阅读说明：本技术 一种基于数据特征识别的智能外呼系统通话数据记录分析管理系统 (Intelligent outbound system call data record analysis and management system based on data feature recognition ) 是由 毛喜斌 周红彪 于 2021-09-13 设计创作，主要内容包括：本发明公开提供的一种基于数据特征识别的智能外呼系统通话数据记录分析管理系统。该基于数据特征识别的智能外呼系统通话数据记录分析管理系统包括通话数据信息获取模块、问卷回访信息获取模块、问卷回访结果统计模块、额外问题信息采集模块、产品负面信息采集模块、云分析平台、数据库、预警提醒终端和信息发送终端；本发明通过对该智能外呼系统各外呼端口对应的问卷受访人员对应的通话记录信息进行详细的分析,有效的解决了现有的智能外呼系统通话数据记录分析管理系统无法实现对问卷受访人员的精准筛分的问题,进而有效的提高了产品问卷的回访效率,同时也大大的提高了产品的市场竞争优势。(The invention discloses an intelligent outbound system call data record analysis and management system based on data feature recognition. The intelligent outbound system call data recording, analyzing and managing system based on data feature recognition comprises a call data information acquisition module, a questionnaire return visit result statistics module, an additional problem information acquisition module, a product negative information acquisition module, a cloud analysis platform, a database, an early warning reminding terminal and an information sending terminal; the call record information corresponding to the questionnaire interviewee corresponding to each outbound port of the intelligent outbound system is analyzed in detail, so that the problem that the call data record analysis and management system of the existing intelligent outbound system cannot realize accurate screening of the questionnaire interviewee is effectively solved, the return visit efficiency of the product questionnaire is effectively improved, and the market competition advantage of the product is greatly improved.)

1. The utility model provides an intelligence system's conversation data record analysis management system that calls out based on data feature identification which characterized in that: the system comprises a call data information acquisition module, a questionnaire return visit result statistics module, an additional problem information acquisition module, a product negative information acquisition module, a cloud analysis platform, a database, an early warning reminding terminal and an information sending terminal;

the call data information acquisition module is used for acquiring call data information corresponding to each outbound port of the intelligent outbound system in the acquisition period, further acquiring the number of outbound ports corresponding to the intelligent outbound system in the acquisition period, and numbering the outbound ports corresponding to the intelligent outbound system in the acquisition period according to a preset sequence, wherein the outbound ports are marked as 1,2,. j,. m in sequence;

the questionnaire return visit information acquisition module is used for acquiring questionnaire return visit information corresponding to the product of the intelligent outbound system in the acquisition period;

the questionnaire return visit result counting module is used for counting return visit result information corresponding to each outbound port of the intelligent outbound system in the acquisition period so as to obtain the return visit result information corresponding to each outbound port of the intelligent outbound system in the acquisition period;

the extra question information acquisition module is used for acquiring extra question information corresponding to the call of the interviewee in each call data of each outbound port according to the call data information and the questionnaire return visit information corresponding to each outbound port of the intelligent outbound system in the acquisition period;

the product negative information acquisition module is used for extracting negative information corresponding to the product according to the call data information corresponding to each outbound port of the intelligent outbound system in the acquisition period;

the cloud analysis platform is used for processing and analyzing call data information of each outbound port of the intelligent outbound system, questionnaire return visit results, extra problem information and product negative information in the acquisition period;

the early warning reminding terminal is used for sending an early warning signal to the product sales management personnel when the product negative information of the intelligent outbound system reaches an early warning value in the acquisition period;

and the information sending terminal is used for sending the result of the analysis of the cloud analysis platform to the product return visit management personnel.

2. The intelligent outbound system call data record analysis and management system based on data feature identification as claimed in claim 1, wherein: the call data information corresponding to each outbound port in the acquisition period comprises the number of call data corresponding to each outbound port of the intelligent outbound system in the acquisition period, and each outbound portThe method comprises the steps of calling voice information corresponding to each call data of ports and call duration corresponding to each call data of each outbound port, numbering the call data corresponding to each outbound port in an acquisition period according to a preset sequence, marking the call data as 1,2, 1.. i.. n in sequence, and constructing a call data information set T of each outbound port_w ^d(T_w ^d1,T_w ^d2,...T_w ^di,...T_w ^dn)，T_w ^di represents information corresponding to the ith call data of the d-th outbound port of the intelligent outbound system in the acquisition period, w represents call data information, and w is a1, a2, a1 and a2 respectively represent voice information corresponding to the call data and call duration corresponding to the call data.

3. The intelligent outbound system call data record analysis and management system based on data feature identification as claimed in claim 1, wherein: the questionnaire return visit information corresponding to the product of the intelligent outbound system in the acquisition period comprises the number of questions of the product questionnaire return visit of the intelligent outbound system in the acquisition period, the return visit sequence corresponding to each questionnaire return visit question, the importance weight corresponding to each questionnaire return visit question and the content corresponding to each questionnaire return visit question, the questionnaire return visit questions of the intelligent outbound system in the acquisition period are numbered according to the return visit sequence, the numbers are sequentially marked as 1,2, 1.x, y, and then an information set H of each questionnaire return visit question of the intelligent outbound system product is constructed_e(H_e1,H_e2,...H_ex,...H_ey)，H_ex represents the e-th information corresponding to the x-th questionnaire back-visit question of the intelligent outbound call system in the acquisition period, e represents questionnaire back-visit question information, e is b1, b2, b3, b1, b2 and b3 respectively represent the back-visit sequence corresponding to each questionnaire back-visit question, the importance weight corresponding to each questionnaire back-visit question and the content corresponding to each questionnaire back-visit question.

