Outbound communication method and device for vehicle insurance information and storage medium

Abstract

The invention discloses an outbound communication method and device for vehicle insurance information and a storage medium, and mainly aims at improving the communication efficiency, reducing the labor cost and improving the communication effect. Comprising the following steps: dividing a car insurance call-out communication process in a call-out conversation robot into a plurality of call-out communication modules according to a process sequence; obtaining current communication information after the current call-out communication module is scheduled to perform call-out communication on the to-be-recommended user; based on the current communication information, judging whether the car insurance outbound communication process meets a preset condition of continuing to advance, and if so, automatically analyzing the current communication information to obtain the next round outbound communication time and the next round outbound communication skill; and based on the next round of call-out communication time, automatically scheduling the next call-out communication module corresponding to the current call-out communication module to carry out car insurance information call-out communication with the to-be-recommended user according to the next round of call-out communication skill, and obtaining a call-out communication result of the car insurance call-out communication process.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile insurance information processing, and in particular to an outbound communication method, device and storage medium for automobile insurance information. Background Art

[0002] With the continuous development of society, the number of vehicles is increasing, and the demand for auto insurance is also increasing. Before the vehicle is insured, effective communication based on auto insurance information is required.

[0003] At present, the communication of auto insurance information is usually driven manually. However, the manual drive method requires manual determination of information such as the communication time, resulting in low communication efficiency and high labor costs. In addition, due to negligence of the staff, the communication time may be incorrectly determined, thus affecting the communication effect. Summary of the invention

[0004] The present invention provides an outbound call communication method, device and storage medium for automobile insurance information, which are mainly capable of improving the outbound call communication efficiency for automobile insurance information, reducing labor costs and improving communication effects.

[0005] According to a first aspect of the present invention, there is provided an outbound communication method for automobile insurance information, comprising:

[0006] Divide the auto insurance outbound call communication process in the outbound call conversation robot into multiple outbound call communication modules according to the process sequence;

[0007] Obtain the current communication information after the outbound call communication module schedules the outbound call communication for the recommended user;

[0008] Based on the current communication information, determine whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance. If so, automatically analyze the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication script;

[0009] Based on the next round of outbound call communication time, the next outbound call communication module corresponding to the current outbound call communication module is automatically scheduled to conduct outbound call communication on auto insurance information with the to-be-recommended user according to the next round of outbound call communication script, and the outbound communication result of the auto insurance outbound call communication process is obtained.

[0010] Optionally, each of the outbound call communication modules includes: a first dial module, a reservation dial module, an underwriting dial module, a promotion dial module, and a communication module, and the current outbound call communication module is any one of the first dial module, the reservation dial module, the underwriting dial module, and the promotion dial module;

[0011] If the current outbound call communication module is the first call module, obtaining the current communication information after scheduling the current outbound call communication module to perform outbound call communication for the recommended user includes:

[0012] In response to a first round of outbound call communication signals for recommending auto insurance information to the user to be recommended, obtaining user feature data and vehicle attribute data of the user to be recommended;

[0013] Based on the user characteristic data and the vehicle attribute data, the first round of outbound call communication words are determined, and the first call module is scheduled to conduct outbound communication on auto insurance information with the recommended user according to the first round of outbound call communication words, so as to obtain the first round of communication information of the auto insurance outbound call communication process, and the first round of communication information is determined as the current communication information.

[0014] Optionally, the method further comprises:

[0015] In the process of scheduling the next outbound call communication module to communicate with the to-be-recommended user about the car insurance information, generating a car insurance demand inquiry speech in response to the car insurance value signal, and scheduling the next outbound call communication module to communicate with the to-be-recommended user about the car insurance demand according to the car insurance demand inquiry speech;

[0016] receiving feedback auto insurance demand information returned by the to-be-recommended user in response to the auto insurance demand inquiry dialogue, and generating auto insurance value call parameters based on the feedback auto insurance demand information;

[0017] Based on the car insurance value calling parameter, the car insurance value is retrieved in the value interface, and the next outbound call communication module is scheduled to transmit the car insurance value to the user terminal to be recommended.

[0018] Optionally, the current communication information includes communication tone information and communication content information of the user to be recommended;

[0019] The automatic analysis of the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication words includes:

[0020] Determine the first vehicle maintenance information of the user to be recommended;

[0021] Determine a first maintenance feature vector corresponding to the vehicle first maintenance information, a tone feature vector corresponding to the communication tone information, and a content feature vector corresponding to the communication content information;

[0022] Cross-processing the first protection feature vector, the tone feature vector, and the content feature vector to obtain a communication cross feature vector;

[0023] The communication cross feature vector is input into the large model for information prediction to obtain the next round of outbound call communication time and the next round of outbound call communication words.

[0024] Optionally, after automatically scheduling the next outbound call communication module corresponding to the current outbound call communication module to perform outbound communication on car insurance information with the to-be-recommended user according to the next round of outbound call communication words and obtaining the outbound communication result of the car insurance outbound call communication process, the method further includes:

[0025] If the next outbound call communication module is the order promotion dialing module, then based on the outbound call communication result corresponding to the order promotion dialing module, it is determined whether the auto insurance outbound call communication process is completed according to the preset result. If so, the communication module is dispatched to send a conversion link of auto insurance information to the user to be recommended.

[0026] Optionally, after scheduling the communication module to send a conversion link of the auto insurance information to the to-be-recommended user, the method further includes:

[0027] After a preset time of sending the conversion link, determining whether feedback information of the to-be-recommended user regarding the conversion link is received;

[0028] If the feedback information is not received, determining the operation process information of the conversion link, and scheduling the promotion and order dialing module to make an outbound call communication with the to-be-recommended user based on the operation process information;

[0029] If the feedback information is received, conversion credential information corresponding to the feedback information is generated, and the communication module is scheduled to send the conversion credential information to the user to be recommended.

