CN113360627A - Online customer service real-time distribution method and system - Google Patents

Online customer service real-time distribution method and system Download PDF

Info

Publication number
CN113360627A
CN113360627A CN202110761668.8A CN202110761668A CN113360627A CN 113360627 A CN113360627 A CN 113360627A CN 202110761668 A CN202110761668 A CN 202110761668A CN 113360627 A CN113360627 A CN 113360627A
Authority
CN
China
Prior art keywords
customer service
feature
session
matching
idle
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110761668.8A
Other languages
Chinese (zh)
Other versions
CN113360627B (en
Inventor
陈�光
李凯
徐笠森
刘焕秋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Ronglian Qimo Technology Co ltd
Original Assignee
Beijing Ronglian Qimo Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Ronglian Qimo Technology Co ltd filed Critical Beijing Ronglian Qimo Technology Co ltd
Priority to CN202110761668.8A priority Critical patent/CN113360627B/en
Publication of CN113360627A publication Critical patent/CN113360627A/en
Application granted granted Critical
Publication of CN113360627B publication Critical patent/CN113360627B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/332—Query formulation
    • G06F16/3329—Natural language query formulation or dialogue systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/01—Customer relationship services

Abstract

The invention provides an online customer service real-time distribution method and system, wherein the method comprises the following steps: step S1: acquiring a plurality of first session queues; step S2: sequentially distributing the first conversations in the first conversation queue to the idle customer service; step S3: summarizing first sessions which are not successfully distributed in a first session queue to obtain a first session set; step S4: preprocessing the first session set to obtain a plurality of second session queues; step S5: and sequentially allocating the second sessions in the second session queue to the idle customer service based on the allocation strategy. The online customer service real-time distribution method and the system can timely deal with and process when a large number of visitors initiate session requests, and complete customer service distribution.