4. The intelligent outbound system call data record analysis and management system based on data feature identification as claimed in claim 1, wherein: the question is rewoundThe access result information comprises the number of answer questions corresponding to the interviewee of the questionnaire, the question numbers corresponding to the answer questions of the interviewee of the questionnaire and the answer contents corresponding to the answer questions of the interviewee of the questionnaire, and then a questionnaire return access result information set F of each piece of call data of each outbound port is constructed_z ^d(F_z ^d1,F_z ^d2,...F_z ^di,...F_z ^dn)，F_z ^di represents the z-th return visit result information corresponding to the ith call data of the d-th outbound port of the intelligent outbound system in the acquisition period, z represents the return visit result information, and z is c1, c2, c3, c1, c2 and c3 respectively represent the number of answer questions corresponding to the interviewee of the questionnaire, the question numbers corresponding to the answer questions of the interviewee of the questionnaire and the answer contents corresponding to the answer questions of the interviewee of the questionnaire.

5. The intelligent outbound system call data record analysis and management system based on data feature identification as claimed in claim 1, wherein: the extra question information corresponding to the call of the interviewee of the questionnaire comprises the number of extra questions corresponding to the call of the interviewee of the questionnaire and the content corresponding to each extra question of the interviewee of the questionnaire, and then according to the extra information corresponding to the call of the interviewee of the questionnaire, an extra question information set E of each call interviewee of each outbound port is constructed_s ^d(E_s ^d1,E_s ^d2,...E_s ^di,...E_s ^dn)，E_s ^di represents the s-th extra question information corresponding to the interviewee of the questionnaire in the ith call data of the d-th outbound port in the acquisition period, s represents the extra question information for the interviewee of the questionnaire, and s is v1, v2, v1 and v2 respectively represent the number of extra questions corresponding to the call of the interviewee of the questionnaire and the content corresponding to each extra question of the interviewee of the questionnaire.

6. The intelligent outbound system call data record analysis and management system based on data feature identification as claimed in claim 1, wherein: and the product negative information acquisition is used for extracting the voice information corresponding to the visited person of each call data questionnaire of each call port of the intelligent call-out system in the acquisition period according to the voice information corresponding to each call data of each call port of the intelligent call-out system in the acquisition period, converting the voice information corresponding to the visited person of each call data questionnaire of each call port of the intelligent call-out system in the acquisition period into a text format, and counting the number corresponding to the product negative vocabulary and the type corresponding to the product negative vocabulary in the return visit of the intelligent call-out system questionnaire in the acquisition period.

7. The intelligent outbound system call data record analysis and management system based on data feature identification as claimed in claim 1, wherein: the cloud analysis platform analyzes the call data information of each outbound port of the intelligent outbound system, and is used for analyzing the call duration corresponding to each call data of each outbound port of the intelligent outbound system, acquiring the call duration corresponding to each call data of each outbound port of the intelligent outbound system in the acquisition period, further acquiring the number of the interviewee participants corresponding to the intelligent outbound system in the acquisition period and the call duration corresponding to each interviewee participant of the questionnaire, comparing the call duration corresponding to each interviewee participant of the intelligent outbound system in the acquisition period with the threshold value of the standard call duration corresponding to the call data of the outbound port, and further counting the call duration of each interviewee participant of the intelligent outbound system to meet the influence coefficient.

8. The intelligent outbound system call data record analysis and management system based on data feature identification as claimed in claim 1, wherein: the cloud analysis platform analyzes the questionnaire return visit results and the extra question information corresponding to each outbound system of the intelligent outbound system, is used for analyzing the matching degree of answering questions of each questionnaire interviewee and the attention degree of products, acquires questionnaire return visit result information and extra question information corresponding to each call data questionnaire interviewee of each outbound port of the intelligent outbound system, and further respectively counts the influence coefficients of answering question matching degrees of each questionnaire interviewee of the intelligent outbound system and the influence coefficients of product attention degrees of each questionnaire interviewee of the intelligent outbound system.

9. The intelligent outbound system call data record analysis and management system based on data feature identification as claimed in claim 1, wherein: the cloud platform analyzes the negative information of the product and is used for analyzing the negative information corresponding to the product in the questionnaire return visit corresponding to the intelligent outbound system in the acquisition period, further obtains the number corresponding to the negative vocabulary of the product in the questionnaire return visit of the intelligent outbound system in the acquisition period and the type corresponding to the negative vocabulary of the product, compares the number corresponding to the negative vocabulary of the product of each type with the standard number corresponding to the negative vocabulary of the product, further counts the negative early warning influence coefficient of the product in the questionnaire return visit of the intelligent outbound system, compares the negative early warning influence coefficient of the product in the questionnaire return visit of the intelligent outbound system with the preset negative early warning influence coefficient of the product, and marks the product as a negative information early warning product if the negative early warning influence coefficient of the product in the questionnaire return visit of the intelligent outbound system is greater than the preset negative early warning influence coefficient of the product.

10. The intelligent outbound system call data record analysis and management system based on data feature identification as claimed in claim 1, wherein: the cloud analysis platform is further used for comprehensively analyzing the call data information, the questionnaire return visit result information and the extra problem information, further counting the product comprehensive trust degree influence coefficients of the interviewees of each questionnaire of the intelligent outbound call system, matching and comparing the product comprehensive trust degree influence coefficients of the interviewees of each questionnaire of the intelligent outbound call system with the product trust degree influence coefficients corresponding to the product trust levels, and obtaining the product trust levels corresponding to the interviewees of each questionnaire of the intelligent outbound call system in the acquisition period.