[0030] Optionally, before determining the first round of outbound call communication skills based on the user feature data and the vehicle attribute data, the method further includes:

[0031] Taking any data in the user feature data and the vehicle attribute data as target data, respectively, determining the data density in a preset neighborhood corresponding to each sub-data in the target data, and determining core data points and non-core data points in each sub-data based on the data density;

[0032] Taking any core data point among the core data points as a target core data point, taking the target core data point as the cluster center, clustering the remaining core data points to obtain core data points under different clustering categories;

[0033] Taking any non-core data point among the non-core data points as a target non-core data point, determining a reference core data point that is closest to the non-core data point among the core data points under each clustering category, and judging whether the distance between the non-core data point and the reference core data point is less than a preset distance threshold;

[0034] If the distance between the non-core data point and the reference core data point is less than the preset distance threshold, the non-core data point is classified into the cluster category to which the reference core data point belongs; otherwise, the non-core data point is determined as abnormal data;

[0035] Abnormal data in the user characteristic data and the vehicle attribute data are respectively removed to obtain cleaned user characteristic data and cleaned vehicle attribute data.

[0036] According to a second aspect of the present invention, there is provided an outbound communication device for automobile insurance information, comprising:

[0037] A division unit, used to divide the auto insurance outbound call communication process in the outbound call dialogue robot into multiple outbound call communication modules according to the process sequence;

[0038] An acquisition unit, used for acquiring current communication information after the outbound call communication module is scheduled to perform outbound call communication with the user to be recommended;

[0039] A judgment unit, configured to judge whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance based on the current communication information, and if so, automatically parse the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication words;

[0040] The scheduling unit is used to automatically schedule the next outbound call communication module corresponding to the current outbound call communication module based on the next round of outbound call communication time to conduct outbound communication on auto insurance information with the recommended user according to the next round of outbound call communication script, and obtain the outbound communication result of the auto insurance outbound call communication process.

[0041] According to a third aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned outbound call communication method for vehicle insurance information.

[0042] According to a fourth aspect of the present invention, there is provided a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned outbound call communication method for vehicle insurance information when executing the program.

[0043] According to a method, device and storage medium for outbound communication of auto insurance information provided by the present invention, compared with the current communication method of auto insurance information driven manually, the present invention divides the auto insurance outbound communication process in the outbound dialogue robot into multiple outbound communication modules according to the process sequence; and obtains the current communication information after scheduling the current outbound communication module to conduct outbound communication with the user to be recommended; then, based on the current communication information, it is judged whether the auto insurance outbound communication process meets the preset conditions for continuing to advance, and if so, the current communication information is automatically parsed to obtain the next round of outbound communication time and the next round of outbound communication words; finally, based on the next round of outbound communication time, the next outbound communication module corresponding to the current outbound communication module is automatically scheduled to conduct outbound communication on auto insurance information with the user to be recommended according to the next round of outbound communication words, and the outbound communication result of the auto insurance outbound communication process is obtained. Since the communication of auto insurance information is a long process and has a long time span, the auto insurance outbound communication process in the outbound dialogue robot is divided into multiple outbound communication modules according to the process sequence, and outbound communication is performed to the user based on each outbound communication module. Each outbound communication module only needs to narrate the corresponding personalized words, and the communication logic is clear, avoiding the problem of stringing processes, thereby improving the outbound communication effect. At the same time, the embodiment of the present invention automatically determines the appropriate communication words and the appropriate communication time through the various communication modules of the outbound robot, and automatically realizes the outbound communication to the user. The whole process is automated without manual participation, thereby improving the outbound communication efficiency and saving labor costs. It can also avoid the situation where manual negligence leads to incorrect determination of communication time and communication content. Therefore, the present invention can improve the accuracy of outbound communication on auto insurance information and enhance user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0045] Figure 1 A flow chart of an outbound communication method for vehicle insurance information provided by an embodiment of the present invention is shown;

[0046] Figure 2 A functional schematic diagram of a large insurance information processing model provided by an embodiment of the present invention is shown;

[0047] Figure 3 A flow chart of another outbound communication method for vehicle insurance information provided by an embodiment of the present invention is shown;

[0048] Figure 4 A schematic diagram of the structure of an outbound communication device for car insurance information provided by an embodiment of the present invention is shown;

[0049] Figure 5 A schematic diagram showing the structure of another outbound communication device for vehicle insurance information provided by an embodiment of the present invention;

[0050] Figure 6 A schematic diagram of the physical structure of a computer device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0051] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0052] At present, the manual outbound communication method for car insurance information leads to low communication efficiency and high labor costs. In addition, due to the negligence of the staff, the communication time may be incorrectly determined, thus affecting the communication effect.

[0053] In order to solve the above problems, the embodiment of the present invention provides an outbound communication method for vehicle insurance information, such as Figure 1 As shown, the method includes:

[0054] 101. Divide the auto insurance outbound communication process in the outbound dialogue robot into multiple outbound communication modules according to the process sequence.

[0055] Among them, multiple outbound call communication modules include: first call module, appointment call module, underwriting call module, promotion call module, communication module, etc. Each module has its own positioning and function, which may be one or more outbound call conversations. Among them, the outbound call dialogue robot also includes a scheduling module. The scheduling module is used to realize round-free. Whether each round of calls ends or the communication module communication ends, the scheduling module will issue the next action, such as determining the time of the next call, or the communication module sending a message, etc., and the next module will be scheduled accordingly according to the scheduling conditions. The entire multi-round outbound call communication process relies on the big model.

[0056] For the embodiment of the present invention, since the communication process of auto insurance information is a process with a long time span, it is difficult for a single module to handle the long-term communication process. Therefore, the auto insurance outbound communication process in the outbound dialogue robot is divided into multiple outbound communication modules according to the process sequence. Each outbound communication module only needs to conduct personalized communication based on its own process and communication content, avoiding the problem of communication string process. Not only is the logic clear, but it can also improve the professionalism and efficiency of communication and enhance the user experience. The entire communication process provided by the embodiment of the present invention is implemented based on the insurance information processing big model. The functions of the insurance information processing big model are as follows: Figure 2 shown.

[0057] 102. Obtain current communication information after the outbound call communication module schedules the outbound call communication for the user to be recommended.

[0058] For the embodiment of the present invention, the current outbound call communication module can be any one of the communication modules of the first dial module, the scheduled dial module, the underwriting dial module, and the order promotion dial module, and the communication information includes the user's tone information, communication content information, communication duration, etc. If the current outbound call communication module is the first dial module, the current communication information is the communication information generated during the communication process between the scheduling first dial module and the user and after the communication; if the current outbound call communication module is the scheduled dial module, the current communication information is the communication information generated during the communication process between the scheduling scheduled dial module and the user and after the communication; if the current outbound call communication module is the underwriting dial module, the current communication information is the communication information generated during the communication process between the scheduling underwriting dial module and the user and after the communication; if the current outbound call communication module is the order promotion dial module, the current communication information is the communication information generated during the communication process between the scheduling order promotion dial module and the user and after the communication.

