An intelligent IP phone outbound call cloud system and method

Abstract

The present invention relates to the field of artificial intelligence technology, and discloses an intelligent IP phone outbound call cloud system and method, the system includes an outbound call user classification module, an outbound call strategy mapping module, a similarity calculation module, an outbound call strategy execution module and a telephone outbound call intention strategy execution module, extracts outbound call user information and outbound call service type based on historical outbound call records, classifies outbound call users according to outbound call user information; generates outbound call parameters according to outbound call service type, generates outbound call strategies of intelligent IP phones according to outbound call parameters and outbound call accounts, maps outbound call strategies with outbound call user categories; calculates similarity between target users and outbound call user categories according to outbound call features corresponding to outbound call information; selects outbound call strategies according to similarity, and performs outbound call operations on intelligent IP phones according to outbound call strategies. The present invention can improve the accuracy of outbound calls made by intelligent IP phones.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent IP telephone outbound call cloud system and method. Background Art

[0002] With the development of cloud computing technology, enterprises can rent computing resources and services through the Internet without having to maintain complex hardware infrastructure themselves, enabling the outbound call system to achieve more intelligent functions, such as voice recognition and natural language processing. However, in order to make outbound calls more accurate, it is necessary to analyze the outbound call strategy to achieve efficient outbound calls of intelligent IP phones.

[0003] The existing intelligent IP phone outbound calling cloud technology uses VoIP (Voice over IP) technology to make and manage calls, allowing enterprises to automatically make large-scale outbound calls. In actual applications, the voice over IP protocol cannot accurately target customer needs for outbound calls, resulting in potential customers being missed or users who are not interested in outbound calls, resulting in low accuracy when making intelligent IP phone outbound calls. Summary of the invention

[0004] The present invention provides an intelligent IP phone outbound call cloud system and method, the main purpose of which is to solve the problem of low accuracy when making intelligent IP phone outbound calls.

[0005] To achieve the above-mentioned purpose, the present invention provides an intelligent IP phone outbound call cloud system, including an outbound call user classification module, an outbound call strategy mapping module, a similarity calculation module, an outbound call strategy execution module and a phone outbound call intention strategy execution module, wherein:

[0006] The outbound user classification module is used to extract outbound user information and outbound service types based on the pre-acquired historical outbound call records, classify outbound users according to the outbound user information, and obtain outbound user categories;

[0007] The outbound call strategy mapping module is used to generate outbound call parameters according to the outbound call service type, generate an outbound call strategy for the intelligent IP phone according to the outbound call parameters and the outbound call account pre-configured on the cloud service platform, map the outbound call strategy with the outbound call user category, and obtain an outbound call mapping relationship, which is specifically used to: divide the outbound call users in the outbound call user category into customer stages according to the behavior frequency of each outbound call user to obtain an outbound call user stage category; generate an outbound call label corresponding to each outbound call user according to the outbound call user stage category and the outbound call service type; generate a policy label corresponding to the external policy according to the service type corresponding to the outbound call strategy and the outbound call user stage category; calculate the label correlation coefficient between the outbound call label and the policy label: in, For the Outbound call label and The label correlation coefficient between the strategy labels, For the Outbound user stage category label in outbound call labels, For the The service type tag in the outbound call tag, For the Outbound user stage category label in the strategy label, For the The business type tag in the strategy tag, is the minimum function;

[0008] Mapping the outbound call strategy with the outbound call user category according to the label correlation coefficient to obtain an outbound call mapping relationship;

[0009] The similarity calculation module is used to obtain the outbound call information of the target user, and calculate the similarity between the target user and the outbound call user category according to the outbound call features corresponding to the outbound call information;

[0010] The outbound call strategy execution module is used to select the outbound call strategy in the outbound call mapping relationship according to the outbound call information when the similarity is a preset first similarity, and perform an outbound call operation on the intelligent IP phone according to the outbound call strategy;

[0011] The outbound call intention strategy execution module is used to identify the outbound call intention of the target user according to the outbound call information when the similarity is a preset second similarity, generate an outbound call intention strategy according to the outbound call event type corresponding to the outbound call intention, and perform an outbound call operation on the smart IP phone according to the outbound call intention strategy.

[0012] Optionally, when the outbound user classification module classifies the outbound user according to the outbound user information to obtain the outbound user category, it is specifically used to:

[0013] Extracting behavior-driven data corresponding to each outbound calling user in the outbound calling user information;

[0014] Counting the behavior frequency of each outbound calling user according to the behavior driving data;

[0015] determining a behavior category of each outbound calling user according to the behavior frequency;

[0016] The outbound user is classified according to the behavior category to obtain an outbound user category.

[0017] Optionally, when the outbound call policy mapping module generates the outbound call parameters according to the outbound call service type, it is specifically used to:

[0018] Determine the outbound calling skills and the outbound calling mode according to the outbound calling service type;

[0019] Calculate the best time for outbound calls based on the frequency of the behavior;

[0020] The optimal time for outbound calls, the service outbound call mode and the service outbound call speech are encapsulated as outbound call parameters.

[0021] Optionally, when the outbound call strategy mapping module generates an outbound call strategy for the intelligent IP phone according to the outbound call parameters and the pre-configured outbound call account on the cloud service platform, it is specifically used to:

[0022] Determine an outbound call rule according to the business outbound call mode in the outbound call parameter;

[0023] Determine the outbound call time according to the outbound call best time in the outbound call parameters;

[0024] Allocate outbound calls to the outbound call account according to the preset outbound call task to obtain an outbound call allocation account;

[0025] Adding the business outbound call script in the outbound call parameters to the outbound call allocation account;

[0026] The outbound call rules, the outbound call time, the outbound call allocation account and the allocated business outbound call speech are combined into an outbound call strategy for the intelligent IP phone.

[0027] Optionally, when the similarity calculation module calculates the similarity between the target user and the outbound user category according to the outbound call feature corresponding to the outbound call information, it is specifically used to:

[0028] Extracting characteristic keywords corresponding to the outbound call characteristics;

[0029] Extracting category keywords corresponding to the outbound user category;

[0030] Select similar feature words between the feature keyword and the category keyword;

[0031] Calculate the similarity between the target user and the outbound user category based on the similar feature words: in, For the target user The outbound user category The similarity between is the number of the characteristic keywords, is the number of keywords in the category, is the number of similar feature words, For the The weight of similar feature words, For the The similarity scores of similar feature words.

[0032] Optionally, when the outbound call strategy execution module selects the outbound call strategy in the outbound call mapping relationship according to the outbound call information, it is specifically used to:

[0033] Determine the target outbound user category according to the similarity;

[0034] Dividing the target outbound user category into user stages according to the user behavior frequency corresponding to the outbound call information to obtain the target outbound user stage category;

[0035] Performing category mapping between the target outbound user stage category and the outbound user stage category in the outbound call mapping relationship;

[0036] Select an outbound calling strategy based on the mapped target outbound user stage category.

[0037] Optionally, when the outbound call strategy execution module performs an outbound call operation on the intelligent IP phone according to the outbound call strategy, it is specifically used to:

[0038] Obtaining in real time the account status corresponding to the outbound call allocation account in the outbound call strategy;

[0039] Calculate the number of outbound calls corresponding to the smart IP phone based on the account status: in, The number of outbound calls issued, is the number of accounts corresponding to the idle state in the account state, is the number of accounts corresponding to the connected state in the account status, is the total number of calls, is the number of successful calls, is the dialing speed adjustment factor, The number of accounts corresponding to the occupied state in the account status;

[0040] Allocating the outbound call strategy to the intelligent IP phone according to the number of outbound calls issued;

[0041] Make outbound calls to target users according to the assigned outbound calling strategy.