Technical Field

The invention belongs to the technical field of call analysis and management of an outbound system, and relates to an intelligent call data record analysis and management system of the outbound system based on data feature recognition.

Background

With the rapid development of communication technology, many merchants use the intelligent outbound system to make questionnaire return visits to the consumed personnel of the products in order to widen the market development space of the products. In order to improve the efficiency of revisiting the questionnaire of the consumed personnel of the product, the call data records of the intelligent outbound system need to be analyzed and managed.

The existing intelligent call-out system call data record analysis and management system mainly analyzes and manages the quality of questionnaire return visits, and does not analyze questionnaire visited persons accurately, so that the existing intelligent call-out system call data record analysis and management system has certain disadvantages, the existing intelligent call-out system call data record analysis and management system cannot realize accurate screening of questionnaire visited persons, and further cannot effectively improve the accuracy of positioning of each questionnaire visited person, on one hand, the existing intelligent call-out system call data record analysis and management system cannot effectively improve the return visit efficiency of product questionnaires, on the other hand, the existing intelligent call-out system call data record analysis and management system cannot effectively improve the market competitive advantages of products.

Disclosure of Invention

In view of the above, in order to solve the problems in the background art, an intelligent outbound system call data record analysis and management system based on data feature recognition is proposed for revisit of repeated consumer product questionnaires, so that accurate analysis and efficient management of the intelligent outbound system call data record are realized;

the purpose of the invention can be realized by the following technical scheme:

the invention provides an intelligent outbound system call data record analysis management system based on data feature recognition, which comprises a call data information acquisition module, a questionnaire return visit result statistics module, an additional problem information acquisition module, a product negative information acquisition module, a cloud analysis platform, a database, an early warning reminding terminal and an information sending terminal, wherein the call data information acquisition module is used for acquiring call data information;

the call data information acquisition module is used for acquiring call data information corresponding to each outbound port of the intelligent outbound system in the acquisition period, further acquiring the number of outbound ports corresponding to the intelligent outbound system in the acquisition period, and numbering the outbound ports corresponding to the intelligent outbound system in the acquisition period according to a preset sequence, wherein the outbound ports are marked as 1,2,. j,. m in sequence;

the questionnaire return visit information acquisition module is used for acquiring questionnaire return visit information corresponding to the product of the intelligent outbound system in the acquisition period;

the questionnaire return visit result counting module is used for counting return visit result information corresponding to each outbound port of the intelligent outbound system in the acquisition period so as to obtain the return visit result information corresponding to each outbound port of the intelligent outbound system in the acquisition period;

the extra question information acquisition module is used for acquiring extra question information corresponding to the call of the interviewee in each call data of each outbound port according to the call data information and the questionnaire return visit information corresponding to each outbound port of the intelligent outbound system in the acquisition period;

the product negative information acquisition module is used for extracting negative information corresponding to the product according to the call data information corresponding to each outbound port of the intelligent outbound system in the acquisition period;

the cloud analysis platform is used for processing and analyzing call data information of each outbound port of the intelligent outbound system, questionnaire return visit results, extra problem information and product negative information in the acquisition period;

the early warning reminding terminal is used for sending an early warning signal to the product sales management personnel when the product negative information of the intelligent outbound system reaches an early warning value in the acquisition period;

and the information sending terminal is used for sending the result of the analysis of the cloud analysis platform to the product return visit management personnel.

Preferably, the call data information corresponding to each outbound port in the collection period includes the number of call data corresponding to each outbound port of the intelligent outbound system in the collection period, the voice information corresponding to each call data of each outbound port, and the call duration corresponding to each call data of each outbound port, and the call data corresponding to each outbound port in the collection period is processed according to a preset sequenceNumbering, sequentially marking as 1,2,. i,. n, and constructing each call data information set T of each outbound port_w ^d(T_w ^d1,T_w ^d2,...T_w ^di,...T_w ^dn)，T_w ^di represents information corresponding to the ith call data of the d-th outbound port of the intelligent outbound system in the acquisition period, w represents call data information, and w is a1, a2, a1 and a2 respectively represent voice information corresponding to the call data and call duration corresponding to the call data.

Preferably, the questionnaire return access information corresponding to the product of the intelligent outbound system in the acquisition period includes the number of questions of the product questionnaire return access of the intelligent outbound system in the acquisition period, the return access sequence corresponding to each questionnaire return access question, the importance weight corresponding to each questionnaire return access question, and the content corresponding to each questionnaire return access question, and then the questionnaire return access questions of the intelligent outbound system in the acquisition period are numbered according to the return access sequence, which are sequentially marked as 1,2,_e(H_e1,H_e2,...H_ex,...H_ey)，H_ex represents the e-th information corresponding to the x-th questionnaire back-visit question of the intelligent outbound call system in the acquisition period, e represents questionnaire back-visit question information, e is b1, b2, b3, b1, b2 and b3 respectively represent the back-visit sequence corresponding to each questionnaire back-visit question, the importance weight corresponding to each questionnaire back-visit question and the content corresponding to each questionnaire back-visit question.

Preferably, the questionnaire return visit result information includes the number of answer questions corresponding to the interviewee, the question number corresponding to each answer question of the interviewee, and the answer content corresponding to each answer question of the interviewee, so as to construct the questionnaire return visit result information set F of each call data of each outbound port_z ^d(F_z ^d1,F_z ^d2,...F_z ^di,...F_z ^dn)，F_z ^di represents the z-th return visit node corresponding to the ith call data of the d-th outbound port of the intelligent outbound system in the acquisition periodThe result information, z, represents return visit result information, and z ═ c1, c2, c3, c1, c2 and c3 represent the number of answer questions corresponding to the interviewee of the questionnaire, the question numbers corresponding to the answer questions of the interviewee of the questionnaire, and the answer contents corresponding to the answer questions of the interviewee of the questionnaire, respectively.