Description

Online customer service real-time distribution method and system
Technical Field
The invention relates to the technical field of online customer service distribution, in particular to a method and a system for online customer service real-time distribution.
Background
At present, when a few visitors initiate session requests, the customer service allocation is simple, but when a large number of visitors initiate session requests, how to allocate the customer service is complex, and therefore a processing method is urgently needed.
Disclosure of Invention
One of the objectives of the present invention is to provide a method and a system for online real-time customer service distribution, which can respond to a large number of visitors in time when initiating a session request, thereby completing customer service distribution.
The embodiment of the invention provides an online customer service real-time distribution method, which comprises the following steps:
step S1: acquiring a plurality of first session queues;
step S2: sequentially distributing the first conversations in the first conversation queue to the idle customer service;
step S3: summarizing first sessions which are not successfully distributed in a first session queue to obtain a first session set;
step S4: preprocessing the first session set to obtain a plurality of second session queues;
step S5: and sequentially allocating the second sessions in the second session queue to the idle customer service based on the allocation strategy.
Preferably, step S4: preprocessing the first session set to obtain a plurality of second session queues, including:
equally dividing or approximately equally dividing the first sessions in the first session set into a plurality of third session queues;
and sequentially placing at least one third conversation of the member users in the third conversation queue at the forefront end of the third conversation queue to obtain a second conversation queue.
Preferably, the allocation policy includes: a rotation distribution strategy, a random distribution strategy, a weight distribution strategy and a workload distribution strategy;
wherein, the alternate distribution strategy comprises:
establishing an idle customer service queue, and adding idle customer service to the idle customer service queue;
allocating the second session at the forefront in the second session queue to the idle customer service at the forefront of the idle customer queue;
when the next second session is allocated, allocating the next second session to the next idle customer service of the last allocated idle customer service in the idle customer service queue;
the random allocation strategy comprises the following steps:
based on a preset workload analysis model, carrying out average distribution analysis on the workload of each idle customer service to obtain a target idle customer service, and allocating a second session to be allocated to the target idle customer service;
the weight distribution strategy comprises the following steps:
acquiring the weight of the idle customer service, and allocating a second session to be allocated to the idle customer service corresponding to the maximum value of the weight;
the workload distribution strategy comprises the following steps:
and acquiring the workload of the idle customer service, and allocating the second session to be allocated to the idle customer service corresponding to the minimum workload.
Preferably, the obtaining of the weight of the idle customer service includes:
determining a plurality of associated customer services associated with the idle customer service based on a preset associated customer service library;
respectively acquiring a plurality of first reception records of idle customer service and a plurality of second reception records of associated customer service;
respectively establishing a first time axis and a second time axis;
expanding the first reception record on a first time axis to obtain a plurality of first record items;
expanding the second reception record on a second time axis to obtain a plurality of second record items;
performing feature extraction on the first record item to obtain a plurality of first features;
acquiring a preset primary screening feature library, matching the first feature with a second feature in the primary screening feature library, and outputting a first matching item if the first feature is matched with the second feature in the primary screening feature library, wherein the first matching item comprises: matching the first and second matched features;
determining a first position of a first record item corresponding to the first characteristic matched with the first record item on a first time axis;
acquiring first record items with a first quantity in front of a first position and/or in a preset mode on a first time axis, and taking the first record items as third record items;
extracting the features of the third record item to obtain a plurality of third features;
acquiring a preset malicious feature library, matching the third feature with a fourth feature in the malicious feature library, and outputting a second matching item if the matching is consistent, wherein the second matching item comprises: matching the third characteristic and the fourth characteristic which are in line, matching a first characteristic type corresponding to the fourth characteristic which is in line, and matching a first matching degree between the third characteristic and the fourth characteristic which are in line;
determining a second position of a third record item corresponding to the matched third feature on the first time axis;
extracting a second position farthest from the first position in front of or behind the second position as a third position;
determining a fourth position on the second time axis corresponding to the third position, and determining a fifth position on the second time axis corresponding to the first position;
acquiring a second record item with a second quantity preset in the direction of the preset fourth position on a second time axis, and taking the second record item as a fourth record item;
acquiring a second recording item between a fourth position and a fifth position on a second time axis, and taking the second recording item as a fifth recording item;
summarizing the fourth record items and the fifth record items, and performing duplication elimination to obtain a plurality of sixth record items;
performing feature extraction on the sixth record item to obtain a plurality of fifth features;
matching the fifth feature with a fourth feature in the malicious feature library, and outputting a third matching item if the matching is matched, wherein the third matching item comprises: matching the fifth characteristic and the fourth characteristic which are in line, matching a second characteristic type corresponding to the fourth characteristic which is in line, and matching a second matching degree between the fifth characteristic and the fourth characteristic which are in line;
determining a first down-regulation amplitude which is commonly corresponding to the first feature type and the first matching degree and determining a second down-regulation amplitude which is commonly corresponding to the second feature type and the second matching degree based on a preset down-regulation amplitude library;
integrating a third record item corresponding to a third feature matched and matched with the fourth feature and a sixth record item corresponding to a fifth feature matched and matched with the fourth feature to obtain reception record data;
extracting target records which meet a preset first condition in the reception record data, and summarizing the number of the target records;
determining third downward adjustment amplitudes corresponding to the quantity based on the downward adjustment amplitude library;
summarizing the first downward regulation amplitude, the second downward regulation amplitude and the third downward regulation amplitude to obtain a downward regulation amplitude sum;
acquiring a preset initial weight, and performing down-regulation on the initial weight based on the down-regulation amplitude;
the adjusted initial weight is used as the weight of the idle customer service to finish the acquisition;
wherein the directions include: front or rear, away from the fifth position;
the first condition includes:
the same customer is received by different receptionists, and the receptionist includes: idle customer service or associated customer service.
Preferably, the online customer service real-time distribution method further includes:
step S6: summarizing second sessions which cannot be successfully distributed in a second session queue to obtain a second session set, summarizing the second sessions which wait for distribution in the second session set for more than or equal to a preset threshold value to obtain a third session set, establishing an outsourcing customer service queue, and sequentially distributing the second sessions in the third session set to outsourcing customer services in the outsourcing customer service queue;
wherein, establish outsourcing customer service queue, include:
acquiring a preset outsourcing customer service set, and distributing a preset questionnaire to outsourcing customer services in the outsourcing customer service set;
acquiring a survey result input by the outsourcing customer service based on a questionnaire, wherein the survey result comprises: the accuracy rate;
determining the satisfaction degree of outsourcing customer service based on a preset satisfaction degree library;
determining the total number of second sessions in a third session set, and determining tendency proportion corresponding to the total number based on a preset tendency proportion library;
extracting outsourced customer services meeting preset second conditions in the outsourced customer service set, and adding the outsourced customer services into an outsourced customer service queue;
when the outsourced customer service meeting the second condition is added into the outsourced customer service queue, the establishment is completed;
wherein the second condition comprises:
the ratio of the accuracy of outsourcing customer service to the satisfaction degree accords with the trend proportion.
The embodiment of the invention provides an online customer service real-time distribution system, which comprises:
the acquisition module is used for acquiring a plurality of first session queues;
the first allocation module is used for sequentially allocating the first sessions in the first session queue to the idle customer service;
the summarizing module is used for summarizing the first session which is not successfully distributed in the first session queue to obtain a first session set;
the preprocessing module is used for preprocessing the first session set to obtain a plurality of second session queues;
and the second distribution module is used for sequentially distributing the second sessions in the second session queue to the idle customer service based on the distribution strategy.
Preferably, the preprocessing module performs the following operations:
equally dividing or approximately equally dividing the first sessions in the first session set into a plurality of third session queues;
and sequentially placing at least one third conversation of the member users in the third conversation queue at the forefront end of the third conversation queue to obtain a second conversation queue.