[0059] 103. Based on the current communication information, determine whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance. If so, automatically parse the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication script.

[0060] 104. Based on the next round of outbound call communication time, automatically schedule the next outbound call communication module corresponding to the current outbound call communication module to conduct outbound communication on auto insurance information with the recommended user according to the next round of outbound call communication script, and obtain the outbound communication result of the auto insurance outbound call communication process.

[0061] For the embodiment of the present invention, based on the communication content information in the current communication information, it can be determined whether the user wants to continue the communication next time. If the user wants to continue the communication next time, it is determined that the auto insurance outbound communication process meets the preset conditions for continuing to advance. At this time, the current communication information is automatically parsed using natural language processing and other technologies to automatically parse out the next round of outbound communication time and the next round of outbound communication words in the current communication information. In another embodiment of the present invention, the basic communication words are pre-stored in the outbound communication module corresponding to the next round, and the communication words parsed in the current communication information are integrated with the basic communication words to obtain the next round of outbound communication words. Then, when the next round of outbound communication time arrives, the scheduling module is used to schedule the next outbound communication module corresponding to the current outbound communication module to conduct outbound communication on auto insurance information with the recommended user according to the next round of outbound communication words, and obtain the outbound communication result of the auto insurance outbound communication process. The embodiment of the present invention automatically determines appropriate communication words and appropriate communication time through various communication modules of the outbound call robot, and automatically implements outbound communication with users. The entire process is automated without the need for human intervention, thereby improving the efficiency of outbound communication and saving labor costs. It can also avoid situations where human negligence leads to incorrect determination of communication time and communication content. Therefore, the present invention can improve the accuracy of outbound communication about auto insurance information and enhance user experience.

[0062] According to a method for outbound communication for auto insurance information provided by the present invention, compared with the current communication method of auto insurance information driven manually, the present invention divides the auto insurance outbound communication process in the outbound dialogue robot into multiple outbound communication modules according to the process sequence; and obtains the current communication information after scheduling the current outbound communication module to conduct outbound communication with the user to be recommended; then, based on the current communication information, it is determined whether the auto insurance outbound communication process meets the preset conditions for continuing to advance, and if so, the current communication information is automatically parsed to obtain the next round of outbound communication time and the next round of outbound communication words; finally, based on the next round of outbound communication time, the next outbound communication module corresponding to the current outbound communication module is automatically scheduled to conduct outbound communication on auto insurance information with the user to be recommended according to the next round of outbound communication words, and the outbound communication result of the auto insurance outbound communication process is obtained. Since the communication of auto insurance information is a long process and has a long time span, the auto insurance outbound communication process in the outbound dialogue robot is divided into multiple outbound communication modules according to the process sequence, and outbound communication is performed to the user based on each outbound communication module. Each outbound communication module only needs to narrate the corresponding personalized words, and the communication logic is clear, avoiding the problem of stringing processes, thereby improving the outbound communication effect. At the same time, the embodiment of the present invention automatically determines the appropriate communication words and the appropriate communication time through the various communication modules of the outbound robot, and automatically realizes the outbound communication to the user. The whole process is automated without manual participation, thereby improving the outbound communication efficiency and saving labor costs. It can also avoid the situation where manual negligence leads to incorrect determination of communication time and communication content. Therefore, the present invention can improve the accuracy of outbound communication on auto insurance information and enhance user experience.

[0063] Further, in order to better illustrate the above process of communicating the vehicle insurance information, as a refinement and extension of the above embodiment, the embodiment of the present invention provides another outbound communication method for vehicle insurance information, such as Figure 3 As shown, the method includes:

[0064] 201. Divide the auto insurance outbound call communication process in the outbound call dialogue robot into multiple outbound call communication modules according to the process sequence.

[0065] Among them, multiple outbound call communication modules include: first dial module, appointment dial module, underwriting dial module, order promotion dial module, and communication module. The communication module can realize functions such as text and picture transmission.

[0066] Specifically, the auto insurance outbound call communication process in the outbound dialogue robot is divided into the first call module, the appointment call module, the underwriting call module, the order promotion call module, the communication module, etc. according to the process sequence, which also includes a scheduling module for realizing the scheduling between various communication modules.

[0067] 202. In response to a first round of outbound call communication signals for recommending auto insurance information to a user to be recommended, obtain user feature data and vehicle attribute data of the user to be recommended.

[0068] Among them, user characteristic data include: gender, age, occupation, interests and hobbies, income, driving experience and other data; vehicle attribute data include vehicle first maintenance date, vehicle usage information, vehicle driving condition, license plate number, frame number, previously insured insurance coverage, previous violation information, etc.

[0069] Specifically, when it is necessary to communicate auto insurance information with the user to be recommended, the communication system sends a communication instruction to the outbound dialogue robot. If it is the first communication, it is necessary to send the first communication instruction, that is, the first round of outbound communication signal, to the first-call module within a preset time range (the preset time range is set according to actual needs) before the first insurance date of the user's vehicle. When the first-call module receives the first-round outbound communication signal, the scheduling module schedules the first-call module to obtain the user characteristic data and vehicle attribute data of the user to be recommended, and determines the first round of communication words based on the user characteristic data and the vehicle attribute data. In order to improve the accuracy of determining the first round of communication words, the predicted communication words can also be determined based on the user characteristic data and the vehicle attribute data, and then the predicted communication words are integrated with the basic communication words pre-stored in the first-call module to obtain the first round of outbound communication words. Then the scheduling module schedules the first-call module to conduct the first round of communication on vehicle insurance information with the user to be recommended according to the first round of outbound communication words. It should be noted that the user feature data and vehicle attribute data obtained in this embodiment are not user privacy data, but data related to the communication of auto insurance information with the recommended user that can be obtained in the auto insurance information communication platform (non-privacy data).

[0070] 203. Based on the user feature data and vehicle attribute data, determine the first round of outbound call communication words, and schedule the first call module to conduct outbound communication on auto insurance information with the recommended user according to the first round of outbound call communication words, obtain the first round of communication information of the auto insurance outbound call communication process, and determine the first round of communication information as the current communication information.