Preferably, the additional question information corresponding to the call of the questionnaire interviewee comprises the number of additional questions corresponding to the call of the questionnaire interviewee and the content corresponding to each additional question of the questionnaire interviewee, and further, according to the additional information corresponding to the call of the questionnaire interviewee, an additional question information set E of each call questionnaire interviewee at each outbound port is constructed_s ^d(E_s ^d1,E_s ^d2,...E_s ^di,...E_s ^dn)，E_s ^di represents the s-th extra question information corresponding to the interviewee of the questionnaire in the ith call data of the d-th outbound port in the acquisition period, s represents the extra question information for the interviewee of the questionnaire, and s is v1, v2, v1 and v2 respectively represent the number of extra questions corresponding to the call of the interviewee of the questionnaire and the content corresponding to each extra question of the interviewee of the questionnaire.

Preferably, the product negative information collection is used for extracting the voice information corresponding to the visited person of each call data questionnaire of each call port of the intelligent call-out system in the collection period according to the voice information corresponding to each call data of each call port of the intelligent call-out system in the collection period, further converting the voice information corresponding to the visited person of each call data questionnaire of each call port of the intelligent call-out system in the collection period into a text format, and counting the number corresponding to the product negative vocabulary and the type corresponding to each product negative vocabulary in the return visit of the intelligent call-out system questionnaire in the collection period.

Preferably, the cloud analysis platform analyzes the call data information of each outbound port of the intelligent outbound system, and is configured to analyze the call duration corresponding to each call data of each outbound port of the intelligent outbound system, acquire the call duration corresponding to each call data of each outbound port of the intelligent outbound system in the acquisition period, further acquire the number of interviewee participants corresponding to the intelligent outbound system in the acquisition period and the call duration corresponding to each interviewee participant of the questionnaire, compare the call duration corresponding to each interviewee participant of the intelligent outbound system in the acquisition period with the threshold of standard call duration corresponding to the call data of the outbound port, and further count that the call duration of each interviewee participant of the intelligent outbound system meets the influence coefficient.

Preferably, the cloud analysis platform analyzes the questionnaire return access results and the extra question information corresponding to each outbound system of the intelligent outbound system, and is used for analyzing the matching degree of answering questions of each interviewee and the attention degree of products of each questionnaire interviewee, so as to obtain questionnaire return access result information and extra question information corresponding to each interviewee of call data questionnaire of each outbound port of the intelligent outbound system, and further respectively count the influence coefficients of the matching degree of answering questions of each interviewee of the intelligent outbound system and the influence coefficients of the attention degree of products of each questionnaire of the intelligent outbound system.

Preferably, the cloud platform analyzes the negative information of the product and is used for analyzing the negative information corresponding to the product in the questionnaire return visit corresponding to the intelligent outbound system in the acquisition period, further acquiring the number corresponding to the negative vocabulary of the product and the type corresponding to the negative vocabulary of each product in the return visit of the intelligent outbound system questionnaire in the acquisition period, comparing the number corresponding to the negative vocabulary of each type of product with the standard number corresponding to the negative vocabulary of the product, and then counting the negative early warning influence coefficient of the return visit product of the intelligent outbound system questionnaire, comparing the negative early warning influence coefficient of the return visit product of the intelligent outbound system questionnaire with a preset negative early warning influence coefficient of a product, and if the negative early warning influence coefficient of the return visit product of the intelligent outbound system questionnaire is larger than the preset negative early warning influence coefficient of the product, recording the product as a negative information early warning product.

Preferably, the cloud analysis platform is further configured to perform comprehensive analysis on call data information, questionnaire return visit result information and additional problem information, further count product comprehensive trust degree influence coefficients of interviewees of the intelligent outbound call system, match and compare the product comprehensive trust degree influence coefficients of the interviewees of the intelligent outbound call system with product trust degree influence coefficients corresponding to the product trust levels, and acquire the product trust levels corresponding to the interviewees of the intelligent outbound call system in the acquisition period.

The invention has the beneficial effects that:

(1) according to the call data record analysis and management system of the intelligent outbound system based on data feature recognition, the questionnaire return visit result counting module, the additional problem information acquisition module and the product negative information acquisition module are combined with the cloud analysis platform to carry out detailed analysis on call record information corresponding to the questionnaire interviewee corresponding to each outbound port of the intelligent outbound system, so that the problems that the accurate screening of the questionnaire interviewee cannot be realized by the existing call data record analysis and management system of the intelligent outbound system, the positioning accuracy of the questionnaire interviewee cannot be effectively improved are effectively solved, the return visit efficiency of the product questionnaire is effectively improved, and the market competition advantage of the product is greatly improved.

(2) According to the invention, in the extra problem information acquisition module, the extra problem information corresponding to the call of the interviewee of the questionnaire in each call data of each outbound port is acquired, so that the accuracy of the classification of the interviewee of the questionnaire is greatly improved, and the comprehensiveness of the acquisition of the return visit information of the product is also greatly improved during call.

(3) According to the invention, at the early warning reminding terminal, the early warning is carried out on the negative information corresponding to the product, so that the response efficiency of a merchant on the negative information of the product is greatly improved, and the processing efficiency of the merchant on the negative information of the product is also greatly improved during communication.

(4) According to the invention, at the information sending terminal, the result of the cloud platform analysis is sent to the product return visit manager, so that the management efficiency of each questionnaire interviewee is greatly improved, and meanwhile, the viscosity of each questionnaire interviewee and the product is also greatly improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a schematic diagram showing the connection of modules of the system of the present invention.

Detailed Description

While the foregoing is directed to embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.

Referring to fig. 1, an intelligent outbound call system call data recording, analyzing and managing system based on data feature recognition comprises a call data information acquisition module, a questionnaire return visit result statistics module, an additional problem information acquisition module, a product negative information acquisition module, a cloud analysis platform, a database, an early warning reminding terminal and an information sending terminal;

the cloud analysis platform is respectively connected with a call data information acquisition module, a questionnaire return visit result statistics module, an additional problem information acquisition module, a product negative information acquisition module, a database, an early warning reminding terminal and an information sending terminal, the call data information acquisition module is respectively connected with the additional problem information acquisition module, the questionnaire return visit result statistics module and the product negative information acquisition module, and the questionnaire return visit information acquisition module is respectively connected with the questionnaire return visit result statistics module and the additional problem information acquisition module;

the call data information acquisition module is used for acquiring call data information corresponding to each outbound port of the intelligent outbound system in the acquisition period, further acquiring the number of outbound ports corresponding to the intelligent outbound system in the acquisition period, and numbering the outbound ports corresponding to the intelligent outbound system in the acquisition period according to a preset sequence, wherein the outbound ports are marked as 1,2,. j,. m in sequence;

specifically, the call data information corresponding to each outbound port in the acquisition period includes the number of call data corresponding to each outbound port of the intelligent outbound system in the acquisition period, the voice information corresponding to each call data of each outbound port, and the call duration corresponding to each call data of each outbound port, the call data corresponding to each outbound port in the acquisition period are numbered according to a preset sequence, which is sequentially marked as 1,2,. i.. n, and a call data information set T corresponding to each outbound port is constructed_w ^d(T_w ^d1,T_w ^d2,...T_w ^di,...T_w ^dn)，T_w ^di represents information corresponding to the ith call data of the d-th outbound port of the intelligent outbound system in the acquisition period, w represents call data information, and w is a1, a2, a1 and a2 respectively represent voice information corresponding to the call data and call duration corresponding to the call data.

The embodiment of the invention provides an effective cushion for the subsequent analysis of the call data information corresponding to each outbound port of the intelligent outbound system by acquiring the call data information corresponding to each outbound port in the acquisition period.

The cloud analysis platform analyzes the call data information of each outbound port of the intelligent outbound system, and is used for analyzing the call duration corresponding to each call data of each outbound port of the intelligent outbound system, acquiring the call duration corresponding to each call data of each outbound port of the intelligent outbound system in the acquisition period, further acquiring the number of the interviewee participants corresponding to the intelligent outbound system in the acquisition period and the call duration corresponding to each interviewee participant of the questionnaire, comparing the call duration corresponding to each interviewee participant of the intelligent outbound system in the acquisition period with the threshold value of the standard call duration corresponding to the call data of the outbound port, and further counting the call duration of each interviewee participant of the intelligent outbound system to meet the influence coefficient.

Wherein, the calculation formula of the influence coefficient that the call duration of each questionnaire of the intelligent outbound call system accords with the call duration of the interviewee isα_rThe communication time corresponding to the visited person of the r-th questionnaire of the intelligent outbound call system is shown to accord with the influence coefficient, a2_rThe call duration corresponding to the visited person of the r-th questionnaire of the intelligent outbound call system is shown as a2_{Standard of merit}The method comprises the steps of representing a threshold value of standard call duration corresponding to call data of an outbound port, representing the number of outbound ports of an intelligent outbound system in an acquisition period, representing the number of interviewees of an intelligent outbound system questionnaire, wherein r is 1, 2.

The questionnaire return visit information acquisition module is used for acquiring questionnaire return visit information corresponding to the product of the intelligent outbound system in the acquisition period;

specifically, the questionnaire return visit information corresponding to the product of the intelligent outbound system in the acquisition period includes the number of questions of the product questionnaire return visit of the intelligent outbound system in the acquisition period, the return visit sequence corresponding to each questionnaire return visit question, the importance weight corresponding to each questionnaire return visit question, and the content corresponding to each questionnaire return visit question, and then the questionnaire return visit questions of the intelligent outbound system in the acquisition period are numbered according to the return visit sequence, which are sequentially marked as 1,2,_e(H_e1,H_e2,...H_ex,...H_ey)，H_ex represents the e-th information corresponding to the x-th questionnaire back-visit question of the intelligent outbound call system in the acquisition period, e represents questionnaire back-visit question information, e is b1, b2, b3, b1, b2 and b3 respectively represent the back-visit sequence corresponding to each questionnaire back-visit question, the importance weight corresponding to each questionnaire back-visit question and the content corresponding to each questionnaire back-visit question.

The questionnaire return visit result counting module is used for counting return visit result information corresponding to each outbound port of the intelligent outbound system in the acquisition period so as to obtain the return visit result information corresponding to each outbound port of the intelligent outbound system in the acquisition period;

specifically, the questionnaire return visit result information includes the number of answer questions corresponding to the interviewee, the question number corresponding to each answer question of the interviewee, and the interviewee answerAnswering contents corresponding to each answering question of the visitor, and further constructing a call data questionnaire return visit result information set F of each outbound port_z ^d(F_z ^d1,F_z ^d2,...F_z ^di,...F_z ^dn)，F_z ^di represents the z-th return visit result information corresponding to the ith call data of the d-th outbound port of the intelligent outbound system in the acquisition period, z represents the return visit result information, and z is c1, c2, c3, c1, c2 and c3 respectively represent the number of answer questions corresponding to the interviewee of the questionnaire, the question numbers corresponding to the answer questions of the interviewee of the questionnaire and the answer contents corresponding to the answer questions of the interviewee of the questionnaire.

The specific statistical process of the questionnaire return visit result information comprises the following steps:

a1, acquiring the quantity of call data corresponding to each outbound port of the intelligent outbound system and call voice information corresponding to each call data in the acquisition period, and converting each call voice information of each outbound port of the intelligent outbound system in the acquisition period into a text format by a voice recognition technology;

a2, matching and comparing text information corresponding to each call of the intelligent outbound system in the acquisition period with content corresponding to a return visit question of the questionnaire of the intelligent outbound system in the acquisition period, and further counting the number of actual return visit problems corresponding to cloud customer services of each outbound port of the intelligent outbound system in the acquisition period and return visit time periods corresponding to each actual return visit problem;

a3, taking a return visit time period corresponding to each actual return visit problem of each cloud customer service of each outbound port of the intelligent outbound system in the acquisition period as a detection time period, and calling voice information corresponding to each detection time period of each piece of call data of each outbound port;

a4, identifying the tone of the speaker in the voice information of each detection time period of each piece of call data of each outbound port through a tone identification technology, if the tone corresponding to the speaker in the voice information of a certain detection time period of a certain piece of call data of a certain outbound port is consistent with the tone corresponding to the visitor of the call questionnaire of the port, further judging that the visitor of the call data of the outbound port answers the return visit question corresponding to the cloud service of the detection time period, further counting the number of the answer questions of the visitor of each call data questionnaire of each outbound port, the question content corresponding to each answer question and the answer content corresponding to each answer question, and further extracting the number corresponding to each answer question according to the question content corresponding to each answer question.

The cloud analysis platform analyzes the questionnaire return visit result corresponding to each outbound system of the intelligent outbound system, and is used for analyzing the matching degree of answering questions of each interviewee of each outbound port, acquiring the call data questionnaire return visit result information set of each outbound port, further acquiring the questionnaire return visit result information corresponding to each call data questionnaire of each outbound port, and counting the influence coefficient of the matching degree of answering questions of each interviewee of the intelligent outbound system.

Wherein, the specific analysis process of the questionnaire return visit result corresponding to each outbound system of the intelligent outbound system comprises the following steps:

b1, obtaining the return visit result information corresponding to each questionnaire interviewee of the intelligent outbound system in the acquisition period according to the return visit result information set of each call data questionnaire of each outbound port, and further obtaining the number of answer questions corresponding to the interviewee of the questionnaire, the question numbers corresponding to each answer question of the interviewee of the questionnaire and the answer contents corresponding to each answer question of the interviewee of the questionnaire;

b2, comparing the number of answer questions corresponding to the interviewee of each questionnaire of the intelligent outbound system with the number of questionnaire return questions corresponding to the intelligent outbound system, and further counting the influence coefficients of the integrity of the access questions of the interviewee of each questionnaire of the intelligent outbound system;

wherein, the calculation formula of the influence coefficient of the integrity of each questionnaire of the intelligent outbound system influenced by the return visit problem of the visitor isβ_rIndicating the intelligent outbound systemThe r-th questionnaire is influenced by the influence coefficient of the integrity of the return visit question corresponding to the visitor, c2_rThe answer question number corresponding to the r-th questionnaire interviewee of the intelligent outbound system is represented, and the question number corresponding to the intelligent outbound system is represented by y.

B3, according to the question numbers corresponding to the answer questions of the interviewee of each questionnaire of the intelligent outbound system, obtaining the importance weights corresponding to the answer questions of the interviewee of each questionnaire of the intelligent outbound system, further obtaining the comprehensive importance weights corresponding to the answer questions of the interviewee of each questionnaire of the intelligent outbound system, comparing the comprehensive importance weights corresponding to the answer questions of the interviewee of each questionnaire of the intelligent outbound system with the comprehensive weights of the answer questions of the interviewee of the intelligent outbound system, and further counting the influence coefficients of the importance of the answer questions of the interviewee of each questionnaire of the intelligent outbound system.

Wherein, the calculation formula of the importance influence coefficient of each questionnaire of the intelligent outbound system on the answer questions of the interviewee isδ_rThe importance influence coefficient Z of the answer questions corresponding to the visitor on the ith questionnaire of the intelligent outbound system_rRepresents the comprehensive importance weight corresponding to the answer question of the r-th questionnaire interviewee of the intelligent outbound system, b2_tAnd the importance weight of the tth questionnaire return question of the intelligent outbound system is represented, t represents the number of the questionnaire return question of the intelligent outbound system, and t is 1, 2.

B4, according to the answer content corresponding to each answer question of each interviewee of the intelligent outbound call system, segmenting the answer content corresponding to each answer question of each interviewee of the intelligent outbound call system, removing stop words in each answer content, further constructing and recording each answer question segmentation set of each interviewee as R, further extracting the segmentation set corresponding to the preset answer result of each interviewee return question from the database, matching and comparing the segmentation corresponding to each answer question of each interviewee of the intelligent outbound call system with the segmentation corresponding to the preset answer result of each answer question of each interviewee of the intelligent outbound call system, and further counting the accurate influence coefficient of each answer question of each interviewee of the intelligent outbound call system.

Wherein, the calculation formula of the precision influence coefficient of each questionnaire of the intelligent outbound system answered by the interviewer is The accurate influence coefficient of answer questions corresponding to the R-th questionnaire of the intelligent outbound call system, R_u ^rA participle set, K, corresponding to the u-th answer question of the r-th questionnaire interviewee of the intelligent outbound system_uThe method includes the steps that a word segmentation set corresponding to a preset answer result of the u-th answer question of an interviewee of the intelligent outbound call system is shown, u represents answer question numbers of the interviewee, and u is 1, 2.

And B5, according to the counted influence coefficients of the return visit question integrity of each questionnaire of the intelligent outbound system, the influence coefficients of the importance of the answer questions of each questionnaire of the intelligent outbound system and the influence coefficients of the accuracy of the answer questions of each questionnaire of the intelligent outbound system, the influence coefficients of the answer question fitness of each questionnaire of the intelligent outbound system are counted.

Wherein, the calculation formula of the influence coefficient of each questionnaire of the intelligent outbound system on the matching degree of the interviewee answering questions isγ_rAnd the influence coefficient of the matching degree of the answer questions corresponding to the visitor is shown for the r-th questionnaire of the intelligent outbound system.

The extra question information acquisition module is used for acquiring extra question information corresponding to the call of the interviewee in each call data of each outbound port according to the call data information and the questionnaire return visit information corresponding to each outbound port of the intelligent outbound system in the acquisition period;

specifically, the extra question information corresponding to the call of the questionnaire interviewee comprises the number of extra questions corresponding to the call of the questionnaire interviewee and the content corresponding to each extra question of the questionnaire interviewee, and further, according to the extra information corresponding to the call of the questionnaire interviewee, an extra question information set of each call questionnaire interviewee of each outbound port is constructedE_s ^di represents the s-th extra question information corresponding to the interviewee of the questionnaire in the ith call data of the d-th outbound port in the acquisition period, s represents the extra question information for the interviewee of the questionnaire, and s is v1, v2, v1 and v2 respectively represent the number of extra questions corresponding to the call of the interviewee of the questionnaire and the content corresponding to each extra question of the interviewee of the questionnaire.

The specific acquisition process of the extra question information corresponding to the call of the interviewee of the questionnaire comprises the following steps: acquiring voice information corresponding to each piece of call data of each outbound port in the acquisition cycle, further extracting voice information corresponding to a questionnaire interviewee in the voice information of each piece of call data of each outbound port by using a tone recognition technology, recording the voice information corresponding to the questionnaire interviewee as target detection voice information, further dividing target detection voice information of each piece of call data of each outbound port into target detection sentences according to corresponding pause time points in the target detection voice information, further identifying tone corresponding to each target detection sentence of each piece of call data of each outbound port by using a tone recognition technology, further acquiring tone types corresponding to each target detection sentence of each piece of call data of each outbound port, and if the tone type corresponding to a certain target detection sentence of an interviewee of a certain outbound port call data questionnaire is a question type, judging that the interviewee of the questionnaire has additional problems, and further counting the number of additional questions corresponding to the visited personnel of each call data questionnaire of each outbound port and the content corresponding to each additional question.

In one embodiment, the mood types include query mood, statement mood, and the like.

In the embodiment of the invention, the extra problem information corresponding to the call of the interviewee in each call data of each outbound port is acquired by the extra problem information acquisition module, so that the accuracy of classifying the interviewee of each questionnaire is greatly improved, and the comprehensiveness of obtaining the return visit information of the product is also greatly improved during communication.

The cloud analysis platform is used for analyzing the attention degree of each interviewee to the product of each interviewee in each call data of each call port of the intelligent call-out system, acquiring an additional problem information set of each call interviewee, further acquiring additional problem information corresponding to each call data interviewee of each call port of the intelligent call-out system, and counting the influence coefficient of the attention degree of each interviewee product of each call interviewee of the intelligent call-out system.

Wherein, the extra problem analysis corresponding to each call data questionnaire interviewee of each outbound port is used for obtaining the extra problem quantity corresponding to each interviewee of the intelligent outbound system and the content corresponding to each extra problem according to the extra problem quantity of each call questionnaire interviewee of each outbound port and the content corresponding to each extra problem, comparing the extra problem quantity corresponding to each interviewee of the intelligent outbound system with the average extra problem quantity corresponding to the interviewee of the questionnaire, further counting the influence coefficient of the product attention of each interviewee of the intelligent outbound system, simultaneously matching and comparing the content corresponding to each extra problem of each interviewee of the intelligent outbound system with the content corresponding to each type of product associated problem in the database, if a certain interviewee extra problem does not belong to the problem stored in the database, and recording the additional problems as the problems to be solved, and counting the number corresponding to the problems to be solved, the content corresponding to each problem to be solved and the interviewee who provides the questionnaire corresponding to each problem to be solved.

Wherein, the calculation formula of the influence coefficient of each questionnaire of the intelligent outbound system on the product attention of the interviewee isμ_rThe r number questionnaire of the intelligent outbound system is shown to be influenced by the product attention degree of the visitor, v1_rIndicating the number of additional questions corresponding to the visited person of the r-th questionnaire of the intelligent outbound system,indicating the average number of additional questions corresponding to the interviewee,h represents the number of interviewee questionnaires corresponding to the intelligent outbound system in the acquisition period.

The product negative information acquisition module is used for extracting negative information corresponding to the product according to the call data information corresponding to each outbound port of the intelligent outbound system in the acquisition period;

specifically, the product negative information collection is used for extracting the voice information corresponding to the visited person of each call data questionnaire of each call port of the intelligent call-out system in the collection period according to the voice information corresponding to each call data of each call port of the intelligent call-out system in the collection period, further converting the voice information corresponding to the visited person of each call data questionnaire of each call port of the intelligent call-out system in the collection period into a text format, and counting the number corresponding to the product negative vocabulary and the type corresponding to each product negative vocabulary in the return visit of the intelligent call-out system questionnaire in the collection period.

Wherein the product negative vocabulary statistical process is as follows: removing corresponding stop words in text information of each call data questionnaire of each outbound port of the intelligent outbound system in the acquisition period, then the text information corresponding to the interviewee of each outbound data questionnaire of each outbound port of the intelligent outbound system in the processed acquisition period is segmented, matching and comparing each participle of each call data questionnaire of each call port of the intelligent call-out system in the acquisition period with each type of negative vocabulary in the database, counting the number corresponding to the product negative vocabulary and the type corresponding to each product negative vocabulary in the text information of each call data questionnaire of each call port of the intelligent call-out system in the acquisition period, and further acquiring the quantity of negative vocabularies corresponding to products in the return visit of the intelligent outbound system questionnaire in the acquisition period and the types corresponding to the negative vocabularies.

In one embodiment, the product negative vocabulary corresponding types include quality type, safety type and the like.

The cloud platform analyzes the negative information of the product and is used for analyzing the negative information corresponding to the product in the questionnaire return visit corresponding to the intelligent outbound system in the acquisition period, further obtains the number corresponding to the negative vocabulary of the product in the questionnaire return visit of the intelligent outbound system in the acquisition period and the type corresponding to the negative vocabulary of the product, compares the number corresponding to the negative vocabulary of the product of each type with the standard number corresponding to the negative vocabulary of the product, further counts the negative early warning influence coefficient of the product in the questionnaire return visit of the intelligent outbound system, compares the negative early warning influence coefficient of the product in the questionnaire return visit of the intelligent outbound system with the preset negative early warning influence coefficient of the product, and marks the product as a negative information early warning product if the negative early warning influence coefficient of the product in the questionnaire return visit of the intelligent outbound system is greater than the preset negative early warning influence coefficient of the product.

Wherein the negative early warning influence coefficient calculation formula of the intelligent outbound call system questionnaire return visit product isXi represents the product negative early warning influence coefficient corresponding to the return visit of the intelligent outbound system questionnaire, F_kRepresenting the quantity corresponding to the negative vocabulary of the kth type of product revisited by the questionnaire of the intelligent outbound system, F_{Standard of merit}The standard number corresponding to the negative vocabulary of the product is shown, and k is the number of the type of the negative vocabulary of the product. K1, 2,. l,. f, f denotes the number of product negative vocabulary types.

The cloud analysis platform is further used for comprehensively analyzing call data information, questionnaire return visit result information and extra question information, further counting product comprehensive trust degree influence coefficients of the interviewees of the intelligent outbound system according to the counted call duration of the interviewees of the intelligent outbound system, the interviewee answer question fitness influence coefficients of the questionnaires of the intelligent outbound system and the interviewee product trust degree influence coefficients of the questionnaires of the intelligent outbound system, matching and comparing the product comprehensive trust degree influence coefficients of the interviewees of the questionnaires of the intelligent outbound system with the product trust degree influence coefficients corresponding to the product trust levels, and obtaining the product trust levels corresponding to the interviewees of the intelligent outbound system in the acquisition period.

Wherein, the calculation formula of the influence coefficient of the comprehensive trust degree of each questionnaire of the intelligent outbound call system isλ_rAnd showing the influence coefficient of the comprehensive trust degree of the product corresponding to the visitor on the nth questionnaire of the intelligent outbound system, wherein epsilon shows the product trust correction coefficient.

The embodiment of the invention effectively solves the problems that the prior call data record analysis management system of the intelligent outbound system can not realize the accurate screening of the interviewee of the questionnaire, and can not effectively improve the accuracy of the positioning of the interviewee of each questionnaire by carrying out the detailed analysis on the call record information corresponding to the interviewee of each outbound port of the intelligent outbound system, thereby effectively improving the return visit efficiency of the product questionnaire and greatly improving the market competition advantages of the product.

The database is used for storing a threshold value of standard call duration corresponding to call data of an outbound port of the intelligent outbound system, contents corresponding to problems associated with various types of products, standard quantity corresponding to negative vocabularies of the products and a preset negative early warning influence coefficient of the products;

the early warning reminding terminal is used for sending an early warning signal to the product sales management personnel when the product negative information of the intelligent outbound system reaches an early warning value in the acquisition period;

specifically, when the negative early warning influence coefficient of the return visit product of the intelligent outbound call system is larger than the preset negative early warning influence coefficient of the product, an early warning signal is sent to the product sales manager.

According to the embodiment of the invention, at the early warning reminding terminal, the early warning is carried out on the negative information corresponding to the product, so that the response efficiency of a merchant on the negative information of the product is greatly improved, and the processing efficiency of the merchant on the negative information of the product is also greatly improved during communication.

And the information sending terminal is used for sending the result of the analysis of the cloud analysis platform to the product return visit management personnel.

Specifically, the information sending terminal is configured to send the number of the problems to be solved, the content corresponding to each problem to be solved, the interviewee who presents the questionnaire corresponding to each problem to be solved, and the product trust level corresponding to each interviewee of the intelligent outbound call system in the acquisition period to the product return visit manager.

According to the embodiment of the invention, at the information sending terminal, the result of the cloud platform analysis is sent to the product return visit manager, so that the management efficiency of each questionnaire interviewee is greatly improved, and meanwhile, the viscosity of each questionnaire interviewee and the product is also greatly improved.

The foregoing is merely exemplary and illustrative of the principles of the present invention and various modifications, additions and substitutions of the specific embodiments described herein may be made by those skilled in the art without departing from the principles of the present invention or exceeding the scope of the claims set forth herein.