Preferably, the allocation policy includes: a rotation distribution strategy, a random distribution strategy, a weight distribution strategy and a workload distribution strategy;
wherein, the alternate distribution strategy comprises:
establishing an idle customer service queue, and adding idle customer service to the idle customer service queue;
allocating the second session at the forefront in the second session queue to the idle customer service at the forefront of the idle customer queue;
when the next second session is allocated, allocating the next second session to the next idle customer service of the last allocated idle customer service in the idle customer service queue;
the random allocation strategy comprises the following steps:
based on a preset workload analysis model, carrying out average distribution analysis on the workload of each idle customer service to obtain a target idle customer service, and allocating a second session to be allocated to the target idle customer service;
the weight distribution strategy comprises the following steps:
acquiring the weight of the idle customer service, and allocating a second session to be allocated to the idle customer service corresponding to the maximum value of the weight;
the workload distribution strategy comprises the following steps:
and acquiring the workload of the idle customer service, and allocating the second session to be allocated to the idle customer service corresponding to the minimum workload.
Preferably, the second allocating module performs the following operations:
determining a plurality of associated customer services associated with the idle customer service based on a preset associated customer service library;
respectively acquiring a plurality of first reception records of idle customer service and a plurality of second reception records of associated customer service;
respectively establishing a first time axis and a second time axis;
expanding the first reception record on a first time axis to obtain a plurality of first record items;
expanding the second reception record on a second time axis to obtain a plurality of second record items;
performing feature extraction on the first record item to obtain a plurality of first features;
acquiring a preset primary screening feature library, matching the first feature with a second feature in the primary screening feature library, and outputting a first matching item if the first feature is matched with the second feature in the primary screening feature library, wherein the first matching item comprises: matching the first and second matched features;
determining a first position of a first record item corresponding to the first characteristic matched with the first record item on a first time axis;
acquiring first record items with a first quantity in front of a first position and/or in a preset mode on a first time axis, and taking the first record items as third record items;
extracting the features of the third record item to obtain a plurality of third features;
acquiring a preset malicious feature library, matching the third feature with a fourth feature in the malicious feature library, and outputting a second matching item if the matching is consistent, wherein the second matching item comprises: matching the third characteristic and the fourth characteristic which are in line, matching a first characteristic type corresponding to the fourth characteristic which is in line, and matching a first matching degree between the third characteristic and the fourth characteristic which are in line;
determining a second position of a third record item corresponding to the matched third feature on the first time axis;
extracting a second position farthest from the first position in front of or behind the second position as a third position;
determining a fourth position on the second time axis corresponding to the third position, and determining a fifth position on the second time axis corresponding to the first position;
acquiring a second record item with a second quantity preset in the direction of the preset fourth position on a second time axis, and taking the second record item as a fourth record item;
acquiring a second recording item between a fourth position and a fifth position on a second time axis, and taking the second recording item as a fifth recording item;
summarizing the fourth record items and the fifth record items, and performing duplication elimination to obtain a plurality of sixth record items;
performing feature extraction on the sixth record item to obtain a plurality of fifth features;
matching the fifth feature with a fourth feature in the malicious feature library, and outputting a third matching item if the matching is matched, wherein the third matching item comprises: matching the fifth characteristic and the fourth characteristic which are in line, matching a second characteristic type corresponding to the fourth characteristic which is in line, and matching a second matching degree between the fifth characteristic and the fourth characteristic which are in line;
determining a first down-regulation amplitude which is commonly corresponding to the first feature type and the first matching degree and determining a second down-regulation amplitude which is commonly corresponding to the second feature type and the second matching degree based on a preset down-regulation amplitude library;
integrating a third record item corresponding to a third feature matched and matched with the fourth feature and a sixth record item corresponding to a fifth feature matched and matched with the fourth feature to obtain reception record data;
extracting target records which meet a preset first condition in the reception record data, and summarizing the number of the target records;
determining third downward adjustment amplitudes corresponding to the quantity based on the downward adjustment amplitude library;
summarizing the first downward regulation amplitude, the second downward regulation amplitude and the third downward regulation amplitude to obtain a downward regulation amplitude sum;
acquiring a preset initial weight, and performing down-regulation on the initial weight based on the down-regulation amplitude;
the adjusted initial weight is used as the weight of the idle customer service to finish the acquisition;
wherein the directions include: front or rear, away from the fifth position;
the first condition includes:
the same customer is received by different receptionists, and the receptionist includes: idle customer service or associated customer service.
Preferably, the online customer service real-time distribution system further includes:
the third distribution module is used for summarizing the second sessions which are not successfully distributed in the second session queue to obtain a second session set, summarizing the second sessions which are waiting for distribution in the second session set and have the distribution time more than or equal to a preset threshold value to obtain a third session set, establishing an outsourcing customer service queue, and sequentially distributing the second sessions in the third session set to outsourcing customer services in the outsourcing customer service queue;
the third distribution module performs the following operations:
acquiring a preset outsourcing customer service set, and distributing a preset questionnaire to outsourcing customer services in the outsourcing customer service set;
acquiring a survey result input by the outsourcing customer service based on a questionnaire, wherein the survey result comprises: the accuracy rate;
determining the satisfaction degree of outsourcing customer service based on a preset satisfaction degree library;
determining the total number of second sessions in a third session set, and determining tendency proportion corresponding to the total number based on a preset tendency proportion library;
extracting outsourced customer services meeting preset second conditions in the outsourced customer service set, and adding the outsourced customer services into an outsourced customer service queue;
when the outsourced customer service meeting the second condition is added into the outsourced customer service queue, the establishment is completed;
wherein the second condition comprises:
the ratio of the accuracy of outsourcing customer service to the satisfaction degree accords with the trend proportion.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a flow chart of a method for real-time online customer service distribution according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an online customer service real-time distribution system according to an embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
The embodiment of the invention provides an online customer service real-time distribution method, as shown in fig. 1, comprising the following steps:
step S1: acquiring a plurality of first session queues;
step S2: sequentially distributing the first conversations in the first conversation queue to the idle customer service;
step S3: summarizing first sessions which are not successfully distributed in a first session queue to obtain a first session set;
step S4: preprocessing the first session set to obtain a plurality of second session queues;
step S5: and sequentially allocating the second sessions in the second session queue to the idle customer service based on the allocation strategy.
The working principle and the beneficial effects of the technical scheme are as follows:
setting a plurality of blank queues, and adding visitors into the blank queues to obtain a plurality of first session queues when a large number of visitors initiate session requests; firstly, sequentially allocating first sessions in a first session queue to idle customer service, and then summarizing the first sessions which cannot be served by the customer service to obtain a first session set; preprocessing the first session set (reasonably splitting) to obtain a plurality of second session queues; and sequentially allocating the second sessions in the second session queue to the idle customer service based on the allocation strategy.
According to the embodiment of the invention, when a large number of visitors initiate session requests, the processing can be responded in time, and the customer service distribution is completed.
The embodiment of the invention provides an online customer service real-time distribution method, which comprises the following steps of S4: preprocessing the first session set to obtain a plurality of second session queues, including:
equally dividing or approximately equally dividing the first sessions in the first session set into a plurality of third session queues;
and sequentially placing at least one third conversation of the member users in the third conversation queue at the forefront end of the third conversation queue to obtain a second conversation queue.
The working principle and the beneficial effects of the technical scheme are as follows:
evenly divide the first sessions in the first set of sessions [ e.g.: 9 first sessions, equally divided into 3 queues, 3 per queue or approximately equally divided [ for example: and (3) approximately evaluating 10 first conversations and dividing the first conversations into 3 queues, wherein the number of the conversations in the queues is 3, 3 and 4 in turn to form a plurality of third conversation queues, and the member users in the third conversation queues are arranged to the forefront and are preferentially distributed.