[0042] Optionally, when the telephone outbound call intention strategy execution module identifies the outbound call intention of the target user according to the outbound call information, it is specifically used to:

[0043] Extracting outbound call features corresponding to the outbound call information;

[0044] Determining an outbound call keyword of the target user according to the outbound call feature;

[0045] Identify the high-level intention and low-level intention corresponding to the target user according to the outbound call keyword;

[0046] The intention probability of the target user is calculated by the high-level intention and the low-level intention: in, For low level intention Given a high-level intent The probability of intention under the condition of is the joint probability of high-level intention and low-level intention occurring simultaneously, is the marginal probability of high-level intention;

[0047] The outbound calling intention of the target user is identified according to the intention probability.

[0048] Optionally, when the outbound call intention strategy execution module generates the outbound call intention strategy according to the outbound call event type corresponding to the outbound call intention, it is specifically used to:

[0049] Determine an outbound call rule and an outbound call time according to the outbound call event type corresponding to the outbound call intention;

[0050] Generate an outbound call intention script according to the outbound call intention;

[0051] Adding the outbound call intention script to a pre-configured outbound call distribution account;

[0052] The outbound call rules, the outbound call time, the outbound call allocation account and the allocated outbound call intention words are combined into an outbound call intention strategy.

[0053] In order to solve the above problems, the present invention also provides an intelligent IP phone outbound cloud method, the method comprising:

[0054] Extracting outbound user information and outbound service types based on the pre-acquired historical outbound call records, and classifying outbound users according to the outbound user information to obtain outbound user categories;

[0055] Generate an outbound call parameter according to the outbound call service type, generate an outbound call strategy for the smart IP phone according to the outbound call parameter and an outbound call account pre-configured on the cloud service platform, and map the outbound call strategy with the outbound call user category to obtain an outbound call mapping relationship;

[0056] Acquire outbound call information of a target user, and calculate the similarity between the target user and the outbound call user category according to outbound call features corresponding to the outbound call information;

[0057] When the similarity is a preset first similarity, selecting an outbound call strategy in the outbound call mapping relationship according to the outbound call information, and performing an outbound call operation on the intelligent IP phone according to the outbound call strategy;

[0058] When the similarity is a preset second similarity, the outbound calling intention of the target user is identified according to the outbound calling information, an outbound calling intention strategy is generated according to the outbound calling event type corresponding to the outbound calling intention, and an outbound calling operation is performed on the smart IP phone according to the outbound calling intention strategy.

[0059] The embodiment of the present invention extracts user information and service types through historical records, and can make analysis and decisions based on real data; classified users can help formulate more accurate outbound call strategies, ensuring that each user group can get the most suitable communication method and content; outbound call parameters are generated according to user classification and service type, making the outbound call strategy more personalized, thereby improving response rate and customer satisfaction; by calculating similarity, the matching degree between the target user and the existing user category can be determined, so as to select the most appropriate outbound call strategy; the most appropriate outbound call strategy is selected according to the similarity, and the outbound call operation is executed according to the strategy, ensuring the efficiency of the outbound call process. Therefore, the intelligent IP phone outbound call cloud system, method and medium proposed by the present invention can solve the problem of low accuracy when making outbound calls by intelligent IP phones. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A functional module diagram of an intelligent IP phone outbound call cloud system provided by an embodiment of the present invention;

[0061] Figure 2 A flowchart of an operating method of an intelligent IP phone outbound call cloud system provided in one embodiment of the present invention.

[0062] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0063] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0064] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0066] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0067] In addition, the step sequence in the following method embodiments is only an example and not a strict limitation.

[0068] In fact, the server-side device deployed by the intelligent IP phone outbound cloud system may be composed of one or more devices. The above-mentioned intelligent IP phone outbound cloud system can be implemented as: a business instance, a virtual machine, and a hardware device. For example, the intelligent IP phone outbound cloud system can be implemented as a business instance deployed on one or more devices in a cloud node. In simple terms, the intelligent IP phone outbound cloud system can be understood as a software deployed on a cloud node, which is used to provide an intelligent IP phone outbound cloud system for each user terminal. Alternatively, the intelligent IP phone outbound cloud system can also be implemented as a virtual machine deployed on one or more devices in a cloud node. Application software for managing each user terminal is installed in the virtual machine. Alternatively, the intelligent IP phone outbound cloud system can also be implemented as a server composed of many hardware devices of the same or different types, and one or more hardware devices are set to provide an intelligent IP phone outbound cloud system for each user terminal.

[0069] In terms of implementation, the intelligent IP phone outbound call cloud system and the user end are adapted to each other. That is, the intelligent IP phone outbound call cloud system is an application installed on the cloud service platform, and the user end is a client that establishes a communication connection with the application; or the intelligent IP phone outbound call cloud system is implemented as a website, and the user end is implemented as a web page; or the intelligent IP phone outbound call cloud system is implemented as a cloud service platform, and the user end is implemented as a small program in the instant messaging application.

[0070] Reference Figure 1 , which is a functional module diagram of an intelligent IP phone outbound call cloud system provided by an embodiment of the present invention.

[0071] The intelligent IP phone outbound cloud system 100 of the present invention can be set in a cloud server. In terms of implementation, it can be used as one or more service devices, or it can be installed as an application on the cloud (such as a server of a mobile service operator, a server cluster, etc.), or it can be developed as a website. According to the functions implemented, the intelligent IP phone outbound cloud system 100 can include an outbound user classification module 101, an outbound call strategy mapping module 102, a similarity calculation module 103, an outbound call strategy execution module 104 and a telephone outbound call intention strategy execution module 105. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by a device processor and can complete fixed functions, which are stored in the memory of the device.

[0072] In the embodiment of the present invention, in the intelligent IP phone outbound call cloud system, each of the above modules can be implemented independently and called with other modules. The call here can be understood as a module that can connect to multiple modules of another type and provide corresponding services to the multiple modules connected to it. For example, the sharing evaluation module can call the same information collection module to obtain the information collected by the information collection module. Based on the above characteristics, in the intelligent IP phone outbound call cloud system provided by the embodiment of the present invention, the scope of application of the intelligent IP phone outbound call cloud system architecture can be adjusted by adding modules and directly calling them without modifying the program code, so as to achieve cluster-based horizontal expansion, so as to achieve the purpose of quickly and flexibly expanding the intelligent IP phone outbound call cloud system. In actual applications, the above modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in cloud servers.

[0073] The following is a description of each component and specific workflow of the intelligent IP phone outbound call cloud system in conjunction with a specific embodiment:

[0074] The outbound user classification module 101 is used to extract outbound user information and outbound service types based on pre-acquired historical outbound call records, and classify outbound users according to the outbound user information to obtain outbound user categories.

[0075] In an embodiment of the present invention, the historical outbound call record refers to a record of outbound call activities conducted in the past, including detailed information of each outbound call, such as the time, telephone number, content, and result of the outbound call, etc., wherein the historical outbound call record can be obtained from a pre-stored storage area through a computer statement with a data capture function (such as a Java statement, a Python statement, etc.), and the storage area includes but is not limited to a database and a blockchain.

[0076] In detail, the outbound user information refers to the basic attributes and behavioral attributes of the outbound user, and the basic attributes include but are not limited to name, telephone number, address, and customer ID. The behavioral attributes include the user's response behavior and complaint behavior during the outbound call process, wherein the user's basic attributes and behavioral attributes can be extracted through natural speech processing (NPL) technology, such as using named entity recognition to identify and extract the user's name from the outbound content, etc. The outbound business type is extracted from the content of the outbound call record or the attached tags, and by analyzing the outbound call record content, the purpose of the outbound call, such as sales, service, investigation, etc., is identified, and then the type of outbound business is determined according to the keywords or subject classification of the outbound content. The text classification algorithm (such as naive Bayes, support vector machine, deep learning model, etc.) can be used to classify the outbound content and identify the business type, and the outbound content can be classified according to a predefined keyword list or rules. For example, the outbound business type includes but is not limited to sales business, customer service business, and investigation business.

[0077] Furthermore, based on the user information of each outbound user and the corresponding outbound service type, different outbound users can be classified, outbound resources and personnel can be reasonably allocated, and outbound efficiency can be improved, thereby obtaining outbound response and conversion effects for different user categories, so as to optimize the outbound strategy.

[0078] In an embodiment of the present invention, the outbound user category refers to the result of dividing the outbound users into different groups according to their behavioral characteristics or interaction patterns. Each category represents a different behavioral pattern exhibited by the user during the outbound call process, such as sales category, customer service category, and survey category, which in turn helps to formulate targeted marketing strategies and optimize outbound resource allocation.

[0079] In the embodiment of the present invention, when the outbound user classification module 101 classifies the outbound user according to the outbound user information to obtain the outbound user category, it is specifically used to:

[0080] Extracting behavior-driven data corresponding to each outbound calling user in the outbound calling user information;

[0081] Counting the behavior frequency of each outbound calling user according to the behavior driving data;

[0082] determining a behavior category of each outbound calling user according to the behavior frequency;

[0083] The outbound user is classified according to the behavior category to obtain an outbound user category.

[0084] In detail, the behavior-driven data corresponding to each user is obtained from the outbound user information, such as the user's response behavior, complaint behavior, and hang-up behavior during the outbound call process, and based on the number of historical outbound calls, the number of responses, complaints, and hang-up times corresponding to the user's sales, customer service, and survey categories are counted, wherein the number of responses refers to the number of responses of the user in each category. A high number of responses may indicate that the user is more concerned about or active in the category; the number of complaints refers to the number of complaints of the user in each category. A large number of complaints may indicate that the user is dissatisfied with the service or product; the number of hang-ups refers to the number of hang-ups of the user in each category. A large number of hang-ups may indicate that the user is not interested in the outbound content or has other negative feedback, then the total response frequency, total complaint frequency, and total hang-up frequency of each user in the sales, customer service, and survey categories are calculated, such as the response rate is (number of responses / total number of outbound calls), the complaint rate is (number of complaints / total number of outbound calls), and the hang-up rate is (number of hang-ups / total number of outbound calls), and then the outbound users are classified according to the statistical behavior frequencies.

[0085] Specifically, users are divided into different behavior categories based on behavior frequency data. For example, if a user's response times to sales are higher than the number of complaints and the number of hang-ups, then the user is classified into the sales category, indicating that the user is more active or interested in sales activities. When analyzing and optimizing strategies, you can give priority to these users and provide more targeted sales plans. If a user's response times to customer service are higher than the number of complaints and the number of hang-ups, then the user is classified into the customer service category, thereby obtaining different outbound user categories.

[0086] Furthermore, in order to ensure the efficiency and targeting of outbound calling activities, outbound calling parameters can be customized by clarifying the business type (such as sales, customer service, and surveys) to meet different business needs, which will help increase user response rates, reduce invalid outbound calls, and ultimately improve business results and customer satisfaction.

[0087] The outbound call strategy mapping module 102 is used to generate outbound call parameters according to the outbound call service type, generate an outbound call strategy for the smart IP phone according to the outbound call parameters and the outbound call account pre-configured on the cloud service platform, and map the outbound call strategy with the outbound call user category to obtain an outbound call mapping relationship.

[0088] In the embodiment of the present invention, the outbound call parameters refer to various settings and information used to guide and optimize outbound call activities, including the best time for outbound calls, outbound call modes (such as automatic or manual dialing), script content, outbound call frequency, etc. The outbound call parameters help ensure the efficiency and effectiveness of outbound call activities, so that outbound calls can better target the target audience and improve the response rate and business success rate.

[0089] In the embodiment of the present invention, when the outbound call policy mapping module 102 generates the outbound call parameters according to the outbound call service type, it is specifically used to:

[0090] Determine the outbound calling skills and the outbound calling mode according to the outbound calling service type;

[0091] Calculate the best time for outbound calls based on the frequency of the behavior;

[0092] The optimal time for outbound calls, the service outbound call mode and the service outbound call speech are encapsulated as outbound call parameters.

[0093] Specifically, design special script templates for different business types (such as sales, customer service, and surveys). For example, sales outbound calls may need to emphasize product advantages and promotional information; customer service outbound calls should focus on solving customer problems and providing support; survey outbound calls need to clarify the purpose and questions of the survey and choose the appropriate outbound call mode (such as automatic dialing, manual dialing, scheduled outbound calls, etc.). Sales may be more suitable for active dialing mode, while customer service may need to adjust the outbound call strategy based on customer feedback.

[0094] Specifically, by analyzing the user's response, complaint and hang-up data, the time when the user is most likely to answer the phone is determined, and the response rate in different time periods is calculated to identify the time periods in which the user is most likely to answer the phone. For example, the answering ratio per hour is counted to find the time period with the highest answering rate. For example, if the statistical data shows that the user's response rate is the highest between 4 pm and 6 pm, this period can be set as the best outbound call time, so that the best outbound call time, business outbound call mode and business outbound call script are encapsulated as outbound call parameters, that is, the time period parameter "start_time" is set to specify the time range suitable for outbound calls, such as {"start_time": "08:00", "end_time": "18:00"}, and the execution mode of the outbound call is defined, such as automatic dialing or manual dialing, such as {"dialing_mode": "auto"}, and the script template used for outbound calls is encapsulated, such as {"script": "Hello, this is XX company, we provide the latest product information, do you have time to learn about it?"}, and then the defined outbound call parameter structure is combined into a configuration object, so as to obtain the outbound call parameters corresponding to different outbound call service types.

[0095] For example, in the sales business, the outbound call script for potential users is to introduce the main advantages of the product and the latest promotions, ask about their needs, and provide trial or demonstration opportunities. For example, the sales representative can say: "Hello, we now have a limited-time discount event that is very suitable for your needs. Can I arrange a free trial for you?" For loyal users, call them to thank them for their long-term support, provide special offers or invite them to exclusive events. For example: "Thank you for your continued support. We have prepared an exclusive discount for you. I hope you like it!" For high-potential users, call to conduct a detailed needs analysis, introduce customized solutions, and provide advanced service options. For example: "We understand your interest in efficient solutions. I can arrange an in-depth product demonstration for you to help you achieve your goals."

[0096] Furthermore, by using the optimal outbound call time in the outbound call parameters, we can ensure that the call is made during the time period when the user is most likely to answer the call, thereby improving the answering rate and success rate. The configuration of the outbound call mode (such as automatic dialing) can optimize the use of outbound call resources, reduce manual intervention, and improve outbound call efficiency. The use of preset speech templates can ensure that the outbound call content is consistent and professional, enhance the user's willingness to answer the call and the quality of the conversation, thereby generating different outbound call strategies based on the outbound call parameters, which can effectively improve the overall effect and efficiency of outbound call activities.

[0097] In the embodiment of the present invention, the outbound call strategy refers to the plans and rules formulated in telemarketing or customer service to improve the efficiency and effectiveness of outbound calls, including outbound call rules, which set specific rules for outbound call operations, such as dialing frequency, dialing order, whether to use an automatic dialing system, etc.; outbound call time, which selects an appropriate time period for outbound calls to ensure that target customers are more likely to answer the phone. For example, avoid disturbing customers in the morning or evening; outbound call scripts, which preset standard dialogue scripts or scripts to ensure information consistency and professionalism during the outbound call process; outbound call allocation, which allocates outbound call tasks to different outbound call accounts or personnel to ensure that each account or personnel is responsible for a specific customer group.

[0098] In the embodiment of the present invention, when the outbound call strategy mapping module 102 generates an outbound call strategy for the smart IP phone according to the outbound call parameters and the pre-configured outbound call account on the cloud service platform, it is specifically used to:

[0099] Determine an outbound call rule according to the business outbound call mode in the outbound call parameter;

[0100] Determine the outbound call time according to the outbound call best time in the outbound call parameters;

[0101] Allocate outbound calls to the outbound call account according to the preset outbound call task to obtain an outbound call allocation account;

[0102] Adding the business outbound call script in the outbound call parameters to the outbound call allocation account;

[0103] The outbound call rules, the outbound call time, the outbound call allocation account and the allocated business outbound call speech are combined into an outbound call strategy for the intelligent IP phone.

[0104] In detail, the outbound call rule refers to the dialing mode, including automatic dialing, setting the automatic dialing frequency, maximum dialing times, automatic hang-up, etc.; manual dialing, setting manual dialing task allocation, dialing time period; smart dialing, setting smart dialing time and content according to user data analysis; mixed dialing, configuring a combination of different dialing modes, optimizing the dialing timing, and then determining the outbound call time according to the best time for outbound calls, such as the best answering time period, 12 noon to 2 pm, 4 pm to 6 pm, working hours, 9 am to 6 pm, rest time, avoiding dialing when the user may be resting, and allocating accounts according to outbound call tasks, setting each In order to set the dialing limit and task volume of each account, it is necessary to clarify the requirements of each outbound call task, such as the number of customers to be dialed, task priority, etc., match and allocate according to the task requirements and the capabilities or resources of each outbound call account, set allocation rules, such as allocating tasks to appropriate outbound call accounts by region, time period or task type, and applying the business outbound call script template to each outbound call allocation account to ensure consistency of the script, and then merge the outbound call rule A, outbound call time B, outbound call allocation account C and script D to form a complete intelligent IP phone outbound call strategy. The intelligent IP phone outbound call strategy refers to {A, B, C, D}.

[0105] For example, assume that the automatic dialing mode is used and relevant accounts are configured on the cloud service platform. For example, the dialing mode is automatic dialing, the outbound calling time is from 10 a.m. to 4 p.m. on weekdays, and the account configuration is that account A is responsible for sales customers and the dialing frequency is set to 50 words per hour. Account B is responsible for customer service customers and the dialing frequency is 30 times per hour. At the same time, it is necessary to ensure that user data is updated in real time, including user behavior data and response records, and configure standard speech synthesis to ensure natural and clear speech.

[0106] Furthermore, in order to ensure that outbound calling activities can provide personalized and accurate communication for different user categories, it is necessary to allocate appropriate outbound calling strategies according to user categories to avoid resource waste and invalid outbound calls.

[0107] In the embodiment of the present invention, the outbound call mapping relationship refers to the relationship between the outbound call strategy and the specific user category in the outbound call, and then the outbound call strategy is adjusted according to the outbound call mapping relationship to more effectively target different user categories and improve the effect and efficiency of the outbound call activity.

[0108] In the embodiment of the present invention, when the outbound call strategy mapping module 102 maps the outbound call strategy with the outbound call user category to obtain an outbound call mapping relationship, it is specifically used to:

[0109] Dividing the outbound users in the outbound user category into customer stages according to the behavior frequency of each outbound user to obtain an outbound user stage category;

[0110] Generate an outbound call label corresponding to each outbound call user according to the outbound call user stage category and the outbound call service type;

[0111] Generate a policy label corresponding to the external policy according to the service type corresponding to the outbound call policy and the outbound call user stage category;

[0112] Calculate the label correlation coefficient between the outbound call label and the strategy label: in, For the Outbound call label and The label correlation coefficient between the strategy labels, For the Outbound user stage category label in outbound call labels, For the The service type tag in the outbound call tag, For the Outbound user stage category label in the strategy label, For the The business type tag in the strategy tag, is the minimum function;

[0113] The outbound call strategy is mapped to the outbound call user category according to the label correlation coefficient to obtain an outbound call mapping relationship.

[0114] In detail, according to the behavior frequency of outbound users, users are divided into different stage categories. For example, users can be divided into potential users, loyal users, and high-potential users. Potential users belong to the "potential market" category and need to be converted into actual customers through marketing and sales strategies; loyal users belong to the "existing customers" category and need to promote continuous purchases by maintaining and enhancing customer relationships; high-potential customers belong to the "key focus customers" category and need to maximize value through refined services and personalized strategies, thereby obtaining different outbound user stage categories, and then for each outbound user, an outbound call label is generated according to its stage category and outbound call business type. The outbound call label identifies the user category status and business type, such as the category status is potential user, loyal user, high-potential user, and the business type is sales, customer service, and investigation; and the corresponding strategy label is generated according to the business type and outbound user stage in the outbound strategy. The strategy label identifies the user stage and business type targeted by each strategy.

[0115] Specifically, the identifier corresponding to potential users is configured as 1, the identifier corresponding to loyal users is configured as 2, the identifier corresponding to high-potential users is configured as 3, the identifier corresponding to sales is configured as 11, the identifier corresponding to customer service is configured as 22, and the identifier corresponding to survey is configured as 33. Then the outbound call tag may be , , etc., while the strategy tags include , , , , , , , , , thereby calculating the correlation coefficient between the outbound call label and the strategy label, and finding the closest strategy label by minimizing the correlation coefficient to match the outbound call label. The outbound user stage category label in the outbound call label is that the label corresponding to the potential user is configured as 1, the label corresponding to the loyal user is configured as 2, and the label corresponding to the high-potential user is configured as 3. The business type tag in the outbound call tag is that the tag corresponding to sales is configured as 11, the tag corresponding to customer service is configured as 22, and the tag corresponding to investigation is configured as 33, so as to calculate the difference between the outbound call tag and the strategy tag. If the outbound call tag is , then match the labels with all the strategy labels respectively, and get the strategy labels If the correlation coefficient between the strategy label and the outbound call label is the smallest, the strategy label is matched with the outbound call label. Based on the strategy label, the outbound call strategy corresponding to the strategy is extracted. Based on this outbound call label, the outbound call user category corresponding to the outbound call is extracted, thereby forming an outbound call mapping relationship. For example, if the sales business includes potential users, loyal users, and high-potential users, the corresponding outbound call strategies are determined according to the business types corresponding to the outbound call strategies.

[0116] Furthermore, in order to allocate outbound calls to real-time users, it is necessary to understand the user's matching degree so that a more effective personalized outbound call plan can be formulated to ensure that the outbound call content is consistent with user needs and improve customer satisfaction and response rate.

[0117] The similarity calculation module 103 is used to obtain the outbound call information of the target user, and calculate the similarity between the target user and the outbound call user category according to the outbound call features corresponding to the outbound call information.

[0118] In an embodiment of the present invention, the outbound call information refers to basic data and behavior data of the target user, such as basic data including name, age, contact information, etc., and behavior data including response, complaint, and hang-up. The outbound call information of the target user can be obtained from a pre-stored storage area through a computer statement with a data capture function (such as a Java statement, a Python statement, etc.), and the storage area includes but is not limited to a database and a blockchain.

[0119] Furthermore, identifying the outbound call category that matches the target user's characteristics helps to increase the chances of closing a deal. Therefore, it is necessary to determine the user category to which the target user belongs so that an outbound call strategy can be assigned to the target user more accurately.

[0120] In the embodiment of the present invention, the similarity refers to the similarity between the target user and the outbound user category, and the similarity reflects the matching degree of the target user characteristics and the outbound user category characteristics in terms of keywords, that is, the outbound user category to which the target user belongs.

[0121] In the embodiment of the present invention, when the similarity calculation module 103 calculates the similarity between the target user and the outbound user category according to the outbound call feature corresponding to the outbound call information, it is specifically used to:

[0122] Extracting characteristic keywords corresponding to the outbound call characteristics;

[0123] Extracting category keywords corresponding to the outbound user category;

[0124] Select similar feature words between the feature keyword and the category keyword;

[0125] Calculate the similarity between the target user and the outbound user category based on the similar feature words: in, For the target user The outbound user category The similarity between is the number of the characteristic keywords, is the number of keywords in the category, is the number of similar feature words, For the The weight of similar feature words, For the The similarity scores of similar feature words.

[0126] In detail, the keywords extracted from the outbound call information represent the outbound call features, such as conversation content, user interests, and needs; the keywords extracted from the outbound user categories represent the characteristics of the user categories, such as sales, customer service, and survey-related words, and then determine the similar words between the feature keywords and the category keywords. The similar words can be found through word embedding models (such as Word2Vec, GloVe) or methods based on text similarity (such as cosine similarity). Similar feature words refer to keywords that appear in both the target user features and the outbound user category features or keywords with correlation, and then calculate the similarity between the target user and the outbound user category through similar feature keywords.

[0127] Specifically, the similarity Indicates the target user Outbound user category The similarity in the characteristic keywords, This similarity is quantified by calculating the common similar feature words and their weights, where is the sum of the weighted similarity scores of all similar feature words, is the normalization factor, which adjusts the total number of feature keywords and category keywords and the number of similar feature words to avoid repeated calculation bias in the calculation results. The normalization factor helps balance the features and category keywords of different orders of magnitude. If If the target user is higher, it means that the target user and the outbound user category have a higher degree of match in the characteristic keywords, that is, the similarity between the target user and the outbound user category is greater. If the similarity is lower, it means the matching degree is lower, that is, the similarity between the target user and the outbound user category is smaller. Similarity is a quantitative measure used to indicate the similarity between the target user and a specific outbound user category in terms of feature keywords. A high similarity value means that the target user and the outbound user category are more consistent or related in terms of features. Similarity scores of similar feature words It usually indicates the degree of match between outbound call feature keywords and user category keywords. Outbound call feature words and category keywords are converted into word vectors. Word embedding models such as Word2Vec or BERT are usually used to calculate the similarity between word vectors. Commonly used measurement methods include cosine similarity or Euclidean distance. For example, cosine similarity can be obtained by calculating the dot product of two vectors and normalizing them. The similarity score is normalized to a standard range for easy comparison and weighting, while the first The weight of similar feature words It is determined based on the frequency of occurrence of feature words. For example, if the frequency of occurrence of similar feature word A is 3 times, then the weight corresponding to similar feature word A is .

[0128] For example, the feature keywords (outbound call features) include "sales", "promotion", and "customer", and the category keywords (user categories) include "market", "advertisement", and "customer". The similar feature words are "sales" and "market", "customer" and "customer", then m = 3 (the total number of feature keywords), n = 3 (the total number of category keywords), d = 2 (the number of similar feature word pairs found), assuming that the similarity scores of similar feature words are and weight For "sales" and "marketing" , , for "clients" and "customers" , , then we get ,Therefore, the similarity between the target user and the outbound user category is 0.76, and the user category with the highest similarity is selected as the user category to which the target user belongs.

[0129] Furthermore, based on the similarity between the target user and the user category, it can be divided into the user category to which the target user belongs that can be found in the existing category, or the user category to which the target user belongs that cannot be found in the existing user category. Therefore, based on the size of the similarity, the target user's choice of outbound call strategy is divided into two parts, thereby improving the overall efficiency of the outbound call activity.

[0130] The outbound call strategy execution module 104 is used to select an outbound call strategy in the outbound call mapping relationship according to the outbound call information when the similarity is a preset first similarity, and execute an outbound call operation on the intelligent IP phone according to the outbound call strategy.

[0131] In an embodiment of the present invention, the selection of the outbound call strategy in the outbound call mapping relationship refers to matching the target user with the strategy in the mapping relationship according to the category and stage of the target user, thereby selecting the most suitable outbound call action plan to ensure that the strategy is consistent with the user's specific needs and behavioral characteristics to improve the outbound call effect.

[0132] In the embodiment of the present invention, when the outbound call strategy execution module 104 selects the outbound call strategy in the outbound call mapping relationship according to the outbound call information, it is specifically used to:

[0133] Determine the target outbound user category according to the similarity;

[0134] Dividing the target outbound user category into user stages according to the user behavior frequency corresponding to the outbound call information to obtain the target outbound user stage category;

[0135] Performing category mapping between the target outbound user stage category and the outbound user stage category in the outbound call mapping relationship;

[0136] Select an outbound calling strategy based on the mapped target outbound user stage category.

[0137] In detail, when the similarity is the preset first similarity, it means that the target user can find the category to which he belongs in the existing user categories, then the first similarity is 100%, so that the outbound call strategy is selected in the created outbound call mapping relationship, and then the outbound user category to which the target user belongs can be determined according to the similarity, such as sales, customer service, and investigation, and then the users in the target outbound user category are divided into user stages according to the user behavior frequency determined by the number of outbound calls, and divided into potential users, loyal users, and high-potential users, which helps to understand the current status and needs of the user, and map the target user stage category with the user stage category in the outbound call mapping relationship to ensure the applicability of the strategy, and select the appropriate outbound call strategy according to the mapping result, so as to achieve the best communication effect and resource utilization.

[0138] Exemplarily, the outbound calling strategy for potential customers is dialing mode: automatic dialing, dialing frequency: 50 times per hour, outbound calling time: 10 a.m. to 4 p.m. on weekdays, account configuration: account A, voice setting: standard voice synthesis, clear and professional, and speech language: the speech language of potential customers; the dialing mode for loyal customers is dialing mode: automatic dialing, dialing frequency: 30 times per hour, outbound calling time: 10 a.m. to 4 p.m. on weekdays, account configuration: account B, voice setting: standard voice synthesis, friendly and personalized, and speech language: the speech language of loyal customers; the dialing mode for high-potential customers is dialing mode: smart dialing, dialing frequency: dynamically adjusted according to behavioral data, outbound calling time: best answering time (12 noon to 2 p.m., 4 p.m. to 6 p.m.), account configuration: account C, voice setting: high-quality voice synthesis, personalized content, and speech language: the speech language of high-potential customers. If the target user is a potential customer in the sales business, select the outbound calling strategy corresponding to the potential customer.

[0139] Furthermore, the outbound calling operation of intelligent IP phones can effectively improve the execution efficiency and effectiveness of outbound calling strategies. Through automated outbound calling, a large number of customers can be quickly covered, and personalized conversations and recommendations can be conducted according to preset strategies, thereby improving conversion rates and customer satisfaction, making the outbound calling process more accurate and controllable, and being able to record and analyze customer feedback in real time to optimize subsequent outbound calling strategies.

[0140] In the embodiment of the present invention, performing an outbound call operation means that the intelligent IP telephone system proactively dials the target user's phone number according to the set strategy to conduct a predetermined communication or promotion activity, including automatic dialing, playing a preset voice message or connecting to a manual customer service for a conversation. The entire process is intended to achieve specific business goals, such as promotion, investigation or customer service.

[0141] In the embodiment of the present invention, when the outbound call strategy execution module 104 performs an outbound call operation on the intelligent IP phone according to the outbound call strategy, it is specifically used to:

[0142] Obtaining in real time the account status corresponding to the outbound call allocation account in the outbound call strategy;

[0143] Calculate the number of outbound calls corresponding to the smart IP phone based on the account status: in, The number of outbound calls issued, is the number of accounts corresponding to the idle state in the account state, is the number of accounts corresponding to the connected state in the account status, is the total number of calls, is the number of successful calls, is the dialing speed adjustment factor, The number of accounts corresponding to the occupied state in the account status;

[0144] Allocating the outbound call strategy to the intelligent IP phone according to the number of outbound calls issued;

[0145] Make outbound calls to target users according to the assigned outbound calling strategy.

[0146] In detail, the current status of the outbound call distribution account is checked in real time, including idle, connected and occupied status, and the number of outbound calls to be made is determined based on the account status. The number of target users is more than one, so the number of outbound calls to be made based on this outbound call strategy needs to be determined. The account status, call success rate, dialing speed adjustment factor and the number of accounts in the occupied state are combined to calculate the appropriate outbound call volume, and the dialing speed adjustment factor Used to adjust the dialing speed to suit different working environments or needs. Indicates the total number of accounts that can currently make outbound calls, including accounts in idle and connected states. Indicates the calculation of the success rate of outbound calls, that is, the ratio of the total number of calls to the number of successful calls, and the dialing speed adjustment coefficient , in order to adjust the outbound call speed, the required number of outbound calls is calculated by comprehensively considering the current account status, successful call ratio, dialing speed and occupied status , thereby deciding how many outbound call requests should be made under current conditions to optimize outbound call efficiency and effectiveness.

[0147] Specifically, effectively use the calculated outbound call volume according to the outbound call strategy To allocate and manage outbound call tasks, the outbound call strategy is allocated to the same number of intelligent IP phones based on the number of outbound calls issued, and the calculated number of outbound calls is Assign to different accounts according to the strategy, schedule tasks according to the account status (idle, connected, occupied), ensure that outbound call tasks are assigned according to priority and status, and then perform actual outbound call operations according to the assigned outbound call strategy. Initialize outbound call tasks for each assigned account, including dialing, connecting to target users and other steps, to realize the outbound call operation of the intelligent IP phone.

[0148] Furthermore, there may be a user category to which the target user belongs that is not found in the existing user categories. Therefore, the intent of the user category to which the target user belongs that is not found is identified based on similarity to ensure that more accurate services are provided or a more effective outbound calling strategy is formulated.

[0149] The outbound call intention strategy execution module 105 is used to identify the outbound call intention of the target user according to the outbound call information when the similarity is a preset second similarity, generate an outbound call intention strategy according to the outbound call event type corresponding to the outbound call intention, and perform an outbound call operation on the smart IP phone according to the outbound call intention strategy.

[0150] In an embodiment of the present invention, when the similarity is a preset second similarity, it means that there is no outbound call strategy matching the current user in the outbound call strategy, and the user's intention needs to be identified in real time, then the second similarity is 0, wherein the outbound call intention refers to the purpose or demand expressed by the user during the outbound call process, for example, it may include seeking help, understanding product information, making complaints, making appointments for services, etc. By identifying the outbound call intention, it is possible to respond to user needs more effectively and provide appropriate services.

[0151] In the embodiment of the present invention, when the outbound calling intention strategy execution module 105 identifies the outbound calling intention of the target user according to the outbound calling information, it is specifically used to:

[0152] Extracting outbound call features corresponding to the outbound call information;

[0153] Determining an outbound call keyword of the target user according to the outbound call feature;

[0154] Identify the high-level intention and low-level intention corresponding to the target user according to the outbound call keyword;

[0155] The intention probability of the target user is calculated by the high-level intention and the low-level intention: in, For low level intention Given a high-level intent The probability of intention under the condition of is the joint probability of high-level intention and low-level intention occurring simultaneously, is the marginal probability of high-level intention;

[0156] The outbound calling intention of the target user is identified according to the intention probability.

[0157] In detail, features such as keywords and context are extracted from outbound call conversations or messages. Voice features: extract features such as pitch, speaking speed, and tone from the conversation, which can help understand the user's emotions or urgency; keywords: identify words that appear repeatedly or are particularly emphasized in the conversation, which may indicate the user's main needs or concerns; contextual information: analyze the context of the conversation, such as the specific problems and background conditions mentioned by the user, to understand the user's actual needs; intent tagging: if the dialogue system has the ability to recognize intent, it can tag the user's intent category (such as query, complaint, purchase, etc.), and can use natural language processing (NLP) technology to identify the target user's outbound keywords, including TF-IDF (term frequency-inverse document frequency) and word embedding (such as Word2Vec or BERT).

[0158] Specifically, high-level intent is usually the main demand or purpose expressed by the user in the conversation. Based on the extracted outbound call keywords (such as "bill", "payment problem"), the high-level intent can be inferred, and the low-level intent refers to the specific action needs or problem details. By analyzing the context and keywords in the conversation, the low-level intent can be further determined. For example, if the user mentions "the bill is wrong" and "payment failed" in the conversation, the high-level intent is that the user cares about the accuracy of the bill and payment issues, and the low-level intent is that the bill is wrong, the user needs to correct the bill or inquire about the bill problem or payment failed, and the user needs to solve the payment failure problem, which may include technical support or refund requests. The high-level intent and low-level intent can be identified through intent recognition models (such as naive Bayes, recurrent neural networks, etc.), and then the probability of the low-level intent occurring under a specific high-level intent is calculated. According to the pre-acquired data set, which contains pairs of high-level intent and low-level intent, the joint probability can be obtained by calculating the frequency of occurrence of these pairs. , ,in It is an event and events The number of simultaneous occurrences (frequency of joint occurrence), is the number of keywords in the total events, It is an event The number of times it occurs at the same time, such as P(account inquiry, customer service request) = 0.3, P(customer service request) = 0.6, then among all "customer service requests", 50% of the requests are "account inquiry", then the intent corresponding to the largest intent probability is selected as the outbound call intent of the target user.

[0159] Furthermore, by identifying outbound call intentions, the content and target of outbound calls can be accurately matched to improve the relevance and efficiency of communication. Strategies targeting different outbound call intentions (such as promotion, collection, customer feedback, etc.) can be formulated to improve the success rate of outbound calls and customer satisfaction.

[0160] In the embodiment of the present invention, the outbound call intention strategy refers to the formulation of specific plans and rules for outbound call activities, including determining the goals and intentions of the outbound calls, writing targeted outbound call scripts, setting dialing rules and times, and allocating execution accounts. Ultimately, these strategies are combined into a systematic solution to optimize the efficiency and effectiveness of outbound calls.

[0161] In the embodiment of the present invention, when the outbound call intention strategy execution module 105 generates the outbound call intention strategy according to the outbound call event type corresponding to the outbound call intention, it is specifically used to:

[0162] Determine an outbound call rule and an outbound call time according to the outbound call event type corresponding to the outbound call intention;

[0163] Generate an outbound call intention script according to the outbound call intention;

[0164] Adding the outbound call intention script to a pre-configured outbound call distribution account;

[0165] The outbound call rules, the outbound call time, the outbound call allocation account and the allocated outbound call intention words are combined into an outbound call intention strategy.

[0166] In detail, formulate specific rules for outbound calls based on outbound call intent and event type. These rules may include the frequency of calls (e.g. daily, weekly), the time period of calls (e.g. 9am to 5pm), the priority of calls (e.g. high priority calls first), etc. Set the time of outbound calls to ensure that outbound calls are made during the best time period. For example, some outbound call activities may need to be made during the customer's working hours, while other activities can be carried out during the customer's free time.

[0167] Specifically, design and write the content and language style of the outbound call according to the outbound call intention. This includes greetings, introductory words, main information points, call purpose, and how to handle possible customer responses. The design of the script should match the goal of the outbound call, such as sales promotion, customer surveys, or service notifications. Then, configure the written outbound script to the pre-set outbound call allocation account. These accounts are usually managed by the outbound call system or call center for specific outbound call activities. After adding the script to the account, the outbound call task can be automatically executed through the system. Integrate all the above elements (outbound call rules, outbound call time, allocated accounts, scripts) into a complete outbound call intention strategy. The outbound call intention strategy will guide the execution of the entire outbound call activity, ensure that all links are coordinated and consistent, achieve the expected business goals, and ensure that the outbound call activity can be executed efficiently and in a targeted manner, thereby improving the success rate and effectiveness of the outbound call.

[0168] Furthermore, the steps of performing an outbound call operation on the intelligent IP phone according to the outbound call intention strategy are consistent with the steps of performing an outbound call operation on the intelligent IP phone according to the outbound call strategy, which will not be repeated here.

[0169] The embodiment of the present invention extracts user information and service types through historical records, and can make analysis and decisions based on real data; classified users can help formulate more accurate outbound call strategies, ensuring that each user group can get the most suitable communication method and content; outbound call parameters are generated according to user classification and service type, making the outbound call strategy more personalized, thereby improving response rate and customer satisfaction; by calculating similarity, the matching degree between the target user and the existing user category can be determined, so as to select the most appropriate outbound call strategy; the most appropriate outbound call strategy is selected according to the similarity, and the outbound call operation is executed according to the strategy, ensuring the efficiency of the outbound call process. Therefore, the intelligent IP phone outbound call cloud system, method and medium proposed by the present invention can solve the problem of low accuracy when making outbound calls by intelligent IP phones.

[0170] Reference Figure 2 FIG. 1 is a flow chart of an operation method of an intelligent IP phone outbound call cloud system provided by an embodiment of the present invention. In this embodiment, the operation method of the intelligent IP phone outbound call cloud system includes:

[0171] S1. Extracting outbound user information and outbound service type based on the pre-acquired historical outbound call records, and classifying the outbound users according to the outbound user information to obtain the outbound user category;

[0172] S2, generating an outbound call parameter according to the outbound call service type, generating an outbound call strategy for the smart IP phone according to the outbound call parameter and an outbound call account pre-configured on the cloud service platform, mapping the outbound call strategy with the outbound call user category to obtain an outbound call mapping relationship;

[0173] S3, obtaining the outbound call information of the target user, and calculating the similarity between the target user and the outbound call user category according to the outbound call features corresponding to the outbound call information;

[0174] S4. When the similarity is a preset first similarity, selecting an outbound call strategy in the outbound call mapping relationship according to the outbound call information, and performing an outbound call operation on the intelligent IP phone according to the outbound call strategy;

[0175] S5. When the similarity is a preset second similarity, the outbound calling intention of the target user is identified according to the outbound calling information, an outbound calling intention strategy is generated according to the outbound calling event type corresponding to the outbound calling intention, and an outbound calling operation is performed on the smart IP phone according to the outbound calling intention strategy.

[0176] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0177] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0178] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0179] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0180] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is not limited only according to the above description, and it is intended that all changes within the meaning and scope of equivalent elements within the scope of protection are included in the present invention.

[0181] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems stated in a system claim can also be implemented by one unit or system through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0182] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An intelligent IP phone outbound call cloud system, characterized in that: The system includes an outbound user classification module, an outbound call strategy mapping module, a similarity calculation module, an outbound call strategy execution module and a telephone outbound call intention strategy execution module, wherein: The outbound user classification module is used to extract outbound user information and outbound service types based on the pre-acquired historical outbound call records, classify outbound users according to the outbound user information, and obtain outbound user categories; The outbound call strategy mapping module is used to generate outbound call parameters according to the outbound call service type, generate an outbound call strategy for the intelligent IP phone according to the outbound call parameters and the outbound call account pre-configured on the cloud service platform, map the outbound call strategy with the outbound call user category, and obtain an outbound call mapping relationship, which is specifically used to: divide the outbound call users in the outbound call user category into customer stages according to the behavior frequency of each outbound call user to obtain an outbound call user stage category; generate an outbound call label corresponding to each outbound call user according to the outbound call user stage category and the outbound call service type; generate a policy label corresponding to the external policy according to the service type corresponding to the outbound call strategy and the outbound call user stage category; calculate the label correlation coefficient between the outbound call label and the policy label: Among them, r ij is the label correlation coefficient between the i-th outbound call label and the j-th strategy label, x i is the outbound user stage category label in the i-th outbound call label, y i is the service type label in the i-th outbound call label, u j is the outbound user stage category label in the jth strategy label, v j is the business type label in the j-th policy label, and min is the minimum value function; Mapping the outbound call strategy with the outbound call user category according to the label correlation coefficient to obtain an outbound call mapping relationship; The similarity calculation module is used to obtain the outbound call information of the target user, and calculate the similarity between the target user and the outbound call user category according to the outbound call features corresponding to the outbound call information; The outbound call strategy execution module is used to select the outbound call strategy in the outbound call mapping relationship according to the outbound call information when the similarity is a preset first similarity, and perform an outbound call operation on the intelligent IP phone according to the outbound call strategy; The outbound call intention strategy execution module is used to identify the outbound call intention of the target user according to the outbound call information when the similarity is a preset second similarity, generate an outbound call intention strategy according to the outbound call event type corresponding to the outbound call intention, and perform an outbound call operation on the smart IP phone according to the outbound call intention strategy.

2. The intelligent IP phone outbound call cloud system according to claim 1, characterized in that: When the outbound user classification module classifies the outbound user according to the outbound user information to obtain the outbound user category, it is specifically used to: Extracting behavior-driven data corresponding to each outbound calling user in the outbound calling user information; Counting the behavior frequency of each outbound calling user according to the behavior driving data; determining a behavior category of each outbound calling user according to the behavior frequency; The outbound user is classified according to the behavior category to obtain an outbound user category.

3. The intelligent IP phone outbound call cloud system according to claim 2, characterized in that: When the outbound call policy mapping module generates outbound call parameters according to the outbound call service type, it is specifically used to: Determine the outbound calling skills and the outbound calling mode according to the outbound calling service type; Calculate the best time for outbound calls based on the frequency of the behavior; The optimal time for outbound calls, the service outbound call mode and the service outbound call speech are encapsulated as outbound call parameters.

4. The intelligent IP phone outbound call cloud system according to claim 1, characterized in that: When the outbound call strategy mapping module generates an outbound call strategy for the intelligent IP phone according to the outbound call parameters and the pre-configured outbound call account on the cloud service platform, it is specifically used to: Determine an outbound call rule according to the business outbound call mode in the outbound call parameter; Determine the outbound call time according to the outbound call best time in the outbound call parameters; Allocate outbound calls to the outbound call account according to the preset outbound call task to obtain an outbound call allocation account; Adding the business outbound call script in the outbound call parameters to the outbound call allocation account; The outbound call rules, the outbound call time, the outbound call allocation account and the allocated business outbound call speech are combined into an outbound call strategy for the intelligent IP phone.

5. The intelligent IP phone outbound call cloud system according to claim 1, characterized in that: When the similarity calculation module calculates the similarity between the target user and the outbound user category according to the outbound call feature corresponding to the outbound call information, it is specifically used to: Extracting characteristic keywords corresponding to the outbound call characteristics; Extracting category keywords corresponding to the outbound user category; Select similar feature words between the feature keyword and the category keyword; Calculate the similarity between the target user and the outbound user category based on the similar feature words: Among them, S(T1, T2) is the similarity between the target user T1 and the outbound user category T2, m is the number of the characteristic keywords, n is the number of the category keywords, d is the number of the similar characteristic words, λ k is the weight of the kth similar feature word, Q k is the similarity score of the kth similar feature word.

6. The intelligent IP phone outbound call cloud system according to claim 5, characterized in that: When the outbound call strategy execution module selects the outbound call strategy in the outbound call mapping relationship according to the outbound call information, it is specifically used to: Determine the target outbound user category according to the similarity; Dividing the target outbound user category into user stages according to the user behavior frequency corresponding to the outbound call information to obtain the target outbound user stage category; Performing category mapping between the target outbound user stage category and the outbound user stage category in the outbound call mapping relationship; Select an outbound calling strategy based on the mapped target outbound user stage category.

7. The intelligent IP phone outbound call cloud system according to claim 1, characterized in that: When the outbound call strategy execution module performs an outbound call operation on the intelligent IP phone according to the outbound call strategy, it is specifically used to: Obtaining in real time the account status corresponding to the outbound call allocation account in the outbound call strategy; Calculate the number of outbound calls corresponding to the smart IP phone based on the account status: Wherein, N is the number of outbound calls, R is the number of accounts corresponding to the idle state in the account status, E is the number of accounts corresponding to the connected state in the account status, G1 is the total number of calls, G2 is the number of successful calls, δ is the dialing speed adjustment coefficient, The number of accounts corresponding to the occupied state in the account status; Allocating the outbound call strategy to the intelligent IP phone according to the number of outbound calls issued; Make outbound calls to target users according to the assigned outbound calling strategy.

8. The intelligent IP phone outbound call cloud system according to claim 1, characterized in that: When the outbound calling intention strategy execution module identifies the outbound calling intention of the target user according to the outbound calling information, it is specifically used to: Extracting outbound call features corresponding to the outbound call information; Determining an outbound call keyword of the target user according to the outbound call feature; Identify the high-level intention and low-level intention corresponding to the target user according to the outbound call keyword; The intention probability of the target user is calculated by the high-level intention and the low-level intention: Among them, P(L|H) is the intention probability of the low-level intention L given the high-level intention H, P(L,H) is the joint probability of the high-level intention and the low-level intention occurring at the same time, and P(H) is the marginal probability of the high-level intention; The outbound calling intention of the target user is identified according to the intention probability.

9. The intelligent IP phone outbound call cloud system according to claim 1, characterized in that: When the outbound call intention strategy execution module generates the outbound call intention strategy according to the outbound call event type corresponding to the outbound call intention, it is specifically used to: Determine an outbound call rule and an outbound call time according to the outbound call event type corresponding to the outbound call intention; Generate an outbound call intention script according to the outbound call intention; Adding the outbound call intention script to a pre-configured outbound call distribution account; The outbound call rules, the outbound call time, the outbound call allocation account and the allocated outbound call intention words are combined into an outbound call intention strategy.

10. An operating method of an intelligent IP phone outbound call cloud system, applied to the intelligent IP phone outbound call cloud system according to any one of claims 1 to 9, the method comprising: Extracting outbound user information and outbound service types based on the pre-acquired historical outbound call records, and classifying outbound users according to the outbound user information to obtain outbound user categories; Generate an outbound call parameter according to the outbound call service type, generate an outbound call strategy for the smart IP phone according to the outbound call parameter and the outbound call account pre-configured on the cloud service platform, map the outbound call strategy with the outbound call user category, and obtain an outbound call mapping relationship, which is specifically used for: dividing the outbound call users in the outbound call user category into customer stages according to the behavior frequency of each outbound call user to obtain an outbound call user stage category; generating an outbound call tag corresponding to each outbound call user according to the outbound call user stage category and the outbound call service type; generating a strategy tag corresponding to the external strategy according to the service type corresponding to the outbound call strategy and the outbound call user stage category; and calculating the tag correlation coefficient between the outbound call tag and the strategy tag: Among them, r ij is the label correlation coefficient between the i-th outbound call label and the j-th strategy label, x i is the outbound user stage category label in the i-th outbound call label, y i is the service type label in the i-th outbound call label, u j is the outbound user stage category label in the jth strategy label, v j is the business type label in the j-th policy label, and min is the minimum value function; Mapping the outbound call strategy with the outbound call user category according to the label correlation coefficient to obtain an outbound call mapping relationship; Acquire outbound call information of a target user, and calculate the similarity between the target user and the outbound call user category according to outbound call features corresponding to the outbound call information; When the similarity is a preset first similarity, selecting an outbound call strategy in the outbound call mapping relationship according to the outbound call information, and performing an outbound call operation on the intelligent IP phone according to the outbound call strategy; When the similarity is a preset second similarity, the outbound calling intention of the target user is identified according to the outbound calling information, an outbound calling intention strategy is generated according to the outbound calling event type corresponding to the outbound calling intention, and an outbound calling operation is performed on the smart IP phone according to the outbound calling intention strategy.

Images