiness. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose a service recommendation method, device, equipment and storage medium thereof, so that when the agent provides service to the customer, it can avoid excessive marketing to the customer within a preset time period, prevent the customer from having negative emotions caused by business marketing harassment, and maintain customer stickiness.

[0005] In order to solve the above technical problems, the present application embodiment provides a service recommendation method, which adopts the following technical solutions:

[0006] A service recommendation method comprises the following steps:

[0007] Obtain target customer profile data;

[0008] Input the target customer portrait data into the pre-trained customer agent matching model to screen out the target service agent;

[0009] Input the target customer portrait data into the pre-trained reach classification model to identify the reach category corresponding to the target customer, wherein each reach category corresponds to a reach strategy;

[0010] Determine the contact strategy corresponding to the target customer based on the contact category, wherein the contact strategy includes the maximum number of communications within a preset time period, optimal communication time period information, optimal communication duration for each communication, and preset communication interval time information;

[0011] Recommending services to the target customer through the target service agent and according to the contact strategy;

[0012] Acquire communication data each time a service recommendation is made to the target customer, and input the communication data into a focus recognition model to obtain focus information that the target customer is concerned about during this service recommendation;

[0013] The script for the next service recommendation is adjusted according to the focus information, and the next service recommendation is made to the target customer according to the script.

[0014] Furthermore, before executing the step of inputting the target customer portrait data into the pre-trained customer agent matching model to screen out the target service agent, the method further includes:

[0015] Obtaining business profile data of all agents, wherein the business profile data includes profile data of customers who have successfully converted after the agents made business recommendations;

[0016] Input the business portrait data of all the seats into the customer-seat matching model to be trained, perform learning training, and learn the customer group portrait data corresponding to all the seats;

[0017] Deploy the correlation relationship between the customer group portrait data corresponding to all seats as reference knowledge to the customer-seat matching model to be trained, so as to obtain a pre-trained customer-seat matching model;

[0018] The step of inputting the target customer portrait data into the pre-trained customer agent matching model to screen out the target service agent specifically includes:

[0019] Step 401, calculating the similarity between the target customer portrait data and all customer group portrait data according to a preset similarity algorithm;

[0020] Step 402, sorting the customer groups to which the target customer portrait data belongs in descending order based on the similarity to obtain a descending sequence;

[0021] Step 403, generating an agent priority sequence for the target customer according to the association relationship between the descending sequence and the customer group portrait data corresponding to all the agents respectively;

[0022] Step 404: Filter out target service agents according to the agent priority sequence and the real-time working status of all agents, wherein the real-time working status includes being on duty, being idle at work, and the predicted end time of being on duty.

[0023] Furthermore, the step of screening out target service seats according to the seat priority sequence and the real-time working status of all seats specifically includes:

[0024] Identify the real-time working status of all agents;

[0025] If all the agents are working, the target customer's waiting time is determined according to the target customer portrait data, and the waiting time is compared with the predicted end time when all the agents are working;

[0026] If the predicted end time of all the agents in work is greater than the waiting time, step 404 is executed again;

[0027] If the predicted end time of the agent's work is not greater than the waiting time, step 404 is executed again, and the communication information of the target customer is added to the preset call waiting sequence pool;

[0028] If there is an agent who is idle at work and the agent who is idle at work is the only one, the agent is used as the target service agent;

[0029] If there is an agent who is not working at a unique time, the agent who is not working at a unique time is selected as the target service agent according to the agent priority sequence.

[0030] Furthermore, before executing the step of inputting the target customer portrait data into the pre-trained reach classification model to identify the reach category corresponding to the target customer, the method further includes:

[0031] Obtain the contact categories of all historical customers and the individual portrait data of all historical customers;

[0032] Perform clustering processing on the individual portrait data of historical customers of the same reach category to obtain clustering processing results;

[0033] According to the clustering processing results, group customer portrait data corresponding to all contact categories are sorted out;

[0034] Deploy the group customer portrait data corresponding to all reach categories as reach classification knowledge into a preset classification model to obtain the pre-trained reach classification model;

[0035] The step of inputting the target customer portrait data into the pre-trained reach classification model to identify the reach category corresponding to the target customer specifically includes:

[0036] Filter out the group customer portrait data that is most similar to the target customer portrait data according to a preset similarity algorithm;

[0037] The reach category corresponding to the group customer portrait data with the highest similarity is used as the reach category corresponding to the target customer.

[0038] Furthermore, the step of recommending services to the target customer through the target service agent according to the contact strategy specifically includes:

[0039] Obtain the number of times the target customer has communicated within a preset time period, as well as the time information of the last communication;

[0040] According to the contact strategy, the maximum number of communications, the best communication time period, the best communication duration for each communication, and the preset communication interval time information corresponding to the target customer within a preset time period are identified;

[0041] Determining whether the number of communications has reached the maximum number of communications;

[0042] If the number of communications has reached the maximum number of communications, then stop recommending services to the target customer;

[0043] If the number of communications has not reached the maximum number of communications, then calculating the interval time information between the current time and the time information of the last communication;

[0044] If the interval time information does not reach the preset communication interval time information, a prompt for delaying service recommendation is sent to the target agent;

[0045] If the interval time information has reached the preset communication interval time information, the target agent is prompted to make service recommendations to the target customer based on the optimal communication time period information and the optimal communication duration during this communication.

[0046] Furthermore, after executing the step of recommending services to the target customer through the target service agent according to the contact strategy, the method further includes:

[0047] Obtaining service feedback information sent by the target customer;

[0048] Identifying the recommendation effect of the service recommendation performed this time according to the service feedback information, wherein the recommendation effect includes three states: positive effect, negative effect and no effect;

[0049] If the recommendation effect is a positive effect, the contact strategy is updated by shortening the preset communication interval time information;

[0050] If the recommendation effect is a reverse effect, a service recommendation is made to the target customer by extending the preset communication interval time information or / and selecting a seat with a higher priority from the seat priority sequence;

[0051] If the recommendation effect is ineffective, a seat with a higher priority is selected from the seat priority sequence to continue to make the next service recommendation to the target customer according to the contact strategy.

[0052] Furthermore, the step of inputting the communication data into a focus recognition model to obtain the focus information that the target customer is concerned about when recommending this service specifically includes:

[0053] Textual processing of the communication data;

[0054] Perform semantic analysis and word segmentation on the text content after text processing;

[0055] The occurrence frequency of each business noun in the word segmentation processing result is counted through a preset business dictionary, and the business verb corresponding to each business noun is identified through the semantic analysis result;

[0056] According to the occurrence frequency of each business noun and the business verb corresponding to each business noun, the focus information that the target customer pays attention to during the service recommendation is determined.

[0057] In order to solve the above technical problems, the embodiment of the present application also provides a service recommendation device, which adopts the following technical solution:

[0058] A service recommendation device, comprising:

[0059] A target customer portrait data acquisition module is used to acquire target customer portrait data;

[0060] A target service agent screening module is used to input the target customer portrait data into the pre-trained customer agent matching model to screen out the target service agent;

[0061] A reach category identification module is used to input the target customer portrait data into a pre-trained reach classification model to identify the reach category corresponding to the target customer, wherein each reach category corresponds to a reach strategy;

[0062] A contact strategy determination module, used to determine the contact strategy corresponding to the target customer based on the contact category, wherein the contact strategy includes the maximum number of communications within a preset time period, the best communication time period information, the best communication duration for each communication, and the preset communication interval time information;

[0063] A service reaching strategy recommendation module, used to recommend services to the target customer through the target service agent and according to the reaching strategy;

[0064] A focus information acquisition module is used to obtain the communication data when recommending services to the target customer each time, and input the communication data into the focus recognition model to obtain the focus information that the target customer pays attention to when recommending services this time;

[0065] The service recommendation adjustment module is used to adjust the text of the next service recommendation according to the focus information, and make the next service recommendation to the target customer according to the text of the text.

[0066] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0067] A computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the above-mentioned service recommendation method when executing the computer-readable instructions.

[0068] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0069] A computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the service recommendation method as described above.

[0070] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0071] The service recommendation method described in the embodiment of the present application adopts artificial intelligence and big data matching to intelligently and automatically screen out service agents that are more suitable for customers. After that, the service agents recommend services to target customers based on the reach strategies, ensuring that the agents strictly follow the reach strategies when recommending services to customers, avoiding excessive marketing to customers within a preset time period, and minimizing the negative emotions of customers being harassed by business marketing, thereby maintaining customer stickiness. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0073] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0074] Figure 2 is a flow chart of an embodiment of a service recommendation method according to the present application;

[0075] Figure 3 yes Figure 2 A flowchart of a specific embodiment of learning and training the customer-seat matching model before step 202 is shown;

[0076] Figure 4 yes Figure 2 A flowchart of a specific embodiment of step 202 is shown;

[0077] Figure 5 yes Figure 4 A flowchart of a specific embodiment of step 404 is shown;

[0078] Figure 6 yes Figure 2 A flowchart of a specific embodiment of learning and training the reach classification model before step 203 is shown;

[0079] Figure 7 yes Figure 2 A flowchart of a specific embodiment of step 205 is shown;

[0080] Figure 8 yes Figure 2 A flowchart of a specific embodiment of obtaining focus information in step 206 is shown;

[0081] Fig. 9 is a structural diagram of an embodiment of a service recommendation device according to the present application;

[0082] Fig.10 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0083] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0084] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0085] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0086] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0087] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0088] The terminal device 101 can be various electronic devices with a display screen and supporting web browsing. In addition to the laptop computer 1011, the tablet computer 1012 or the mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop computer and a desktop computer, etc.

[0089] The server 103 may be a server that provides various services, such as a background server that provides support for a web page displayed on the terminal device 101 .

[0090] It should be noted that the service recommendation method provided in the embodiment of the present application is generally executed by a terminal device, and accordingly, the service recommendation device is generally arranged in the terminal device.

[0091] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0092] Continue to refer Figure 2 , shows a flow chart of an embodiment of a service recommendation method according to the present application. The service recommendation method comprises the following steps:

[0093] Step 201, obtaining target customer portrait data;

[0094] In this embodiment, the target customer portrait data can be acquired in two different ways: active acquisition or passive acquisition. It can be understood that the active acquisition is that the acquirer actively sends an acquisition instruction to acquire the target customer portrait data fed back by the data provider, while the passive acquisition is that the acquirer passively receives the target customer portrait data fed back by the data provider while maintaining the transmission connection under the premise that the acquirer does not actively initiate the acquisition instruction.

[0095] Specifically, the customer portrait data refers to data that reflects multi-dimensional comprehensive information about customers, such as basic information about customers, purchase history, communication records, feedback information, APP activity, behavioral preferences, etc. By obtaining target customer portrait data, we can interpret the customer's characterization information in an all-round way, so as to better recommend services to customers.

[0096] Step 202: input the target customer portrait data into the pre-trained customer-seat matching model to screen out the target service agent;

[0097] In this embodiment, the target customer portrait data is input into the pre-trained customer agent matching model. The input can be understood as the target customer portrait data acquired by the previous acquisition component being transmitted to the pre-trained customer agent matching model in the form of data transmission.

[0098] Specifically, by using the pre-trained customer-agent matching model, target service agents are screened out, and artificial intelligence and big data matching methods are used to intelligently and automatically screen out service agents that are more suitable for customers, thereby reducing subjective errors in manual screening based on experience. At the same time, the machine screening method also saves screening time and reduces the consumption of manpower and material resources during manual screening.

[0099] Step 203: input the target customer portrait data into the pre-trained reach classification model to identify the reach category corresponding to the target customer, wherein each reach category corresponds to a reach strategy;

[0100] In this embodiment, the "reach" mentioned above refers to a way for agents to establish an active access relationship with customers, for example, agents reach customers by outbound calls, agents reach customers by chatting, and agents reach customers by push messaging. There is no limitation on the reach method here, and the purpose of "reach" is to establish a service marketing relationship with customers. The reach category mentioned above refers to a one-to-one category set according to different reach strategies, not the above-mentioned communication method category.

[0101] Specifically, by identifying the reach category corresponding to the target customer in the pre-trained reach classification model, it can help determine the reach strategy corresponding to the target customer based on a one-to-one matching relationship in the post-processing step, thereby improving computer execution efficiency.

[0102] Step 204, determining a contact strategy corresponding to the target customer based on the contact category, wherein the contact strategy includes a maximum number of communications within a preset time period, optimal communication time period information, optimal communication duration for each communication, and preset communication interval time information;

[0103] Specifically, by pre-setting the reach strategy, that is, setting the maximum number of communications within the preset time period, the best communication time period information, the best communication duration for each communication, and the preset communication interval time information, it is possible to avoid excessive marketing to customers within the preset time period, prevent customers from having negative emotions caused by business marketing harassment, and maintain customer stickiness.

[0104] Step 205: recommending services to the target customer through the target service agent and according to the contact strategy;

[0105] Specifically, by making service recommendations to the target customers through the target service agent and based on the contact strategy, it is ensured that the agent strictly follows the contact strategy when making service recommendations to customers, and can avoid excessive marketing to customers within a preset time period, try to prevent customers from having negative emotions caused by business marketing harassment, and maintain customer stickiness.

[0106] Step 206, obtaining communication data each time a service recommendation is made to the target customer, and inputting the communication data into a focus recognition model to obtain focus information that the target customer is concerned about during this service recommendation;

[0107] In this embodiment, the communication data obtained each time a service recommendation is made to the target customer includes call voice data or chat record data. The method of obtaining the communication data varies greatly depending on the communication method. For example, when the communication data is call voice data, the method of obtaining the communication data may be monitoring or real-time voice recording. Correspondingly, when the communication data is chat record data, the method of obtaining the communication data may be chat log record file obtaining method.

[0108] Specifically, the communication data is input into a focus recognition model to obtain the focus information that the target customer is concerned about when recommending the service this time. This can improve the next service recommendation based on the focus information that the target customer is concerned about when recommending the service this time, and provide better recommended services to customers.

[0109] Step 207, adjusting the next service recommendation speech text according to the focus information, and making the next service recommendation to the target customer according to the speech text.

[0110] In this embodiment, by adopting artificial intelligence and big data matching methods, service agents who are more suitable for customers are intelligently and automatically screened out. After that, the service agents recommend services to target customers based on the contact strategy, ensuring that the agents strictly follow the contact strategy when recommending services to customers, avoiding excessive marketing to customers within a preset time period, and preventing customers from having negative emotions caused by business marketing harassment as much as possible, thereby maintaining customer stickiness.

[0111] Continue to refer Figure 3 In some optional implementations, before step 202, a step of learning and training the customer-seat matching model is also included. Figure 3 yes Figure 2The flowchart of a specific embodiment of learning and training the customer-seat matching model before step 202 includes the following steps:

[0112] Step 301, obtaining business portrait data of all agents, wherein the business portrait data includes portrait data of customers who are successfully converted after the agents make business recommendations;

[0113] Step 302: input the business portrait data of all the seats into the customer-seat matching model to be trained, perform learning training, and learn the customer group portrait data corresponding to all the seats;

[0114] Specifically, the customer group portrait data corresponding to all seats are learned, and the specific implementation method is to screen the common customer portrait data for all customer portrait data that have been successfully converted after business recommendations by all seats, and use the common customer portrait data corresponding to different seats as the customer group portrait data corresponding to the corresponding seats.

[0115] Step 303: deploy the association relationship between the customer group portrait data corresponding to all seats as reference knowledge into the customer-seat matching model to be trained, so as to obtain a pre-trained customer-seat matching model.

[0116] A customer-seat matching model is trained through the business portrait data of all agents, so that service agents can be screened using artificial intelligence methods combined with automated models, making the processing method more intelligent and automated.

[0117] Continue to refer Figure 4 , Figure 4 yes Figure 2 The flowchart of a specific embodiment of step 202 shown includes the following steps:

[0118] Step 401, calculating the similarity between the target customer portrait data and all customer group portrait data according to a preset similarity algorithm;

[0119] Specifically, the preset similarity algorithm, for example, a cosine similarity algorithm, an edit distance algorithm, etc., can be selected according to the calculation or encoding method.

[0120] Step 402, sorting the customer groups to which the target customer portrait data belongs in descending order based on the similarity to obtain a descending sequence;

[0121] Step 403, generating an agent priority sequence for the target customer according to the association relationship between the descending sequence and the customer group portrait data corresponding to all the agents respectively;

[0122] Step 404: Filter out target service agents according to the agent priority sequence and the real-time working status of all agents, wherein the real-time working status includes being on duty, being idle at work, and the predicted end time of being on duty.

[0123] By generating an agent priority sequence for the target customer, a better service agent can be selected for the customer in combination with the agent priority.

[0124] Continue to refer Figure 5 , Figure 5 yes Figure 4 The flowchart of a specific embodiment of step 404 shown includes the following steps:

[0125] Step 501, identifying the real-time working status of all agents;

[0126] Step 502: If all the agents are working, determine the target customer's waiting time according to the target customer portrait data, and compare the waiting time with the predicted end time of all the agents working;

[0127] Step 503: if the predicted end time of all the agents in work is greater than the waiting time, re-execute step 404;

[0128] Step 504: if the predicted end time of the agent in the work is not greater than the waiting time, re-execute step 404 and add the communication information of the target customer to the preset call waiting sequence pool;

[0129] Step 505: If there is an agent who is in idle time and the agent who is in idle time is unique, the agent is used as the target service agent;

[0130] Step 506: If there is an agent who is not working at the same time and the agent who is not working at the same time is not unique, then select the agent who is working at the same time and has the highest priority as the target service agent according to the agent priority sequence.

[0131] By identifying the real-time working status of all agents, we can screen out the best service agents who are not working properly based on the agent priority, thereby improving customers' service experience.

[0132] Continue to refer Figure 6 In some optional implementations, before step 203, a step of learning and training the reach classification model is also included. Figure 6 yes Figure 2 The flowchart of a specific embodiment of learning and training the reach classification model before step 203 shown includes the following steps:

[0133] Step 601, obtaining the contact categories of all historical customers and the individual portrait data of all historical customers;

[0134] Step 602, clustering the individual portrait data of historical customers of the same reach category to obtain a clustering result;

[0135] Step 603, sorting out group customer portrait data corresponding to all reach categories according to the clustering processing results;

[0136] Step 604, deploying the group customer portrait data corresponding to all reach categories into a preset classification model as reach classification knowledge to obtain the pre-trained reach classification model.

[0137] Specifically, the group customer portrait data is clustered and sorted according to all contact categories, and there is a big difference in the classification starting point between the group customer portrait data sorted according to the business portrait data of all seats in step 302. Therefore, the sorted group portrait data also have a big difference.

[0138] By training the reach classification model, it is possible to subsequently use artificial intelligence methods and combine them with automated models to perform reach category screening, making the screening method more intelligent and automated.

[0139] In this embodiment, the step 203, inputting the target customer portrait data into the pre-trained reach classification model, and identifying the reach category corresponding to the target customer, specifically includes: screening out the group customer portrait data with the highest similarity to the target customer portrait data according to a preset similarity algorithm; and taking the reach category corresponding to the group customer portrait data with the highest similarity as the reach category corresponding to the target customer.

[0140] Specifically, the preset similarity algorithm preferentially adopts the cosine similarity algorithm, and screens out the group customer portrait data with the highest similarity to the target customer portrait data according to the cosine similarity algorithm; and uses the reach category corresponding to the group customer portrait data with the highest similarity as the reach category corresponding to the target customer.

[0141] Continue to refer Figure 7 , Figure 7 yes Figure 2 The flowchart of a specific embodiment of step 205 shown includes the following steps:

[0142] Step 701, obtaining the number of times the target customer has communicated within a preset time period, and the time information of the last communication;

[0143] Step 702, identifying the maximum number of communications, optimal communication time period information, optimal communication duration for each communication, and preset communication interval time information corresponding to the target customer within a preset time period according to the contact strategy;

[0144] Step 703, determining whether the number of communications has reached the maximum number of communications;

[0145] Step 704, if the number of communications has reached the maximum number of communications, then stop recommending services to the target customer;

[0146] Step 705, if the number of communications has not reached the maximum number of communications, then calculate the interval time information between the current time and the time information of the last communication;

[0147] Step 706: If the interval time information does not reach the preset communication interval time information, a prompt of delaying service recommendation is sent to the target agent;

[0148] Step 706: If the interval time information has reached the preset communication interval time information, the target agent is prompted to make service recommendations to the target customer based on the optimal communication time period information and the optimal communication duration of this communication.

[0149] By making service recommendations to the target customers through the target service agent and based on the contact strategy, it is ensured that the agent strictly follows the contact strategy when recommending services to customers, and can avoid excessive marketing to customers within a preset time period, try to prevent customers from having negative emotions caused by business marketing harassment, and maintain customer stickiness.

[0150] In some optional implementations, after the step of recommending services to the target customer through the target service agent and according to the contact strategy in step 205, the following steps are also included: obtaining service feedback information sent by the target customer; identifying the recommendation effect of this service recommendation based on the service feedback information, wherein the recommendation effect includes three states: positive effect, negative effect and no effect; if the recommendation effect is a positive effect, the contact strategy is updated by shortening the preset communication interval time information; if the recommendation effect is a negative effect, the service recommendation is made to the target customer by extending the preset communication interval time information or / and selecting an agent with a higher priority from the agent priority sequence; if the recommendation effect is no effect, a seat with a higher priority is selected from the agent priority sequence to continue to make the next service recommendation to the target customer according to the contact strategy.

[0151] By obtaining the service feedback information sent by the target customers and updating and adjusting the contact strategy according to the service feedback information, it is ensured that when the agent makes service recommendations to customers, the contact strategy is adjusted in time according to customer feedback, which can enhance the customer's business experience, fully respect the customer's wishes, and maintain customer stickiness to the greatest extent.

[0152] Continue to refer Figure 8 , Figure 8 yes Figure 2 The flowchart of a specific embodiment of obtaining the focus information in step 206 includes the following steps:

[0153] Step 801, text processing the communication data;

[0154] The text processing, for example, if the communication data is call voice data, the communication data is processed into text by voice-to-text conversion; if the communication data is chat record data, if it is a voice chat, the communication data is processed into text by voice-to-text conversion; if it is a text chat, the chat text is standardized according to a preset text template to process the communication data into text. There is no limitation here, it depends on the actual situation.

[0155] Step 802, performing semantic analysis and word segmentation on the text content after text processing;

[0156] Step 803, counting the occurrence frequency of each business noun in the word segmentation processing result through a preset business dictionary, and identifying the business verb corresponding to each business noun through the semantic analysis result;

[0157] Step 804: Determine the focus information that the target customer is concerned about during this service recommendation based on the occurrence frequency of each business noun and the business verb corresponding to each business noun.

[0158] Specifically, the focus information of the target customers during the service recommendation is determined based on the occurrence frequency of each business noun and the business verbs corresponding to each business noun. This can not only identify the business names that the customers are concerned about, but also analyze the customers' inclination towards the business with a high probability based on the business verbs, thereby facilitating the adjustment of the service recommendation strategy according to customer needs.

[0159] This application uses artificial intelligence and big data matching methods to intelligently and automatically screen out service agents that are more suitable for customers. After that, the service agents recommend services to target customers based on the reach strategies, ensuring that the agents strictly follow the reach strategies when recommending services to customers, avoiding excessive marketing to customers within a preset time period, and minimizing the risk of customers developing negative emotions caused by business marketing harassment, thereby maintaining customer stickiness.

[0160] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0161] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0162] In the embodiment of the present application, by adopting artificial intelligence and big data matching methods, service agents who are more suitable for customers are intelligently and automatically screened out. After that, the service agents recommend services to target customers based on the reach strategies, ensuring that the agents strictly follow the reach strategies when recommending services to customers, avoiding excessive marketing to customers within a preset time period, and preventing customers from having negative emotions caused by being harassed by business marketing, thereby maintaining customer stickiness.

[0163] Further references Fig. 9 , as a response to the above Figure 2 The present application provides an embodiment of a service recommendation device, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0164] like Fig. 9 As shown, the service recommendation device 900 described in this embodiment includes: a target customer portrait data acquisition module 901, a target service agent screening module 902, a reach category identification module 903, a reach strategy determination module 904, a service reach strategy recommendation module 905, a focus information acquisition module 906 and a service recommendation adjustment module 907. Among them:

[0165] The target customer portrait data acquisition module 901 is used to acquire the target customer portrait data;

[0166] The target service agent screening module 902 is used to input the target customer portrait data into the pre-trained customer agent matching model to screen out the target service agent;

[0167] The reach category identification module 903 is used to input the target customer portrait data into the pre-trained reach classification model to identify the reach category corresponding to the target customer, wherein each reach category corresponds to a reach strategy;

[0168] A contact strategy determination module 904 is used to determine the contact strategy corresponding to the target customer based on the contact category, wherein the contact strategy includes the maximum number of communications within a preset time period, the best communication time period information, the best communication duration for each communication, and the preset communication interval time information;

[0169] A service reaching strategy recommendation module 905 is used to recommend services to the target customer through the target service agent and according to the reaching strategy;

[0170] The focus information acquisition module 906 is used to obtain the communication data when recommending services to the target customer each time, and input the communication data into the focus recognition model to obtain the focus information that the target customer pays attention to during the current service recommendation;

[0171] The service recommendation adjustment module 907 is used to adjust the text of the next service recommendation according to the focus information, and make the next service recommendation to the target customer according to the text of the text.

[0172] This application uses artificial intelligence and big data matching methods to intelligently and automatically screen out service agents that are more suitable for customers. After that, the service agents recommend services to target customers based on the reach strategies, ensuring that the agents strictly follow the reach strategies when recommending services to customers, avoiding excessive marketing to customers within a preset time period, and minimizing the risk of customers developing negative emotions caused by business marketing harassment, thereby maintaining customer stickiness.

[0173] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0174] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0175] To solve the above technical problems, the present application also provides a computer device. Fig.10 , Fig.10 This is a basic structural block diagram of the computer device in this embodiment.

[0176] The computer device 10 includes a memory 10a, a processor 10b, and a network interface 10c which are interconnected and communicated via a system bus. It should be noted that: Fig.10 Only a computer device 10 having a component memory 10a, a processor 10b, and a network interface 10c is shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to a microprocessor, an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital processor (DSP), an embedded device, etc.

[0177] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0178] The memory 10a includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 10a can be an internal storage unit of the computer device 10, such as a hard disk or memory of the computer device 10. In other embodiments, the memory 10a can also be an external storage device of the computer device 10, such as a plug-in hard disk equipped on the computer device 10, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the memory 10a can also include both the internal storage unit of the computer device 10 and its external storage device. In this embodiment, the memory 10a is generally used to store an operating system and various application software installed on the computer device 10, such as a computer-readable instruction of a service recommendation method, etc. In addition, the memory 10a can also be used to temporarily store various data that have been output or will be output.

[0179] The processor 10b may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 10b is generally used to control the overall operation of the computer device 10. In this embodiment, the processor 10b is used to run the computer-readable instructions stored in the memory 10a or process data, such as computer-readable instructions for running the service recommendation method.

[0180] The network interface 10c may include a wireless network interface or a wired network interface. The network interface 10c is generally used to establish a communication connection between the computer device 10 and other electronic devices.

[0181] The computer device proposed in this embodiment belongs to the field of artificial intelligence technology and is applied to service management recommendation image scenarios. This application uses artificial intelligence and big data matching methods to intelligently and automatically screen out service agents that are more suitable for customers. After that, the service agents recommend services to target customers based on the contact strategy, ensuring that the agents strictly follow the contact strategy when recommending services to customers, avoiding excessive marketing to customers within a preset time period, and trying to prevent customers from having negative emotions caused by business marketing harassment, thereby maintaining customer stickiness.

[0182] The present application also provides another implementation, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by a processor to enable the processor to perform the steps of the service recommendation method as described above.

[0183] The computer-readable storage medium proposed in this embodiment belongs to the field of artificial intelligence technology and is applied to service management recommendation image scenarios. This application uses artificial intelligence and big data matching methods to intelligently and automatically screen out service agents that are more suitable for customers. After that, the service agents recommend services to target customers based on the contact strategy, ensuring that the agents strictly follow the contact strategy when recommending services to customers, avoiding excessive marketing to customers within a preset time period, and trying to prevent customers from having negative emotions caused by business marketing harassment, thereby maintaining customer stickiness.

[0184] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0185] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A service recommendation method, characterized in that: The steps include: Obtain target customer profile data; Input the target customer portrait data into the pre-trained customer-seat matching model to screen out the target service agent; Input the target customer portrait data into the pre-trained reach classification model to identify the reach category corresponding to the target customer, wherein each reach category corresponds to a reach strategy; Determine the contact strategy corresponding to the target customer based on the contact category, wherein the contact strategy includes the maximum number of communications within a preset time period, optimal communication time period information, optimal communication duration for each communication, and preset communication interval time information; Recommending services to the target customer through the target service agent and according to the contact strategy; Acquire communication data each time a service recommendation is made to the target customer, and input the communication data into a focus recognition model to obtain focus information that the target customer is concerned about during this service recommendation; The script for the next service recommendation is adjusted according to the focus information, and the next service recommendation is made to the target customer according to the script.

2. The service recommendation method according to claim 1, characterized in that: Before executing the step of inputting the target customer portrait data into the pre-trained customer agent matching model to screen out the target service agent, the method further includes: Obtaining business profile data of all agents, wherein the business profile data includes profile data of customers who have successfully converted after the agents made business recommendations; Input the business portrait data of all the seats into the customer-seat matching model to be trained, perform learning training, and learn the customer group portrait data corresponding to all the seats; Deploy the correlation relationship between the customer group portrait data corresponding to all seats as reference knowledge to the customer-seat matching model to be trained, so as to obtain a pre-trained customer-seat matching model; The step of inputting the target customer portrait data into the pre-trained customer agent matching model to screen out the target service agent specifically includes: Step 401, calculating the similarity between the target customer portrait data and all customer group portrait data according to a preset similarity algorithm; Step 402, sorting the customer groups to which the target customer portrait data belongs in descending order based on the similarity to obtain a descending sequence; Step 403, generating an agent priority sequence for the target customer according to the association relationship between the descending sequence and the customer group portrait data corresponding to all the agents respectively; Step 404: Filter out target service agents according to the agent priority sequence and the real-time working status of all agents, wherein the real-time working status includes being on duty, being idle at work, and the predicted end time of being on duty.

3. The service recommendation method according to claim 2, characterized in that: The step of screening out target service seats according to the seat priority sequence and the real-time working status of all seats specifically includes: Identify the real-time working status of all agents; If all the agents are working, the target customer's waiting time is determined according to the target customer portrait data, and the waiting time is compared with the predicted end time when all the agents are working; If the predicted end time of all the agents in work is greater than the waiting time, step 404 is executed again; If the predicted end time of the agent's work is not greater than the waiting time, step 404 is executed again, and the communication information of the target customer is added to the preset call waiting sequence pool; If there is an agent who is idle at work and the agent who is idle at work is the only one, the agent is used as the target service agent; If there is an agent who is not working at a unique time, the agent who is not working at a unique time is selected as the target service agent according to the agent priority sequence.

4. The service recommendation method according to claim 1, characterized in that: Before executing the step of inputting the target customer portrait data into the pre-trained reach classification model to identify the reach category corresponding to the target customer, the method further includes: Obtain the contact categories of all historical customers and the individual portrait data of all historical customers; Perform clustering processing on the individual portrait data of historical customers of the same reach category to obtain clustering processing results; According to the clustering processing results, group customer portrait data corresponding to all contact categories are sorted out; Deploy the group customer portrait data corresponding to all reach categories as reach classification knowledge into a preset classification model to obtain the pre-trained reach classification model; The step of inputting the target customer portrait data into the pre-trained reach classification model to identify the reach category corresponding to the target customer specifically includes: Filter out the group customer portrait data that is most similar to the target customer portrait data according to a preset similarity algorithm; The reach category corresponding to the group customer portrait data with the highest similarity is used as the reach category corresponding to the target customer.

5. The service recommendation method according to claim 1, characterized in that: The step of recommending services to the target customer through the target service agent and according to the contact strategy specifically includes: Obtain the number of times the target customer has communicated within a preset time period, as well as the time information of the last communication; According to the contact strategy, the maximum number of communications, the best communication time period, the best communication duration for each communication, and the preset communication interval time information corresponding to the target customer within a preset time period are identified; Determining whether the number of communications has reached the maximum number of communications; If the number of communications has reached the maximum number of communications, then stop recommending services to the target customer; If the number of communications has not reached the maximum number of communications, then calculating the interval time information between the current time and the time information of the last communication; If the interval time information does not reach the preset communication interval time information, a prompt for delaying service recommendation is sent to the target agent; If the interval time information has reached the preset communication interval time information, the target agent is prompted to make service recommendations to the target customer based on the optimal communication time period information and the optimal communication duration during this communication.

6. The service recommendation method according to claim 2, characterized in that: After executing the step of recommending services to the target customer through the target service agent and according to the contact strategy, the method further includes: Obtaining service feedback information sent by the target customer; Identifying the recommendation effect of the service recommendation performed this time according to the service feedback information, wherein the recommendation effect includes three states: positive effect, negative effect and no effect; If the recommendation effect is a positive effect, the contact strategy is updated by shortening the preset communication interval time information; If the recommendation effect is a reverse effect, a service recommendation is made to the target customer by extending the preset communication interval time information or / and selecting a seat with a higher priority from the seat priority sequence; If the recommendation effect is ineffective, a seat with a higher priority is selected from the seat priority sequence to continue to make the next service recommendation to the target customer according to the contact strategy.

7. The service recommendation method according to claim 1, characterized in that: The step of inputting the communication data into the focus recognition model to obtain the focus information that the target customer is concerned about when recommending this service specifically includes: Textual processing of the communication data; Perform semantic analysis and word segmentation on the text content after text processing; The occurrence frequency of each business noun in the word segmentation processing result is counted through a preset business dictionary, and the business verb corresponding to each business noun is identified through the semantic analysis result; According to the occurrence frequency of each business noun and the business verb corresponding to each business noun, the focus information that the target customer pays attention to during the service recommendation is determined.

8. A service recommendation device, characterized in that: include: A target customer portrait data acquisition module is used to acquire target customer portrait data; A target service agent screening module is used to input the target customer portrait data into the pre-trained customer agent matching model to screen out the target service agent; A reach category identification module is used to input the target customer portrait data into a pre-trained reach classification model to identify the reach category corresponding to the target customer, wherein each reach category corresponds to a reach strategy; A contact strategy determination module, used to determine the contact strategy corresponding to the target customer based on the contact category, wherein the contact strategy includes the maximum number of communications within a preset time period, the best communication time period information, the best communication duration for each communication, and the preset communication interval time information; A service reaching strategy recommendation module, used to recommend services to the target customer through the target service agent and according to the reaching strategy; A focus information acquisition module is used to obtain the communication data when recommending services to the target customer each time, and input the communication data into the focus recognition model to obtain the focus information that the target customer pays attention to when recommending services this time; The service recommendation adjustment module is used to adjust the text of the next service recommendation according to the focus information, and make the next service recommendation to the target customer according to the text of the text.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the service recommendation method according to any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the service recommendation method according to any one of claims 1 to 7 are implemented.

Images