The embodiment of the invention provides an online customer service real-time distribution method, and a distribution strategy comprises the following steps: a rotation distribution strategy, a random distribution strategy, a weight distribution strategy and a workload distribution strategy;
wherein, the alternate distribution strategy comprises:
establishing an idle customer service queue, and adding idle customer service to the idle customer service queue;
allocating the second session at the forefront in the second session queue to the idle customer service at the forefront of the idle customer queue;
when the next second session is allocated, allocating the next second session to the next idle customer service of the last allocated idle customer service in the idle customer service queue;
the random allocation strategy comprises the following steps:
based on a preset workload analysis model, carrying out average distribution analysis on the workload of each idle customer service to obtain a target idle customer service, and allocating a second session to be allocated to the target idle customer service;
the weight distribution strategy comprises the following steps:
acquiring the weight of the idle customer service, and allocating a second session to be allocated to the idle customer service corresponding to the maximum value of the weight;
the workload distribution strategy comprises the following steps:
and acquiring the workload of the idle customer service, and allocating the second session to be allocated to the idle customer service corresponding to the minimum workload.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset workload analysis module specifically comprises: a model generated after a large number of records of manual work which are analyzed for carrying out average distribution on the workload of each customer service (the workload of each customer service is almost) is utilized for learning, and after the model is analyzed, target idle customer service needing conversation distribution is output;
and a plurality of customer service distribution strategies are set, so that the response capability of the system can be improved, and the customer service distribution can be better completed.
The embodiment of the invention provides an online customer service real-time distribution method, which obtains the weight of idle customer service and comprises the following steps:
determining a plurality of associated customer services associated with the idle customer service based on a preset associated customer service library;
respectively acquiring a plurality of first reception records of idle customer service and a plurality of second reception records of associated customer service;
respectively establishing a first time axis and a second time axis;
expanding the first reception record on a first time axis to obtain a plurality of first record items;
expanding the second reception record on a second time axis to obtain a plurality of second record items;
performing feature extraction on the first record item to obtain a plurality of first features;
acquiring a preset primary screening feature library, matching the first feature with a second feature in the primary screening feature library, and outputting a first matching item if the first feature is matched with the second feature in the primary screening feature library, wherein the first matching item comprises: matching the first and second matched features;
determining a first position of a first record item corresponding to the first characteristic matched with the first record item on a first time axis;
acquiring first record items with a first quantity in front of a first position and/or in a preset mode on a first time axis, and taking the first record items as third record items;
extracting the features of the third record item to obtain a plurality of third features;
acquiring a preset malicious feature library, matching the third feature with a fourth feature in the malicious feature library, and outputting a second matching item if the matching is consistent, wherein the second matching item comprises: matching the third characteristic and the fourth characteristic which are in line, matching a first characteristic type corresponding to the fourth characteristic which is in line, and matching a first matching degree between the third characteristic and the fourth characteristic which are in line;
determining a second position of a third record item corresponding to the matched third feature on the first time axis;
extracting a second position farthest from the first position in front of or behind the second position as a third position;
determining a fourth position on the second time axis corresponding to the third position, and determining a fifth position on the second time axis corresponding to the first position;
acquiring a second record item with a second quantity preset in the direction of the preset fourth position on a second time axis, and taking the second record item as a fourth record item;
acquiring a second recording item between a fourth position and a fifth position on a second time axis, and taking the second recording item as a fifth recording item;
summarizing the fourth record items and the fifth record items, and performing duplication elimination to obtain a plurality of sixth record items;
performing feature extraction on the sixth record item to obtain a plurality of fifth features;
matching the fifth feature with a fourth feature in the malicious feature library, and outputting a third matching item if the matching is matched, wherein the third matching item comprises: matching the fifth characteristic and the fourth characteristic which are in line, matching a second characteristic type corresponding to the fourth characteristic which is in line, and matching a second matching degree between the fifth characteristic and the fourth characteristic which are in line;
determining a first down-regulation amplitude which is commonly corresponding to the first feature type and the first matching degree and determining a second down-regulation amplitude which is commonly corresponding to the second feature type and the second matching degree based on a preset down-regulation amplitude library;
integrating a third record item corresponding to a third feature matched and matched with the fourth feature and a sixth record item corresponding to a fifth feature matched and matched with the fourth feature to obtain reception record data;
extracting target records which meet a preset first condition in the reception record data, and summarizing the number of the target records;
determining third downward adjustment amplitudes corresponding to the quantity based on the downward adjustment amplitude library;
summarizing the first downward regulation amplitude, the second downward regulation amplitude and the third downward regulation amplitude to obtain a downward regulation amplitude sum;
acquiring a preset initial weight, and performing down-regulation on the initial weight based on the down-regulation amplitude;
the adjusted initial weight is used as the weight of the idle customer service to finish the acquisition;
wherein the directions include: front or rear, away from the fifth position;
the first condition includes:
the same customer is received by different receptionists, and the receptionist includes: idle customer service or associated customer service.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset associated customer service library specifically comprises the following steps: a database, which stores the association relation (the association relation is divided into active association and passive association) with each customer service, the active association means that the customer service sets a certain customer service as association, the passive association means that other customer services set themselves as association; the preset primary screening feature library specifically comprises the following steps: a database having stored therein a plurality of prescreening characteristics [ for example: that is, the characteristics of vital energy; the preset first number specifically comprises: for example, 10; the preset malicious feature library specifically comprises the following steps: a database having stored therein a number of malicious features [ for example: bad speech characteristics in customer service level); the preset second number is specifically as follows: for example, 20; the preset down-regulation amplitude library specifically comprises: a database, in which the down-regulation amplitude corresponding to each feature type and the number of target records [ for example: one tenth, one hundredth, etc.;
acquiring a first reception record (conversation record with access) of idle customer service and a second reception record which is in accordance with the association; establishing a first time axis, and expanding the first reception record on the first time axis (the generation time of the record corresponds to the time node of the time axis); the second reception record is also expanded on a second time axis; extracting a first feature, and if the first feature is matched with the preliminary screening feature, it indicates that a malicious event may occur before and after the first position of the first entry corresponding to the first feature [ for example: acquiring more third record items, extracting third features, and if the third features are matched with the fourth features, seating malicious events;
when the customer service is in the standby state, there are switching customer service functions [ for example: visitor initiatively requires to switch customer service, and customer service requires to switch customer service for visitor, and when a customer meets a customer, the attitude is poor, and the customer service is switched, because the customer service belongs to the relationship of colleagues, the last customer service may maliciously inform its bad information [ for example: the client is tricky, you are also poor in attitude, and the like ] and the malicious events occur;
based on the first position, the second position where the third record item is corresponding to the matched third feature is determined; extracting a third position which is farthest from the first position from the second position; determining a fourth position and a fifth position on a second time axis; acquiring a sixth record item based on the fourth position and the fifth position, extracting fifth features, matching the fifth features with the fourth features, determining that the malicious events possibly occur if the fifth features are matched with the fourth features, trying to extract a target record, and if the fifth features are extracted, indicating that the malicious events really occur;
continuously reducing the initial weight, and reducing the initial weight after summarizing a first reduction amplitude generated by the idle customer service, a second reduction amplitude generated by the associated customer service associated with the idle customer service, and a third reduction amplitude generated by a malicious event [ for example: multiplying the initial weight by the down-regulation amplitude, and after down-regulation, finishing acquisition;
the embodiment of the invention sets the association mechanism, the customer service generates the bad record per se, the bad record per se is influenced, the associated customer service associated with the customer service is also influenced, the cost of bad customer service attitude is improved, and the occurrence rate of events with bad customer service attitude is reduced to a certain extent.