[0071] For the embodiment of the present invention, after obtaining user feature data and vehicle attribute data, in order to improve data quality and thereby improve the accuracy of determining the first round of communication words, it is necessary to pre-process the user feature data and vehicle attribute data. Based on this, the method includes: taking any data in the user feature data and the vehicle attribute data as a target data, determining the data density in a preset neighborhood corresponding to each sub-data in the target data, and based on the data density, determining core data points and non-core data points in each sub-data; taking any core data point in the core data points as a target core data point, clustering the remaining core data points with the target core data point as the clustering center, and obtaining core data under different clustering categories. point; taking any non-core data point among the non-core data points as a target non-core data point, determining the reference core data point closest to the non-core data point among the core data points under each clustering category, and judging whether the distance between the non-core data point and the reference core data point is less than a preset distance threshold; if the distance between the non-core data point and the reference core data point is less than the preset distance threshold, classifying the non-core data point into the clustering category to which the reference core data point belongs, otherwise, determining the non-core data point as abnormal data; respectively eliminating the abnormal data in the user feature data and the vehicle attribute data to obtain the cleaned user feature data and the cleaned vehicle attribute data.

[0072] Among them, the preset neighborhood is set according to actual needs, and the preset distance threshold is set according to actual needs. The data density can be specifically the amount of data. Specifically, if the data density in the preset neighborhood corresponding to a certain sub-data in the user feature data is greater than the preset density threshold, the sub-data is determined as a core data point. On the contrary, if the data density in the preset neighborhood corresponding to a certain sub-data is less than or equal to the preset density threshold, the sub-data is determined as a non-core data point, and then each core data point is used as a clustering center, and the remaining core data points are clustered to obtain different clustering categories. After that, it is determined whether each non-core data point can be clustered into the above-mentioned clustering categories, and finally the non-core points that cannot be clustered into the above-mentioned clustering categories are determined as abnormal data. Similarly, according to the above method, the abnormal data in the vehicle attribute data can be determined, and the abnormal data in the vehicle attribute data can be eliminated to obtain the cleaned vehicle attribute data. The embodiment of the present invention can improve the data quality, avoid the waste of computing resources caused by useless data participating in the calculation, and avoid the situation where the abnormal data participates in the calculation and causes the communication speech prediction error. That is, the embodiment of the present invention can improve data quality, enhance model performance, and reduce computing costs by preprocessing user feature data and vehicle attribute data.

[0073] Furthermore, after preprocessing the user feature data and the vehicle attribute data, it is also necessary to determine the first round of communication words according to the preprocessed user feature data and the vehicle attribute data. Based on this, step 203 specifically includes: determining the user feature vector corresponding to the user feature data and the vehicle feature vector corresponding to the vehicle attribute data; cross-processing the user feature vector and the vehicle feature vector to obtain a cross-feature vector of words; inputting the cross-feature vector of words into a preset word prediction model to perform word prediction to obtain the first round of communication words, wherein the preset word prediction model is pre-trained based on a sample data set of sample user feature data and sample vehicle attribute data with labeled information. The labeled information is the communication words that meet the requirements of the communication effect. The embodiment of the present invention predicts the communication words through a model, which can improve the accuracy and efficiency of determining the communication words.

[0074] 204. Based on the current communication information, determine whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance. If so, automatically parse the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication script.

[0075] Among them, the current communication information includes the communication tone information and communication content information of the user to be recommended. For the embodiment of the present invention, by analyzing the current communication information, it is possible to determine whether the user to be recommended is willing to conduct the next round of communication. If so, it is determined that the auto insurance outbound communication process meets the preset conditions for continuing to advance. At this time, it is necessary to automatically parse the current communication information to obtain the next round of outbound communication time and the next round of outbound communication words. Based on this, step 204 specifically includes: determining the first insurance information of the vehicle to be recommended; determining the first insurance feature vector corresponding to the vehicle first insurance information, the tone feature vector corresponding to the communication tone information, and the content feature vector corresponding to the communication content information; cross-processing the first insurance feature vector, the tone feature vector, and the content feature vector to obtain a communication cross feature vector; inputting the communication cross feature vector into the large model for information prediction to obtain the next round of outbound communication time and the next round of outbound communication words. Among them, the specific cross-processing process includes: performing feature-level cross-processing on the first-preserved feature vector, the tone feature vector, and the content feature vector to obtain a feature cross-vector; performing element-level cross-processing on the first-preserved feature vector, the tone feature vector, and the content feature vector to obtain an element cross-vector; performing low-order cross-processing on the first-preserved feature vector, the tone feature vector, and the content feature vector to obtain a low-order cross-vector; combining the feature cross-vector, the element cross-vector, and the low-order cross-vector to obtain a communication cross-feature vector.

[0076] Among them, the first maintenance information of the vehicle includes the first maintenance date of the vehicle pre-insured by the recommended user, the first maintenance items and other information. Specifically, firstly, the first maintenance feature vector corresponding to the vehicle first maintenance information, the tone feature vector corresponding to the communication tone information, and the content feature vector corresponding to the communication content information are determined by using word embedding and other methods. Then, in order to make full use of the relationship between the data, extract more implicit features, and take into account both high-order and low-order processing, so that the data can be used more fully and the prediction results obtained later are more accurate to meet the needs of actual application scenarios, it is necessary to cross-process the first maintenance feature vector, tone feature vector, and content feature vector. The specific cross-processing method is as follows: if the first maintenance feature vector is (a1, a2), the tone feature vector is (b1, b2), and the content feature vector is (c1, c2), the specific cross-processing method includes: feature-level crossover between different feature vectors, that is, after taking the Hadamard product of all elements between the vectors, a convolution transformation is performed under certain weights. , the characteristic cross vector is f(w*(a1*b1*c1,a2*b2*c2,a3*b3*c3)); at the same time, all the characteristic vector data are cross-linked at the element level, that is, each element between the vectors is cross-producted, and different weight values ​​are assigned to the results of each product, and then a linear transformation is performed, and the obtained element cross vector is f(w1*a1*b1*c1,w2*a2*b2*c2,w3*a3*b3*c3); in addition, all the characteristic vectors are subjected to low-order cross processing, and weight coefficients are assigned to the results after the cross processing, and then a linear transformation is performed, and the obtained low-order cross vector is f(w2(a1,a2,b1,b2,c1,c2)); finally, the above characteristic cross vectors, element cross vectors, and low-order cross vectors are combined together, such as horizontal splicing, to obtain a communication cross characteristic vector. It should be noted that the above examples are only illustrative and do not limit the embodiments of the present application. Therefore, by cross-processing the first feature vector, tone feature vector, and content feature vector, different features can be automatically or explicitly combined to generate new feature combinations. These combined features may contain complex nonlinear relationships between the original features, allowing the model to capture more detailed and rich information in the data, that is, it can make full use of the relationship between various data, extract more implicit features, and take into account both high-order and low-order processing, so that data utilization is more complete, and the communication results obtained later are more accurate, meeting the needs of actual application scenarios.