The embodiment of the invention provides an online customer service real-time distribution method, which further comprises the following steps:
step S6: summarizing second sessions which cannot be successfully distributed in a second session queue to obtain a second session set, summarizing the second sessions which wait for distribution in the second session set for more than or equal to a preset threshold value to obtain a third session set, establishing an outsourcing customer service queue, and sequentially distributing the second sessions in the third session set to outsourcing customer services in the outsourcing customer service queue;
wherein, establish outsourcing customer service queue, include:
acquiring a preset outsourcing customer service set, and distributing a preset questionnaire to outsourcing customer services in the outsourcing customer service set;
acquiring a survey result input by the outsourcing customer service based on a questionnaire, wherein the survey result comprises: the accuracy rate;
determining the satisfaction degree of outsourcing customer service based on a preset satisfaction degree library;
determining the total number of second sessions in a third session set, and determining tendency proportion corresponding to the total number based on a preset tendency proportion library;
extracting outsourced customer services meeting preset second conditions in the outsourced customer service set, and adding the outsourced customer services into an outsourced customer service queue;
when the outsourced customer service meeting the second condition is added into the outsourced customer service queue, the establishment is completed;
wherein the second condition comprises:
the ratio of the accuracy of outsourcing customer service to the satisfaction degree accords with the trend proportion.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset threshold specifically includes: for example, 8 minutes; a preset outsourcing customer service set: a set of multiple online idle outsourcing combinations; the preset questionnaire specifically comprises: for example, the use scene is a clothing sale online shop, the questionnaire is a clothing familiarity survey question, and the outsourcing is answered, the higher the accuracy rate is, the more familiar the outsourcing is, the easier the outsourcing is to get to the hands; the preset satisfaction degree library specifically comprises: a database, in which satisfaction statistics of each outsourced customer service are stored [ big data can be obtained, and the big data is satisfaction evaluation data given by a management layer after the outsource works in each work place based on big data judgment ]; the preset tendency matching library specifically comprises the following steps: a database, in which trend proportions corresponding to the total number of the second sessions in the third session set are stored, for example: the number of waiting people is more at present, and the tendency ratio is 2: 8, when the number of people is large, the accuracy of the questionnaire can be blocked by a plurality of outsourcing customer services, so that the number of the outsourcing customer services is insufficient, and more people tend to satisfy the degree; the proportion meeting the tendency is specifically as follows: the tendency ratio is 2: 8, the accuracy of a certain outsourcing customer service is 0.2, and the satisfaction is 0.8, namely, when the accuracy is 0.3, and the satisfaction is 0.8, the accuracy is also 0.2, and the satisfaction is 0.9, [ the determined tendency ratio is 2: 8, the most consistent accuracy and satisfaction under the matching are 0.2 and 0.8, the accuracy of the external envelope customer service is more than or equal to 0.2, and the satisfaction is more than or equal to 0.8 and meets the trend matching;
the embodiment of the invention extracts the session with longer waiting time, distributes the session to the outsourcing customer service, sets the trend matching library, and adaptively adjusts the matching based on the number of the waiting sessions, thereby meeting the requirement of not allowing the visitor to wait for a long time to the greatest extent, improving the experience of the visitor and further improving the response capability of the system.
An embodiment of the present invention provides an online customer service real-time distribution system, as shown in fig. 2, including:
an obtaining module 1, configured to obtain a plurality of first session queues;
the first allocation module 2 is used for sequentially allocating the first sessions in the first session queue to the idle customer service;
the summarizing module 3 is used for summarizing the first sessions which are not successfully distributed in the first session queue to obtain a first session set;
the preprocessing module 4 is configured to preprocess the first session set to obtain a plurality of second session queues;
and the second allocating module 5 is configured to sequentially allocate the second sessions in the second session queue to the idle customer service based on the allocation policy.
The working principle and the beneficial effects of the technical scheme are as follows:
setting a plurality of blank queues, and adding visitors into the blank queues to obtain a plurality of first session queues when a large number of visitors initiate session requests; firstly, sequentially allocating first sessions in a first session queue to idle customer service, and then summarizing the first sessions which cannot be served by the customer service to obtain a first session set; preprocessing the first session set (reasonably splitting) to obtain a plurality of second session queues; and sequentially allocating the second sessions in the second session queue to the idle customer service based on the allocation strategy.
According to the embodiment of the invention, when a large number of visitors initiate session requests, the processing can be responded in time, and the customer service distribution is completed.
The embodiment of the invention provides an online customer service real-time distribution system, and a preprocessing module 4 executes the following operations:
equally dividing or approximately equally dividing the first sessions in the first session set into a plurality of third session queues;
and sequentially placing at least one third conversation of the member users in the third conversation queue at the forefront end of the third conversation queue to obtain a second conversation queue.
The working principle and the beneficial effects of the technical scheme are as follows:
evenly divide the first sessions in the first set of sessions [ e.g.: 9 first sessions, equally divided into 3 queues, 3 per queue or approximately equally divided [ for example: and (3) approximately evaluating 10 first conversations and dividing the first conversations into 3 queues, wherein the number of the conversations in the queues is 3, 3 and 4 in turn to form a plurality of third conversation queues, and the member users in the third conversation queues are arranged to the forefront and are preferentially distributed.
The embodiment of the invention provides an online customer service real-time distribution system, and a distribution strategy comprises the following steps: a rotation distribution strategy, a random distribution strategy, a weight distribution strategy and a workload distribution strategy;
wherein, the alternate distribution strategy comprises:
establishing an idle customer service queue, and adding idle customer service to the idle customer service queue;
allocating the second session at the forefront in the second session queue to the idle customer service at the forefront of the idle customer queue;
when the next second session is allocated, allocating the next second session to the next idle customer service of the last allocated idle customer service in the idle customer service queue;
the random allocation strategy comprises the following steps:
based on a preset workload analysis model, carrying out average distribution analysis on the workload of each idle customer service to obtain a target idle customer service, and allocating a second session to be allocated to the target idle customer service;
the weight distribution strategy comprises the following steps:
acquiring the weight of the idle customer service, and allocating a second session to be allocated to the idle customer service corresponding to the maximum value of the weight;
the workload distribution strategy comprises the following steps:
and acquiring the workload of the idle customer service, and allocating the second session to be allocated to the idle customer service corresponding to the minimum workload.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset workload analysis module specifically comprises: a model generated after a large number of records of manual work which are analyzed for carrying out average distribution on the workload of each customer service (the workload of each customer service is almost) is utilized for learning, and after the model is analyzed, target idle customer service needing conversation distribution is output;
and a plurality of customer service distribution strategies are set, so that the response capability of the system can be improved, and the customer service distribution can be better completed.
The embodiment of the invention provides an online customer service real-time distribution system, wherein a second distribution module 5 executes the following operations:
determining a plurality of associated customer services associated with the idle customer service based on a preset associated customer service library;
respectively acquiring a plurality of first reception records of idle customer service and a plurality of second reception records of associated customer service;
respectively establishing a first time axis and a second time axis;
expanding the first reception record on a first time axis to obtain a plurality of first record items;
expanding the second reception record on a second time axis to obtain a plurality of second record items;
performing feature extraction on the first record item to obtain a plurality of first features;
acquiring a preset primary screening feature library, matching the first feature with a second feature in the primary screening feature library, and outputting a first matching item if the first feature is matched with the second feature in the primary screening feature library, wherein the first matching item comprises: matching the first and second matched features;
determining a first position of a first record item corresponding to the first characteristic matched with the first record item on a first time axis;
acquiring first record items with a first quantity in front of a first position and/or in a preset mode on a first time axis, and taking the first record items as third record items;
extracting the features of the third record item to obtain a plurality of third features;
acquiring a preset malicious feature library, matching the third feature with a fourth feature in the malicious feature library, and outputting a second matching item if the matching is consistent, wherein the second matching item comprises: matching the third characteristic and the fourth characteristic which are in line, matching a first characteristic type corresponding to the fourth characteristic which is in line, and matching a first matching degree between the third characteristic and the fourth characteristic which are in line;
determining a second position of a third record item corresponding to the matched third feature on the first time axis;
extracting a second position farthest from the first position in front of or behind the second position as a third position;
determining a fourth position on the second time axis corresponding to the third position, and determining a fifth position on the second time axis corresponding to the first position;
acquiring a second record item with a second quantity preset in the direction of the preset fourth position on a second time axis, and taking the second record item as a fourth record item;
acquiring a second recording item between a fourth position and a fifth position on a second time axis, and taking the second recording item as a fifth recording item;
summarizing the fourth record items and the fifth record items, and performing duplication elimination to obtain a plurality of sixth record items;
performing feature extraction on the sixth record item to obtain a plurality of fifth features;
matching the fifth feature with a fourth feature in the malicious feature library, and outputting a third matching item if the matching is matched, wherein the third matching item comprises: matching the fifth characteristic and the fourth characteristic which are in line, matching a second characteristic type corresponding to the fourth characteristic which is in line, and matching a second matching degree between the fifth characteristic and the fourth characteristic which are in line;
determining a first down-regulation amplitude which is commonly corresponding to the first feature type and the first matching degree and determining a second down-regulation amplitude which is commonly corresponding to the second feature type and the second matching degree based on a preset down-regulation amplitude library;
integrating a third record item corresponding to a third feature matched and matched with the fourth feature and a sixth record item corresponding to a fifth feature matched and matched with the fourth feature to obtain reception record data;
extracting target records which meet a preset first condition in the reception record data, and summarizing the number of the target records;
determining third downward adjustment amplitudes corresponding to the quantity based on the downward adjustment amplitude library;
summarizing the first downward regulation amplitude, the second downward regulation amplitude and the third downward regulation amplitude to obtain a downward regulation amplitude sum;
acquiring a preset initial weight, and performing down-regulation on the initial weight based on the down-regulation amplitude;
the adjusted initial weight is used as the weight of the idle customer service to finish the acquisition;
wherein the directions include: front or rear, away from the fifth position;
the first condition includes:
the same customer is received by different receptionists, and the receptionist includes: idle customer service or associated customer service.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset associated customer service library specifically comprises the following steps: a database, which stores the association relation (the association relation is divided into active association and passive association) with each customer service, the active association means that the customer service sets a certain customer service as association, the passive association means that other customer services set themselves as association; the preset primary screening feature library specifically comprises the following steps: a database having stored therein a plurality of prescreening characteristics [ for example: that is, the characteristics of vital energy; the preset first number specifically comprises: for example, 10; the preset malicious feature library specifically comprises the following steps: a database having stored therein a number of malicious features [ for example: bad speech characteristics in customer service level); the preset second number is specifically as follows: for example, 20; the preset down-regulation amplitude library specifically comprises: a database, in which the down-regulation amplitude corresponding to each feature type and the number of target records [ for example: one tenth, one hundredth, etc.;
acquiring a first reception record (conversation record with access) of idle customer service and a second reception record which is in accordance with the association; establishing a first time axis, and expanding the first reception record on the first time axis (the generation time of the record corresponds to the time node of the time axis); the second reception record is also expanded on a second time axis; extracting a first feature, and if the first feature is matched with the preliminary screening feature, it indicates that a malicious event may occur before and after the first position of the first entry corresponding to the first feature [ for example: acquiring more third record items, extracting third features, and if the third features are matched with the fourth features, seating malicious events;
when the customer service is in the standby state, there are switching customer service functions [ for example: visitor initiatively requires to switch customer service, and customer service requires to switch customer service for visitor, and when a customer meets a customer, the attitude is poor, and the customer service is switched, because the customer service belongs to the relationship of colleagues, the last customer service may maliciously inform its bad information [ for example: the client is tricky, you are also poor in attitude, and the like ] and the malicious events occur;
based on the first position, the second position where the third record item is corresponding to the matched third feature is determined; extracting a third position which is farthest from the first position from the second position; determining a fourth position and a fifth position on a second time axis; acquiring a sixth record item based on the fourth position and the fifth position, extracting fifth features, matching the fifth features with the fourth features, determining that the malicious events possibly occur if the fifth features are matched with the fourth features, trying to extract a target record, and if the fifth features are extracted, indicating that the malicious events really occur;
continuously reducing the initial weight, and reducing the initial weight after summarizing a first reduction amplitude generated by the idle customer service, a second reduction amplitude generated by the associated customer service associated with the idle customer service, and a third reduction amplitude generated by a malicious event [ for example: multiplying the initial weight by the down-regulation amplitude, and after down-regulation, finishing acquisition;
the embodiment of the invention sets the association mechanism, the customer service generates the bad record per se, the bad record per se is influenced, the associated customer service associated with the customer service is also influenced, the cost of bad customer service attitude is improved, and the occurrence rate of events with bad customer service attitude is reduced to a certain extent.
The embodiment of the invention provides an online customer service real-time distribution system, which further comprises:
the third distribution module is used for summarizing the second sessions which are not successfully distributed in the second session queue to obtain a second session set, summarizing the second sessions which are waiting for distribution in the second session set and have the distribution time more than or equal to a preset threshold value to obtain a third session set, establishing an outsourcing customer service queue, and sequentially distributing the second sessions in the third session set to outsourcing customer services in the outsourcing customer service queue;
the third distribution module performs the following operations:
acquiring a preset outsourcing customer service set, and distributing a preset questionnaire to outsourcing customer services in the outsourcing customer service set;
acquiring a survey result input by the outsourcing customer service based on a questionnaire, wherein the survey result comprises: the accuracy rate;
determining the satisfaction degree of outsourcing customer service based on a preset satisfaction degree library;
determining the total number of second sessions in a third session set, and determining tendency proportion corresponding to the total number based on a preset tendency proportion library;
extracting outsourced customer services meeting preset second conditions in the outsourced customer service set, and adding the outsourced customer services into an outsourced customer service queue;
when the outsourced customer service meeting the second condition is added into the outsourced customer service queue, the establishment is completed;
wherein the second condition comprises:
the ratio of the accuracy of outsourcing customer service to the satisfaction degree accords with the trend proportion.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset threshold specifically includes: for example, 8 minutes; a preset outsourcing customer service set: a set of multiple online idle outsourcing combinations; the preset questionnaire specifically comprises: for example, the use scene is a clothing sale online shop, the questionnaire is a clothing familiarity survey question, and the outsourcing is answered, the higher the accuracy rate is, the more familiar the outsourcing is, the easier the outsourcing is to get to the hands; the preset satisfaction degree library specifically comprises: a database, in which satisfaction statistics of each outsourced customer service are stored [ big data can be obtained, and the big data is satisfaction evaluation data given by a management layer after the outsource works in each work place based on big data judgment ]; the preset tendency matching library specifically comprises the following steps: a database, in which trend proportions corresponding to the total number of the second sessions in the third session set are stored, for example: the number of waiting people is more at present, and the tendency ratio is 2: 8, when the number of people is large, the accuracy of the questionnaire can be blocked by a plurality of outsourcing customer services, so that the number of the outsourcing customer services is insufficient, and more people tend to satisfy the degree; the proportion meeting the tendency is specifically as follows: the tendency ratio is 2: 8, the accuracy of a certain outsourcing customer service is 0.2, and the satisfaction is 0.8, namely, when the accuracy is 0.3, and the satisfaction is 0.8, the accuracy is also 0.2, and the satisfaction is 0.9, [ the determined tendency ratio is 2: 8, the most consistent accuracy and satisfaction under the matching are 0.2 and 0.8, the accuracy of the external envelope customer service is more than or equal to 0.2, and the satisfaction is more than or equal to 0.8 and meets the trend matching;
the embodiment of the invention extracts the session with longer waiting time, distributes the session to the outsourcing customer service, sets the trend matching library, and adaptively adjusts the matching based on the number of the waiting sessions, thereby meeting the requirement of not allowing the visitor to wait for a long time to the greatest extent, improving the experience of the visitor and further improving the response capability of the system.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (10)