[0077] 205. Based on the next round of outbound call communication time, automatically schedule the next outbound call communication module corresponding to the current outbound call communication module to conduct outbound communication on auto insurance information with the recommended user according to the next round of outbound call communication script, and obtain the outbound communication result of the auto insurance outbound call communication process.

[0078] For the embodiment of the present invention, the entire auto insurance outbound call communication process is described in the order of the first dial module, the appointment dial module, the underwriting dial module, the order promotion dial module, and the communication module. The method includes: in response to the first round of outbound call communication signal for recommending auto insurance information to the user to be recommended, obtaining the user characteristic data and vehicle attribute data of the user to be recommended; based on the user characteristic data and the vehicle attribute data, determining the first round of outbound call communication words, and scheduling the first dial module to communicate the auto insurance information outbound with the user to be recommended according to the first round of outbound call communication words, and obtaining the first round of communication information of the auto insurance outbound call communication process; based on the first round of communication information, judging Whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance, if so, the first round of communication information is automatically parsed to obtain the next round of outbound call communication time and the next round of outbound call communication words, wherein the next round of outbound call communication time and the next round of outbound call communication words are the appointment dialing outbound call communication time and appointment dialing outbound call communication words corresponding to the appointment dialing module; based on the appointment dialing outbound call communication time, the appointment dialing module is scheduled to conduct an outbound call communication on auto insurance information with the to-be-recommended user according to the appointment dialing outbound call communication words, and the appointment dialing communication information of the auto insurance outbound call communication process is obtained; based on the appointment dialing communication information, it is determined whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance. The preset conditions for continuing to advance are met. If so, the appointment call communication information is automatically parsed to obtain the underwriting call communication time and the underwriting call communication language; based on the underwriting call communication time, the underwriting call module is automatically scheduled to conduct an outbound call communication on auto insurance information with the to-be-recommended user according to the underwriting call communication language to obtain the underwriting call communication information of the auto insurance outbound call communication process; based on the underwriting call communication information, it is determined whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance. If so, the underwriting call communication information is automatically parsed to obtain the order promotion outbound call communication time and the order promotion outbound call communication language; based on the order promotion outbound call communication time During the communication time, the order promotion dialing communication module is automatically scheduled to communicate with the user to be recommended about the car insurance information according to the order promotion dialing communication language, and the order promotion dialing communication information of the car insurance outbound call communication process is obtained; based on the order promotion dialing communication information, it is judged whether the car insurance outbound call communication process meets the preset conditions for continuing to advance, and if so, the order promotion dialing communication information is automatically parsed to obtain the communication module communication time and the communication module communication language; based on the communication module communication time, the communication module is automatically scheduled to communicate with the user to be recommended about the car insurance information according to the communication module communication language, and the outbound call communication result of the car insurance outbound call communication process is obtained. Among them, the outbound call communication result can be the recommendation result of the car insurance information. After each communication process ends, if the user does not intend to communicate next time, that is, it is judged that the car insurance outbound call communication process does not meet the preset conditions for continuing to advance, in order to improve the user experience, it is prohibited to schedule the next outbound call communication module to communicate with the user to be recommended.Furthermore, in any communication process, if the user to be recommended sends an inquiry signal for auto insurance value information, in order to improve the accuracy of feedback information to the user, the method includes: in the process of scheduling the next outbound communication module to conduct outbound communication on auto insurance information with the user to be recommended, in response to the auto insurance value signal, generating an auto insurance demand inquiry script, and scheduling the next outbound communication module to conduct an outbound communication on auto insurance demand with the user to be recommended according to the auto insurance demand inquiry script; receiving feedback auto insurance demand information returned by the user to be recommended in response to the auto insurance demand inquiry script, and generating auto insurance value call parameters based on the feedback auto insurance demand information; based on the auto insurance value call parameters, retrieving the auto insurance value in the value interface, and scheduling the next outbound communication module to transmit the auto insurance value to the terminal of the user to be recommended.

[0079] Specifically, the first call module: mainly communicates before the first insurance date, the purpose is to contact the user in advance, and add auxiliary contact methods to prepare for the subsequent process of the auto insurance information after the first insurance date; the appointment call module: after the insurance date is reached, further communication is established with the user to encourage the user to confirm the auto insurance information, or encourage the user to convert the auto insurance information; the underwriting call module: after the user has the intention to convert the auto insurance information, a process for approving the auto insurance information is provided to the user; the order promotion call module: in the last step of the auto insurance information outbound call communication process, when the user agrees to the conversion of the auto insurance information, but has not converted, it is necessary to call the user to remind the conversion process. Furthermore, after the outbound call communication process of the order promotion call module is completed, the workflow of the communication module needs to be started. Based on this, the method includes: if the next outbound call communication module is the order promotion call module, then based on the outbound call communication result corresponding to the order promotion call module, it is determined whether the auto insurance outbound call communication process is completed according to the preset result. If so, the communication module is dispatched to send the conversion link of the auto insurance information to the recommended user.

[0080] Among them, the preset result means that the outbound communication result of auto insurance meets the expected expectations, such as waiting for the recommended user to agree to convert the auto insurance information; the communication module contains the company's own communication procedures.

[0081] Specifically, if the outbound call communication result obtained by the promotion dialing module is parsed, it is determined that the user to be recommended agrees to the conversion of the auto insurance information, but the user has not implemented the conversion operation. At this time, it is necessary to schedule the communication module to send the user a conversion link of the auto insurance information, such as sending a conversion identifier, such as a QR code, etc., and receive the user's information to meet the conversion needs of the auto insurance information. After the user to be recommended receives the conversion link, in order to achieve the final conversion effect, it is also necessary to perform conversion-related operations. Based on this, the method includes: after a preset time of sending the conversion link, determining whether feedback information from the user to be recommended regarding the conversion link is received; if the feedback information is not received, determining the operation process information of the conversion link, and scheduling the promotion dialing module to conduct an outbound call communication with the user to be recommended based on the operation process information; if the feedback information is received, generating conversion voucher information corresponding to the feedback information, and scheduling the communication module to send the conversion voucher information to the user to be recommended.

[0082] Among them, the preset time is set according to actual needs. Specifically, after the preset time of the conversion link for sending the auto insurance information to the user to be recommended, the user to be recommended has not realized the conversion of the auto insurance information for the conversion link. At this time, it is necessary to start the promotion dialing module to send the operation process information of the conversion link to the user, so as to be used for the conversion of the auto insurance information based on the operation process information. If the user completes the conversion of the auto insurance information based on the conversion link, it is necessary to generate conversion voucher information and send the conversion voucher information to the user to be recommended through the communication module. Among them, the conversion voucher information can be transaction invoice, receipt and other voucher information.

[0083] Furthermore, in the communication process of each outbound communication module in the auto insurance communication process, if the user raises an inquiry about the auto insurance value information, based on the inquiry information, the corresponding communication module generates an auto insurance demand inquiry script, such as: asking the user what types of auto insurance items the user needs, and then automatically generating a user demand in a preset format (set according to actual needs) for the auto insurance demand information fed back by the user, and generating auto insurance value call parameters based on the user demand in the preset format, and using the auto insurance value call parameters to retrieve the auto insurance value in the value interface, such as transmitting the auto insurance value call parameters to the CRM (Customer Relationship Management) system through RPA (Robotic Process Automation, outbound robot process automation technology), and the CRM system calculates the value according to the auto insurance value call parameters, and then transmits the value to the corresponding communication module, and the corresponding communication module conveys the auto insurance value to the user to be recommended. The embodiment of the present invention does not directly generate value information, but calls the value information in the background after receiving the user demand, which can improve the accuracy of value provision.

[0084] According to another outbound communication method for auto insurance information provided by the present invention, compared with the current communication method of auto insurance information driven manually, the present invention divides the auto insurance outbound communication process in the outbound dialogue robot into multiple outbound communication modules according to the process sequence; and obtains the current communication information after scheduling the current outbound communication module to conduct outbound communication with the user to be recommended; then, based on the current communication information, it is determined whether the auto insurance outbound communication process meets the preset conditions for continuing to advance, and if so, the current communication information is automatically parsed to obtain the next round of outbound communication time and the next round of outbound communication words; finally, based on the next round of outbound communication time, the next outbound communication module corresponding to the current outbound communication module is automatically scheduled to conduct outbound communication on auto insurance information with the user to be recommended according to the next round of outbound communication words, and the outbound communication result of the auto insurance outbound communication process is obtained. Since the communication of auto insurance information is a long process and has a long time span, the auto insurance outbound communication process in the outbound dialogue robot is divided into multiple outbound communication modules according to the process sequence, and outbound communication is performed to the user based on each outbound communication module. Each outbound communication module only needs to narrate the corresponding personalized words, and the communication logic is clear, avoiding the problem of stringing processes, thereby improving the outbound communication effect. At the same time, the embodiment of the present invention automatically determines the appropriate communication words and the appropriate communication time through the various communication modules of the outbound robot, and automatically realizes the outbound communication to the user. The whole process is automated without manual participation, thereby improving the outbound communication efficiency and saving labor costs. It can also avoid the situation where manual negligence leads to incorrect determination of communication time and communication content. Therefore, the present invention can improve the accuracy of outbound communication on auto insurance information and enhance user experience.

[0085] Further, as Figure 1 In a specific implementation, an embodiment of the present invention provides an outbound communication device for vehicle insurance information, such as Figure 4 As shown, the device includes: a dividing unit 31, an acquiring unit 32, a judging unit 33, and a scheduling unit 34.

[0086] The division unit 31 can be used to divide the auto insurance outbound call communication process in the outbound call dialogue robot into multiple outbound call communication modules according to the process sequence.

[0087] The acquisition unit 32 may be used to acquire current communication information after the outbound call communication module is scheduled to perform outbound communication with the user to be recommended.

[0088] The judgment unit 33 can be used to judge whether the car insurance outbound call communication process meets the preset conditions for continuing to advance based on the current communication information. If so, the current communication information is automatically analyzed to obtain the next round of outbound call communication time and the next round of outbound call communication words.

[0089] The scheduling unit 34 can be used to automatically schedule the next outbound call communication module corresponding to the current outbound call communication module based on the next round of outbound call communication time to conduct outbound call communication on car insurance information with the recommended user according to the next round of outbound call communication script, and obtain the outbound communication result of the car insurance outbound call communication process.

[0090] In a specific application scenario, each of the outbound call communication modules includes: a first call module, an appointment call module, an underwriting call module, a promotion call module, and a communication module. In order to obtain current communication information, such as Figure 5 As shown, the acquisition unit 32 includes an acquisition module 321 and an outbound call communication module 322 .

[0091] The acquisition module 321 may be used to acquire user feature data and vehicle attribute data of the user to be recommended in response to a first round of outbound call communication signals for recommending auto insurance information to the user to be recommended.

[0092] The outbound call communication module 322 can be used to determine the first round of outbound call communication words based on the user feature data and the vehicle attribute data, and schedule the first call module to conduct outbound communication on car insurance information with the recommended user according to the first round of outbound call communication words, obtain the first round of communication information of the car insurance outbound call communication process, and determine the first round of communication information as the current communication information.

[0093] In a specific application scenario, in order to display the car insurance value information to the user, the device further includes: a generating unit 35 and a retrieving unit 36 ​​.

[0094] The generating unit 35 can be used to generate a car insurance demand inquiry script in response to a car insurance value signal during the process of scheduling the next outbound communication module to conduct an outbound communication on car insurance information with the user to be recommended, and schedule the next outbound communication module to conduct an outbound communication on car insurance demand with the user to be recommended according to the car insurance demand inquiry script.

[0095] The generating unit 35 may be specifically configured to receive the feedback auto insurance demand information returned by the to-be-recommended user in response to the auto insurance demand inquiry dialogue, and generate an auto insurance value calling parameter based on the feedback auto insurance demand information.

[0096] The retrieving unit 36 ​​may be used to retrieve the auto insurance value in the value interface based on the auto insurance value calling parameter, and schedule the next outbound call communication module to transmit the auto insurance value to the user terminal to be recommended.

[0097] In a specific application scenario, the current communication information includes the communication tone information and communication content information of the recommended user. In order to determine the next round of outbound communication time and the next round of outbound communication language, the judgment unit 33 includes a determination module 331, a cross-processing module 332, and a prediction module 333.

[0098] The determination module 331 may be used to determine the first vehicle maintenance information of the user to be recommended.

[0099] The determination module 331 may also be used to determine a first maintenance feature vector corresponding to the vehicle first maintenance information, a tone feature vector corresponding to the communication tone information, and a content feature vector corresponding to the communication content information.

[0100] The cross-processing module 332 may be used to perform cross-processing on the first protection feature vector, the tone feature vector, and the content feature vector to obtain a communication cross-feature vector.

[0101] The prediction module 333 can be used to input the communication cross feature vector into the large model to perform information prediction, and obtain the next round of outbound call communication time and the next round of outbound call communication words.

[0102] In a specific application scenario, in order to send transaction information to the user, the device further includes: a sending unit 37.

[0103] The sending unit 37 can be used to determine whether the auto insurance outbound communication process is completed according to the preset result based on the outbound communication result corresponding to the outbound communication module if the next outbound communication module is the order promotion module. If so, the communication module is scheduled to send a conversion link of auto insurance information to the user to be recommended.

[0104] In a specific application scenario, in order to send conversion voucher information to the user, the sending unit 37 can also be used to determine whether feedback information from the user to be recommended regarding the conversion link is received after a preset time of sending the conversion link; if the feedback information is not received, the operation process information of the conversion link is determined, and the promotion dialing module is scheduled to make an outbound call with the user to be recommended based on the operation process information; if the feedback information is received, the conversion voucher information corresponding to the feedback information is generated, and the communication module is scheduled to send the conversion voucher information to the user to be recommended.

[0105] In a specific application scenario, in order to pre-process the user feature data and the vehicle attribute data, the acquisition unit 32 further includes a pre-processing module 323 .

[0106] The preprocessing module 323 can be used to take any data in the user feature data and the vehicle attribute data as target data, determine the data density in the preset neighborhood corresponding to each sub-data in the target data, and determine the core data points and non-core data points in each sub-data based on the data density; take any core data point in the core data points as a target core data point, cluster the remaining core data points with the target core data point as the cluster center, and obtain core data points under different clustering categories; take any non-core data point in the non-core data points as a target non-core data point, and cluster the remaining core data points with the target core data point as the cluster center. Determine the reference core data point that is closest to the non-core data point among the core data points under each clustering category, and judge whether the distance between the non-core data point and the reference core data point is less than a preset distance threshold; if the distance between the non-core data point and the reference core data point is less than the preset distance threshold, classify the non-core data point into the clustering category to which the reference core data point belongs, otherwise, determine the non-core data point as abnormal data; respectively eliminate the abnormal data in the user feature data and the vehicle attribute data to obtain the cleaned user feature data and the cleaned vehicle attribute data.

[0107] It should be noted that for other corresponding descriptions of the functional modules involved in the outbound call communication device for automobile insurance information provided by the embodiment of the present invention, reference can be made to Figure 1 The corresponding description of the method shown will not be repeated here.

[0108] Based on the above Figure 1 The method shown, accordingly, an embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the following steps are implemented: dividing the auto insurance outbound call communication process in the outbound call dialogue robot into multiple outbound call communication modules according to the process sequence; obtaining the current communication information after scheduling the current outbound call communication module to conduct outbound call communication with the user to be recommended; based on the current communication information, judging whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance, and if so, automatically parsing the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication words; based on the next round of outbound call communication time, automatically scheduling the next outbound call communication module corresponding to the current outbound call communication module to conduct outbound communication on auto insurance information with the user to be recommended according to the next round of outbound call communication words, and obtaining the outbound communication result of the auto insurance outbound call communication process.

[0109] Based on the above Figure 1 The method shown and Figure 4 The embodiment of the device shown in the figure, the embodiment of the present invention also provides a physical structure diagram of a computer device, such as Figure 6 As shown, the computer device includes: a processor 41, a memory 42, and a computer program stored in the memory 42 and executable on the processor, wherein the memory 42 and the processor 41 are both arranged on a bus 43, and the processor 41 implements the following steps when executing the program: dividing the auto insurance outbound call communication process in the outbound call dialogue robot into multiple outbound call communication modules according to the process sequence; obtaining the current communication information after scheduling the current outbound call communication module to conduct outbound call communication with the user to be recommended; based on the current communication information, judging whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance, and if so, automatically parsing the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication words; based on the next round of outbound call communication time, automatically scheduling the next outbound call communication module corresponding to the current outbound call communication module to conduct outbound communication on auto insurance information with the user to be recommended according to the next round of outbound call communication words, and obtaining the outbound communication result of the auto insurance outbound call communication process.

[0110] Through the technical solution of the present invention, the present invention divides the auto insurance outbound call communication process in the outbound call dialogue robot into multiple outbound call communication modules according to the process sequence; and obtains the current communication information after scheduling the current outbound call communication module to conduct outbound call communication with the user to be recommended; then, based on the current communication information, it is judged whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance. If so, the current communication information is automatically parsed to obtain the next round of outbound communication time and the next round of outbound communication speech; finally, based on the next round of outbound communication time, the next outbound call communication module corresponding to the current outbound call communication module is automatically scheduled to conduct outbound communication on auto insurance information with the user to be recommended according to the next round of outbound communication speech, and the outbound communication result of the auto insurance outbound communication process is obtained. Since the communication of auto insurance information is a long process and has a long time span, the auto insurance outbound communication process in the outbound dialogue robot is divided into multiple outbound communication modules according to the process sequence, and outbound communication is performed to the user based on each outbound communication module. Each outbound communication module only needs to narrate the corresponding personalized words, and the communication logic is clear, avoiding the problem of stringing processes, thereby improving the outbound communication effect. At the same time, the embodiment of the present invention automatically determines the appropriate communication words and the appropriate communication time through the various communication modules of the outbound robot, and automatically realizes the outbound communication to the user. The whole process is automated without manual participation, thereby improving the outbound communication efficiency and saving labor costs. It can also avoid the situation where manual negligence leads to incorrect determination of communication time and communication content. Therefore, the present invention can improve the accuracy of outbound communication on auto insurance information and enhance user experience.

[0111] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An outbound communication method for automobile insurance information, characterized in that: include: Divide the auto insurance outbound call communication process in the outbound call conversation robot into multiple outbound call communication modules according to the process sequence; Obtain the current communication information after the outbound call communication module schedules the outbound call communication for the recommended user; Based on the current communication information, determine whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance. If so, automatically analyze the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication script; Based on the next round of outbound call communication time, the next outbound call communication module corresponding to the current outbound call communication module is automatically scheduled to conduct outbound call communication on auto insurance information with the to-be-recommended user according to the next round of outbound call communication script, and the outbound communication result of the auto insurance outbound call communication process is obtained.

2. The method according to claim 1, characterized in that Each of the outbound call communication modules includes: a first call module, a reservation call module, an underwriting call module, a promotion call module, and a communication module. The current outbound call communication module is any one of the first call module, the reservation call module, the underwriting call module, and the promotion call module. If the current outbound call communication module is the first call module, obtaining the current communication information after scheduling the current outbound call communication module to perform outbound call communication for the recommended user includes: In response to a first round of outbound call communication signals for recommending auto insurance information to the user to be recommended, obtaining user feature data and vehicle attribute data of the user to be recommended; Based on the user characteristic data and the vehicle attribute data, the first round of outbound call communication words are determined, and the first call module is scheduled to conduct outbound communication on auto insurance information with the recommended user according to the first round of outbound call communication words, so as to obtain the first round of communication information of the auto insurance outbound call communication process, and the first round of communication information is determined as the current communication information.

3. The method according to claim 1, characterized in that The method further comprises: In the process of scheduling the next outbound call communication module to communicate with the to-be-recommended user about the car insurance information, generating a car insurance demand inquiry speech in response to the car insurance value signal, and scheduling the next outbound call communication module to communicate with the to-be-recommended user about the car insurance demand according to the car insurance demand inquiry speech; receiving feedback auto insurance demand information returned by the to-be-recommended user in response to the auto insurance demand inquiry dialogue, and generating auto insurance value call parameters based on the feedback auto insurance demand information; Based on the car insurance value calling parameter, the car insurance value is retrieved in the value interface, and the next outbound call communication module is scheduled to transmit the car insurance value to the user terminal to be recommended.

4. The method according to claim 1, characterized in that: The current communication information includes the communication tone information and communication content information of the user to be recommended; The automatic analysis of the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication words includes: Determine the first vehicle maintenance information of the user to be recommended; Determine a first maintenance feature vector corresponding to the vehicle first maintenance information, a tone feature vector corresponding to the communication tone information, and a content feature vector corresponding to the communication content information; Cross-processing the first protection feature vector, the tone feature vector, and the content feature vector to obtain a communication cross feature vector; The communication cross feature vector is input into the large model for information prediction to obtain the next round of outbound call communication time and the next round of outbound call communication words.

5. The method according to claim 2, characterized in that: After automatically scheduling the next outbound call communication module corresponding to the current outbound call communication module to perform outbound communication on car insurance information with the to-be-recommended user according to the next round of outbound call communication words and obtaining the outbound communication result of the car insurance outbound call communication process, the method further includes: If the next outbound call communication module is the order promotion dialing module, then based on the outbound call communication result corresponding to the order promotion dialing module, it is determined whether the auto insurance outbound call communication process is completed according to the preset result. If so, the communication module is dispatched to send a conversion link of auto insurance information to the user to be recommended.

6. The method according to claim 5, characterized in that After scheduling the communication module to send a conversion link of the auto insurance information to the to-be-recommended user, the method further includes: After a preset time of sending the conversion link, determining whether feedback information of the to-be-recommended user regarding the conversion link is received; If the feedback information is not received, determining the operation process information of the conversion link, and scheduling the promotion and order dialing module to make an outbound call communication with the to-be-recommended user based on the operation process information; If the feedback information is received, conversion credential information corresponding to the feedback information is generated, and the communication module is scheduled to send the conversion credential information to the user to be recommended.

7. The method according to claim 2, characterized in that Before determining the first round of outbound call communication skills based on the user feature data and the vehicle attribute data, the method further includes: Taking any data in the user feature data and the vehicle attribute data as target data, respectively, determining the data density in a preset neighborhood corresponding to each sub-data in the target data, and determining core data points and non-core data points in each sub-data based on the data density; Taking any core data point among the core data points as a target core data point, taking the target core data point as the cluster center, clustering the remaining core data points to obtain core data points under different clustering categories; Taking any non-core data point among the non-core data points as a target non-core data point, determining a reference core data point that is closest to the non-core data point among the core data points under each clustering category, and judging whether the distance between the non-core data point and the reference core data point is less than a preset distance threshold; If the distance between the non-core data point and the reference core data point is less than the preset distance threshold, the non-core data point is classified into the cluster category to which the reference core data point belongs; otherwise, the non-core data point is determined as abnormal data; Abnormal data in the user characteristic data and the vehicle attribute data are respectively removed to obtain cleaned user characteristic data and cleaned vehicle attribute data.

8. An outbound communication device for automobile insurance information, characterized in that: include: A division unit, used to divide the auto insurance outbound call communication process in the outbound call dialogue robot into multiple outbound call communication modules according to the process sequence; An acquisition unit, used for acquiring current communication information after the outbound call communication module is scheduled to perform outbound call communication with the user to be recommended; A judgment unit, configured to judge whether the auto insurance outbound call communication process meets the preset conditions for continuing to advance based on the current communication information, and if so, automatically parse the current communication information to obtain the next round of outbound call communication time and the next round of outbound call communication words; The scheduling unit is used to automatically schedule the next outbound call communication module corresponding to the current outbound call communication module based on the next round of outbound call communication time to conduct outbound communication on auto insurance information with the recommended user according to the next round of outbound call communication script, and obtain the outbound communication result of the auto insurance outbound call communication process.

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

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images