1. An online customer service real-time distribution method is characterized by comprising the following steps:
step S1: acquiring a plurality of first session queues;
step S2: sequentially distributing the first conversations in the first conversation queue to idle customer services;
step S3: summarizing the first conversation which is not successfully distributed in the first conversation queue to obtain a first conversation set;
step S4: preprocessing the first session set to obtain a plurality of second session queues;
step S5: and sequentially allocating the second sessions in the second session queue to idle customer services based on an allocation strategy.
2. The method for real-time distribution of online customer service according to claim 1, wherein the step S4: preprocessing the first session set to obtain a plurality of second session queues, including:
evenly dividing or approximately equally dividing the first session in the first set of sessions into a plurality of third session queues;
and sequentially placing at least one third conversation of the member users in the third conversation queue at the forefront end of the third conversation queue to obtain a second conversation queue.
3. The method of claim 1, wherein the allocation policy comprises: a rotation distribution strategy, a random distribution strategy, a weight distribution strategy and a workload distribution strategy;
wherein, the alternate allocation strategy comprises:
establishing an idle customer service queue, and adding idle customer service to the idle customer service queue;
assigning the second session at the head of the second session queue to the idle customer service at the head of the idle customer queue;
when allocating the next second session, allocating the next second session to the next idle service in the idle service queue, which is allocated to the idle service last time;
the random allocation strategy comprises:
based on a preset workload analysis model, carrying out average distribution analysis on the workload of each idle customer service to obtain target idle customer services, and allocating the second session to be allocated to the target idle customer services;
the weight distribution strategy comprises the following steps:
acquiring the weight of the idle customer service, and allocating the second session to be allocated to the idle customer service corresponding to the maximum value of the weight;
the workload distribution strategy comprises:
and acquiring the workload of the idle customer service, and distributing the second session to be distributed to the idle customer service corresponding to the minimum workload.
4. The method of claim 3, wherein obtaining the weight of the idle customer service comprises:
determining a plurality of associated customer services associated with the idle customer service based on a preset associated customer service library;
respectively acquiring a plurality of first reception records of the idle customer service and a plurality of second reception records of the associated customer service;
respectively establishing a first time axis and a second time axis;
expanding the first reception record on the first time axis to obtain a plurality of first record items;
expanding the second reception record on the second time axis to obtain a plurality of second record items;
performing feature extraction on the first record item to obtain a plurality of first features;
acquiring a preset primary screening feature library, matching the first features with the second features in the primary screening feature library, and outputting a first matching item if the first features are matched with the second features in the primary screening feature library, wherein the first matching item comprises: matching the first feature and the second feature that are coincident;
determining that the first feature that matches corresponds to a first location of the first entry on the first time axis;
acquiring first record items of a first quantity before and/or in a preset first position on the first time axis, and taking the first record items as third record items;
performing feature extraction on the third record item to obtain a plurality of third features;
acquiring a preset malicious feature library, matching the third feature with a fourth feature in the malicious feature library, and outputting a second matching item if the third feature matches with the fourth feature in the malicious feature library, wherein the second matching item comprises: matching the third feature and the fourth feature which are in line, matching a first feature type corresponding to the fourth feature which is in line, and matching a first matching degree between the third feature and the fourth feature which are in line;
determining that the third feature that matches corresponds to a second location of the third entry on the first time axis;
extracting the second position farthest in front or back from the first position among the second positions as a third position;
determining a fourth position on the second time axis corresponding to the third position, determining a fifth position on the second time axis corresponding to the first position;
acquiring a second number of second recording items preset in the direction preset by the fourth position on the second time axis, and taking the second recording items as fourth recording items;
acquiring the second recording item between the fourth position and the fifth position on the second time axis, and taking the second recording item as a fifth recording item;
summarizing the fourth record items and the fifth record items, and performing duplication elimination processing to obtain a plurality of sixth record items;
performing feature extraction on the sixth record item to obtain a plurality of fifth features;
matching the fifth feature with the fourth feature in the malicious feature library, and outputting a third matching item if the fifth feature matches the fourth feature in the malicious feature library, wherein the third matching item comprises: matching the fifth feature and the fourth feature which are in line, matching a second feature type corresponding to the fourth feature which is in line, and matching a second matching degree between the fifth feature and the fourth feature which are in line;
determining a first downward adjustment amplitude which is commonly corresponding to the first feature type and the first matching degree based on a preset downward adjustment amplitude library, and determining a second downward adjustment amplitude which is commonly corresponding to the second feature type and the second matching degree;
integrating a third characteristic matched with the fourth characteristic and corresponding to the third record item and a fifth characteristic matched with the fourth characteristic and corresponding to the sixth record item to obtain reception record data;
extracting target records which meet a preset first condition in reception record data, and summarizing the number of the target records;
determining a third down-regulation amplitude corresponding to the quantity based on the down-regulation amplitude library;
summarizing the first downward regulation amplitude, the second downward regulation amplitude and the third downward regulation amplitude to obtain a downward regulation amplitude sum;
acquiring a preset initial weight, and performing down-regulation on the initial weight based on the down-regulation amplitude;
taking the adjusted initial weight as the weight of the idle customer service to finish the acquisition;
wherein the direction comprises: front or rear, away from the fifth position;
the first condition includes:
the same customer is received by different receptionists, the receptionist includes: the idle customer service or the associated customer service.
5. The method of claim 1, further comprising:
step S6: summarizing the second sessions which are not successfully distributed in the second session queue to obtain a second session set, summarizing the second sessions which are waiting for distribution in the second session set and have the time more than or equal to a preset threshold value to obtain a third session set, establishing an outsourcing customer service queue, and sequentially distributing the second sessions in the third session set to outsourcing customer services in the outsourcing customer service queue;
wherein, establish outsourcing customer service queue, include:
acquiring a preset outsourcing customer service set, and distributing a preset questionnaire to outsourcing customer services in the outsourcing customer service set;
obtaining a survey result input by the outsourcing customer service based on the questionnaire, wherein the survey result comprises: the accuracy rate;
determining the satisfaction degree of the outsourcing customer service based on a preset satisfaction degree library;
determining the total number of the second sessions in the third session set, and determining tendency proportion corresponding to the total number based on a preset tendency proportion library;
extracting outsourced customer service meeting a preset second condition in the outsourced customer service set, and adding the outsourced customer service into an outsourced customer service queue;
when the outsourcing customer service meeting the second condition is added into an outsourcing customer service queue, finishing the establishment;
wherein the second condition comprises:
the ratio of the accuracy of the outsourcing customer service to the satisfaction degree accords with the tendency matching.
6. An online customer service real-time distribution system, comprising:
the acquisition module is used for acquiring a plurality of first session queues;
the first allocation module is used for sequentially allocating the first sessions in the first session queue to idle customer services;
the summarizing module is used for summarizing the first session which is not successfully distributed in the first session queue to obtain a first session set;
the preprocessing module is used for preprocessing the first session set to obtain a plurality of second session queues;
and the second distribution module is used for sequentially distributing the second sessions in the second session queue to the idle customer service based on a distribution strategy.
7. The system of claim 6, wherein the preprocessing module performs the following operations:
evenly dividing or approximately equally dividing the first session in the first set of sessions into a plurality of third session queues;
and sequentially placing at least one third conversation of the member users in the third conversation queue at the forefront end of the third conversation queue to obtain a second conversation queue.
8. The system of claim 6, wherein the allocation policy comprises: a rotation distribution strategy, a random distribution strategy, a weight distribution strategy and a workload distribution strategy;
wherein, the alternate allocation strategy comprises:
establishing an idle customer service queue, and adding idle customer service to the idle customer service queue;
assigning the second session at the head of the second session queue to the idle customer service at the head of the idle customer queue;
when allocating the next second session, allocating the next second session to the next idle service in the idle service queue, which is allocated to the idle service last time;
the random allocation strategy comprises:
based on a preset workload analysis model, carrying out average distribution analysis on the workload of each idle customer service to obtain target idle customer services, and allocating the second session to be allocated to the target idle customer services;
the weight distribution strategy comprises the following steps:
acquiring the weight of the idle customer service, and allocating the second session to be allocated to the idle customer service corresponding to the maximum value of the weight;
the workload distribution strategy comprises:
and acquiring the workload of the idle customer service, and distributing the second session to be distributed to the idle customer service corresponding to the minimum workload.
9. The system of claim 8, wherein the second distribution module performs the following operations:
determining a plurality of associated customer services associated with the idle customer service based on a preset associated customer service library;
respectively acquiring a plurality of first reception records of the idle customer service and a plurality of second reception records of the associated customer service;
respectively establishing a first time axis and a second time axis;
expanding the first reception record on the first time axis to obtain a plurality of first record items;
expanding the second reception record on the second time axis to obtain a plurality of second record items;
performing feature extraction on the first record item to obtain a plurality of first features;
acquiring a preset primary screening feature library, matching the first features with the second features in the primary screening feature library, and outputting a first matching item if the first features are matched with the second features in the primary screening feature library, wherein the first matching item comprises: matching the first feature and the second feature that are coincident;
determining that the first feature that matches corresponds to a first location of the first entry on the first time axis;
acquiring first record items of a first quantity before and/or in a preset first position on the first time axis, and taking the first record items as third record items;
performing feature extraction on the third record item to obtain a plurality of third features;
acquiring a preset malicious feature library, matching the third feature with a fourth feature in the malicious feature library, and outputting a second matching item if the third feature matches with the fourth feature in the malicious feature library, wherein the second matching item comprises: matching the third feature and the fourth feature which are in line, matching a first feature type corresponding to the fourth feature which is in line, and matching a first matching degree between the third feature and the fourth feature which are in line;
determining that the third feature that matches corresponds to a second location of the third entry on the first time axis;
extracting the second position farthest in front or back from the first position among the second positions as a third position;
determining a fourth position on the second time axis corresponding to the third position, determining a fifth position on the second time axis corresponding to the first position;
acquiring a second number of second recording items preset in the direction preset by the fourth position on the second time axis, and taking the second recording items as fourth recording items;
acquiring the second recording item between the fourth position and the fifth position on the second time axis, and taking the second recording item as a fifth recording item;
summarizing the fourth record items and the fifth record items, and performing duplication elimination processing to obtain a plurality of sixth record items;
performing feature extraction on the sixth record item to obtain a plurality of fifth features;
matching the fifth feature with the fourth feature in the malicious feature library, and outputting a third matching item if the fifth feature matches the fourth feature in the malicious feature library, wherein the third matching item comprises: matching the fifth feature and the fourth feature which are in line, matching a second feature type corresponding to the fourth feature which is in line, and matching a second matching degree between the fifth feature and the fourth feature which are in line;
determining a first downward adjustment amplitude which is commonly corresponding to the first feature type and the first matching degree based on a preset downward adjustment amplitude library, and determining a second downward adjustment amplitude which is commonly corresponding to the second feature type and the second matching degree;
integrating a third characteristic matched with the fourth characteristic and corresponding to the third record item and a fifth characteristic matched with the fourth characteristic and corresponding to the sixth record item to obtain reception record data;
extracting target records which meet a preset first condition in reception record data, and summarizing the number of the target records;
determining a third down-regulation amplitude corresponding to the quantity based on the down-regulation amplitude library;
summarizing the first downward regulation amplitude, the second downward regulation amplitude and the third downward regulation amplitude to obtain a downward regulation amplitude sum;
acquiring a preset initial weight, and performing down-regulation on the initial weight based on the down-regulation amplitude;
taking the adjusted initial weight as the weight of the idle customer service to finish the acquisition;
wherein the direction comprises: front or rear, away from the fifth position;
the first condition includes:
the same customer is received by different receptionists, the receptionist includes: the idle customer service or the associated customer service.
10. The on-line real-time customer service distribution system of claim 6, further comprising:
the third distribution module is used for summarizing the second sessions which are not successfully distributed in the second session queue to obtain a second session set, summarizing the second sessions which are waiting for distribution in the second session set and have the time more than or equal to a preset threshold value to obtain a third session set, establishing an outsourcing service queue, and sequentially distributing the second sessions in the third session set to outsourcing service in the outsourcing service queue;
the third distribution module performs the following operations:
acquiring a preset outsourcing customer service set, and distributing a preset questionnaire to outsourcing customer services in the outsourcing customer service set;
obtaining a survey result input by the outsourcing customer service based on the questionnaire, wherein the survey result comprises: the accuracy rate;
determining the satisfaction degree of the outsourcing customer service based on a preset satisfaction degree library;
determining the total number of the second sessions in the third session set, and determining tendency proportion corresponding to the total number based on a preset tendency proportion library;
extracting outsourced customer service meeting a preset second condition in the outsourced customer service set, and adding the outsourced customer service into an outsourced customer service queue;
when the outsourcing customer service meeting the second condition is added into an outsourcing customer service queue, finishing the establishment;
wherein the second condition comprises:
the ratio of the accuracy of the outsourcing customer service to the satisfaction degree accords with the tendency matching.
CN202110761668.8A 2021-07-06 2021-07-06 Online customer service real-time distribution method and system Active CN113360627B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110761668.8A CN113360627B (en) 2021-07-06 2021-07-06 Online customer service real-time distribution method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110761668.8A CN113360627B (en) 2021-07-06 2021-07-06 Online customer service real-time distribution method and system

Publications (2)

Publication Number Publication Date
CN113360627A true CN113360627A (en) 2021-09-07
CN113360627B CN113360627B (en) 2021-12-21

Family

ID=77538348

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110761668.8A Active CN113360627B (en) 2021-07-06 2021-07-06 Online customer service real-time distribution method and system

Country Status (1)

Country Link
CN (1) CN113360627B (en)

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150170259A1 (en) * 2010-09-03 2015-06-18 Edward Katzin System and method for custom service markets
CN105898087A (en) * 2016-06-20 2016-08-24 上海久科信息技术有限公司 Full-routing customer service distribution method
CN106096975A (en) * 2016-06-03 2016-11-09 北京壹代家庭服务有限公司 A kind of online customer service system and method
CN106453977A (en) * 2016-09-14 2017-02-22 广东亿迅科技有限公司 Comprehensive queuing method for multimedia comprehensive customer service representatives
CN107332902A (en) * 2017-06-29 2017-11-07 北京鸿联九五信息产业有限公司 User's request distribution method, device and the computing device of online customer service system
CN107426448A (en) * 2016-05-23 2017-12-01 上海神计信息系统工程有限公司 A kind of automatic calling allocation method being lined up based on dynamic priority
CN107968897A (en) * 2017-11-03 2018-04-27 平安科技（深圳）有限公司 Customer service session distribution method, electronic device and computer-readable recording medium
CN109670691A (en) * 2018-12-07 2019-04-23 泰康保险集团股份有限公司 Method, equipment and the customer service system distributed for customer service queue management and customer service
CN110648060A (en) * 2019-09-19 2020-01-03 达疆网络科技(上海)有限公司 Method for automatically allocating tasks to customer service
CN111031113A (en) * 2019-11-29 2020-04-17 湖南御家科技有限公司 User queuing method for supporting platform-level customer service system
CN111163238A (en) * 2019-11-26 2020-05-15 国网江苏省电力有限公司电力科学研究院 Intelligent scheduling method of online customer service system
CN111726462A (en) * 2019-03-22 2020-09-29 阿里巴巴集团控股有限公司 Resource scheduling system, method and device in customer service system and electronic equipment

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150170259A1 (en) * 2010-09-03 2015-06-18 Edward Katzin System and method for custom service markets
CN107426448A (en) * 2016-05-23 2017-12-01 上海神计信息系统工程有限公司 A kind of automatic calling allocation method being lined up based on dynamic priority
CN106096975A (en) * 2016-06-03 2016-11-09 北京壹代家庭服务有限公司 A kind of online customer service system and method
CN105898087A (en) * 2016-06-20 2016-08-24 上海久科信息技术有限公司 Full-routing customer service distribution method
CN106453977A (en) * 2016-09-14 2017-02-22 广东亿迅科技有限公司 Comprehensive queuing method for multimedia comprehensive customer service representatives
CN107332902A (en) * 2017-06-29 2017-11-07 北京鸿联九五信息产业有限公司 User's request distribution method, device and the computing device of online customer service system
CN107968897A (en) * 2017-11-03 2018-04-27 平安科技（深圳）有限公司 Customer service session distribution method, electronic device and computer-readable recording medium
CN109670691A (en) * 2018-12-07 2019-04-23 泰康保险集团股份有限公司 Method, equipment and the customer service system distributed for customer service queue management and customer service
CN111726462A (en) * 2019-03-22 2020-09-29 阿里巴巴集团控股有限公司 Resource scheduling system, method and device in customer service system and electronic equipment
CN110648060A (en) * 2019-09-19 2020-01-03 达疆网络科技(上海)有限公司 Method for automatically allocating tasks to customer service
CN111163238A (en) * 2019-11-26 2020-05-15 国网江苏省电力有限公司电力科学研究院 Intelligent scheduling method of online customer service system
CN111031113A (en) * 2019-11-29 2020-04-17 湖南御家科技有限公司 User queuing method for supporting platform-level customer service system

Also Published As

Publication number Publication date
CN113360627B (en) 2021-12-21

Similar Documents

Publication Publication Date Title
AU2017404110B2 (en) Agent allocation method and device, server, and storage medium
CN103684874B (en) Method and device for automatically distributing online customer service executives to conduct customer service
CN111079557B (en) Face recognition-based automatic distribution method and system for power business hall customer service terminals
US20060015388A1 (en) Method and apparatus for supporting individualized selection rules for resource allocation
Aksin et al. To sell or not to sell: Determining the trade-offs between service and sales in retail banking phone centers
US20100254527A1 (en) System and method for matching service representatives with customers
Al-Jumaily et al. Automatic queuing model for banking applications
CN113904912B (en) Method and device for realizing high availability of service of cloud management platform
CN112418799A (en) Work order assignment method and device, electronic equipment and readable storage medium
CN107231393A (en) A kind of conversation processing method and device
CN110570215A (en) Intelligent customer service system
CN111985786A (en) Agent-based task allocation method and device, computer equipment and storage medium
CN107888660A (en) Cloud service resource allocation method, medium, device and computing device
CN107786758A (en) Distribution method of attending a banquet and device
CN114462849A (en) Task balanced distribution method and task balanced distribution device
CN113360627B (en) Online customer service real-time distribution method and system
CN109583615B (en) Conference room booking method, conference room booking system, conference room booking server and computer readable storage medium
CN110458410A (en) A kind of method and apparatus of task distribution
CN110233840A (en) A kind of method for processing business, device, equipment and storage medium
Billiet et al. Understanding and improving response rates
Whitt A multi-class fluid model for a contact center with skill-based routing
Ackere et al. Simulation and queueing theory applied to a single-server queue with advertising and balking
CN115394003A (en) Service handling time length estimation method, device, equipment, storage medium and product
CN111163237B (en) Call service flow control method and related device
CN106776947A (en) Resource acquiring method, device and terminal

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant