CN110414823A - Service quality optimization management system based on community system - Google Patents
Service quality optimization management system based on community system Download PDFInfo
- Publication number
- CN110414823A CN110414823A CN201910664645.8A CN201910664645A CN110414823A CN 110414823 A CN110414823 A CN 110414823A CN 201910664645 A CN201910664645 A CN 201910664645A CN 110414823 A CN110414823 A CN 110414823A
- Authority
- CN
- China
- Prior art keywords
- emotion
- employee
- user
- service
- community
- 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
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 22
- 230000008451 emotion Effects 0.000 claims abstract description 179
- 238000004891 communication Methods 0.000 claims abstract description 30
- 238000004364 calculation method Methods 0.000 claims abstract description 23
- 238000004458 analytical method Methods 0.000 claims abstract description 6
- 238000003860 storage Methods 0.000 claims description 16
- 230000005540 biological transmission Effects 0.000 claims description 12
- 238000000034 method Methods 0.000 claims description 12
- 238000009826 distribution Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 230000008909 emotion recognition Effects 0.000 description 4
- 230000008520 organization Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- 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
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Finance (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Accounting & Taxation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The present invention provides a kind of to be analyzed based on conversation message of the community system to employee and user, to be managed by service quality of these conversation messages to employee, further also optimize the service quality optimization management system of the Service Quality Management by sentiment analysis, it is characterised by comprising: service managing apparatus；At least one employee's terminal is provided with community module corresponding with community system and employee's account for access；Multiple user terminals are provided with community module and the user account for access；One administrator terminal, it is provided with community module and administrator's account for access, service managing apparatus logins community by administrator's account and periodically obtains the conversation message in community, wherein, there is service managing apparatus dialog information acquisition unit, dialogue identification part, emotion calculation part, session affection computation portion, service quality to be divided to determine that generating unit, market content determine generating unit, prompt information generating unit and service side communication unit.
Description
Technical field
The invention belongs to field of client relation management, and in particular to a kind of service quality optimization management based on community system
System.
Background technique
With the development of social media, there are some enterprises that can realize by establishing online community and interacted with consumer, from
And reinforce its competitiveness in the market, such as the service provider of infant's product, employee will be allowed by community to consumer
Some the problem of should be noted that when bringing up about infant and the relevant product of some infants of push are provided.Statistics indicate that
The support of consumer can be improved 25% by the complaint in social media only by simple reply consumer by enterprise,
Therefore this mode, which not only facilitates, improves consumer to the satisfaction of enterprise, additionally aids the popularity and its production for promoting enterprise
Product.Even be conducive to consumer and understand daily the problem of not paying attention to or obtaining without channel in time.
However, the level of skill of employee plays the role of very important during linking up with consumer.But by
There is no direct data in the management of service, enterprise is difficult to carry out the service quality of each employee the judgement of one quantization,
It just can not preferably be managed and decision.
Summary of the invention
To solve the above problems, provide it is a kind of analyzed based on conversation message of the community system to employee and user, from
And be managed by service quality of these conversation messages to employee, the service quality is also further optimized by sentiment analysis
The service quality optimization management system of management, present invention employs following technical solutions:
The present invention provides a kind of service quality optimization management systems based on community system characterized by comprising clothes
Business managing device；At least one employee's terminal, is provided with community module corresponding with community system, and each employee's terminal has one
A employee's account for being used to access community module, each employee's account are corresponding with a community, and employee's account is for making employee logical
Employee's account is crossed to login community and provide a user question and answer marketing service information；Multiple user terminals are provided with community module,
With the user account for accessing community module, for allowing user to login community by user account and obtaining employee's offer
Question and answer marketing service information；One administrator terminal, is provided with community module, has the management for accessing all community modules
Member's account, service managing apparatus login community by administrator's account and periodically obtain the dialog text in community, wherein service
There is managing device dialog information acquisition unit, dialogue identification part, emotion point calculation part, session affection computation portion, service quality to sentence
Determine generating unit, market content determines generating unit, prompt information generating unit and service side communication unit, dialog information acquisition unit are regular
The conversation message stream for accessing community module by administrator's account and obtaining corresponding each community and being made of a plurality of conversation message,
Dialogue identification part engages in the dialogue identification to obtain multiple related with content type to each conversation message in conversation message stream
And include the session of multiple respective dialog message, session is divided into service conversation and marketing session, emotion divide calculation part successively right
Each conversation message carries out the message emotion point that emotion point calculates and obtains corresponding to each conversation message, session affection computation portion root
Corresponding each service conversation is calculated according to the conversation message and corresponding message emotion for corresponding to user account in each session point
Service conversation emotion point and the marketing session emotion point of corresponding each marketing session, service quality determine generating unit according to each
The average value of all service conversation emotions of employee's account point determines the service quality of each employee to generate corresponding service
Quality judging information, market content determine generating unit according to each marketing session emotion point determine each marketing session whether not
It is welcome to determine that information, prompt information generating unit determine that information generates phase according to service quality to generate corresponding market content
The service quality prompt information answered, and determine that information generates corresponding market content and determines prompt information, clothes according to market content
Service quality prompt information and market content are determined that prompt information is sent to accordingly according to employee's account by business side communication unit
Employee's terminal.
Service quality administrative system provided by the invention based on community system, can also have such technical characteristic,
Wherein, service managing apparatus also has user feeling Department of Statistics, user's position determination unit, negative user storage unit, negative user
Search part and emotion divide re-computation portion, and user feeling Department of Statistics counts each user account, and corresponding dialogue disappears whithin a period of time
The user feeling point that the message emotion of breath is allocated as each user, user's position determination unit determine that user sentences according to user feeling point
Whether fixed is negative user, and negative user storage unit is deposited the user account for being determined as negative user as negative account
Storage, after emotion divides calculation part to calculate message emotion point, negative user search part retrieval negative user storage unit is to be born
Face account, emotion point re-computation portion count the corresponding message emotion point of negative account according to a preset distribution weight again
Calculation obtains new message emotion point, and the value range for distributing weight is 0.1~0.9.
Service quality administrative system provided by the invention based on community system, can also have such technical characteristic,
Wherein, service managing apparatus also has employee's affection computation portion, and employee's affection computation portion is according to each all transmissions of employee's account
The mean value of message emotion point of conversation message successively calculate employee's emotion that be averaged of each employee's account and score, service quality is sentenced
Generating unit is determined in the service quality for determining each employee according to the average value of all service conversation emotions of each employee's account point
When, determine the service quality of each employee and generate corresponding service quality to determine information also according to employee's emotion scoring that is averaged.
Service quality administrative system provided by the invention based on community system, can also have such technical characteristic,
Wherein, service quality determines information for service quality scoring, and service quality determines generating unit in the institute for determining each employee's account
When the emotion that the average value and corresponding employee for having service conversation emotion to divide is averaged scores, is generated and serviced according to the average value of the two
Quality score.
Service quality administrative system provided by the invention based on community system, can also have such technical characteristic,
Wherein, conversation message includes at least dialog text, and emotion point calculation part includes: text participle unit, disappears respectively to the dialogue of each item
The dialog text of breath is segmented；Participle emotion divides computing unit, carries out emotion to each participle according to the sentiment dictionary prestored
Analysis is to obtain the emotion point of each participle；Message emotion divides setup unit, according to participles all in each conversation message
Emotion point is divided with the message emotion for setting the conversation message.
Service quality administrative system provided by the invention based on community system, can also have such technical characteristic,
Wherein, sentiment dictionary includes text emotion dictionary and expression emotion dictionary.
Service quality administrative system provided by the invention based on community system, can also have such technical characteristic,
Wherein, session affection computation portion is calculating service conversation emotion point and marketing session emotion timesharing, the total session of each session
Emotion divides Cust_SentiitAre as follows:
In formula, C indicates that the sum of each session, J indicate the number of user, Cust_SentiijtcFor primary member in the t weeks
The message emotion of the user point in the dialog procedure c of work i and user j, wherein i represents an employee, and j indicates a consumer,
T is represented one week, and c represents some session in one week.
Invention action and effect
Service quality optimization management system according to the present invention based on community system, due to having dialogue identification part, because
This can carry out identification to the conversation message stream that dialog information acquisition unit is obtained from community to obtain service conversation and battalion
Sell session.Also calculation part is divided to carry out emotion recognition, therefore session feelings to each conversation message in conversation message stream with emotion
Sense calculation part can further calculate the session emotion point of each session according to the message emotion of these conversation messages point, make to service
Quality judging generating unit determines the service quality of employee according to the session emotion of service conversation point, market content is made to determine generating unit
It is not popular with users point to determine whether there is market content according to the session emotion of marketing session, by by these data-pushings to member
Work, it will be able to make employee recognize the service effectiveness of itself and the preference of user, to pointedly improve the service of itself
Quality.The present invention excavates the user feeling in conversation message by arrangement and emotion recognition to community message, thus right
Employee by community service user when service quality play the role of monitoring, be more advantageous to enterprise and employee according to
To information service quality is improved, the professional ability of the final satisfaction for improving consumer and employee.
Detailed description of the invention
Fig. 1 is the structural block diagram of service quality optimization management system in the embodiment of the present invention；
Fig. 2 is the structural block diagram of service managing apparatus in the embodiment of the present invention；
Fig. 3 is the structural block diagram of employee's terminal in the embodiment of the present invention；
Fig. 4 is the structural block diagram of user terminal in the embodiment of the present invention；
Fig. 5 is the structural block diagram of administrator terminal in the embodiment of the present invention；And
Fig. 6 is the flow chart of service quality optimization management process in the embodiment of the present invention.
Specific embodiment
In order to be easy to understand the technical means, the creative features, the aims and the efficiencies achieved by the present invention, tie below
Examples and drawings are closed to be specifically addressed the service quality optimization management system of the invention based on community system.
<embodiment>
Fig. 1 is the structural block diagram of service quality optimization management system in the embodiment of the present invention.
As shown in Figure 1, service quality optimization management system 100 include service managing apparatus 1, it is multiple employee's terminals 2, multiple
User terminal 3 and administrator terminal 4.
Wherein, service managing apparatus 1 is by needing the organization for carrying out service management to be held, employee's terminal 2 respectively by
Employee with different employee's accounts is held, and user terminal 3 is held by the user with different user account, and administrator is whole
Held by the administrator of organization at end 4.
In the present embodiment, the Intelligent mobile equipment (such as smart phone) that user terminal 3 is all held for corresponding personnel, management
Member's terminal 4 and service managing apparatus 1 are respectively the operating terminal (such as computer) for being in communication with each other connection.In other embodiments
In, each component parts of administrator terminal 4 and service managing apparatus 1 can also be the server held as organization
In a part.
In the present embodiment, it is fitted with community module on employee's terminal 2, user terminal 3 and administrator terminal 4, the society
Group's module is and community system (such as wechat, QQ etc. provide system for exchanging of community, are provided with the community communication server 200)
Corresponding software application, each terminal can be carried out with the community communication server 200 by respective community module respectively respectively
It communicates (such as in a manner of through communication network access community communication server 200).Each terminal using community module, pass through account
Number logining community, (dialogue that i.e. employee provides question and answer marketing service disappears to send and receive question and answer marketing service information mutually
Breath).
In the present embodiment, each employee's terminal, user terminal and administrator terminal are respectively provided with respective employee's account
Number, user account and administrator's account.Each employee's account and community correspond, and employee logins society by employee's account
Group and user's transmission question and answer marketing service information into the community；Each community is corresponding with multiple user accounts, Yong Hutong
It crosses the user account and accesses corresponding community and employee's transmission enquirement message or acquisition question and answer marketing service into the community
Information；Administrator's account only has one, can be used in logining all community and receives conversation message therein.In this community
Structure in, user (or consumer) receives service message usually as member and is putd question in community, and employee is usual
Group master as community is responsible for daily community management and question and answer service is exchanged and provided with user, and administrator's account is generally used
In the permission for allowing service managing apparatus 1 to obtain each community of access.
Fig. 2 is the structural block diagram of service managing apparatus in the embodiment of the present invention.
As shown in Fig. 2, there is service managing apparatus 1 dialog information acquisition unit 11, dialogue identification part 12, emotion to divide calculation part
13, user feeling Department of Statistics 14, user's position determination unit 15, negative user storage unit 16, negative user search part 17, emotion point
Re-computation portion 18, session affection computation portion 19, employee's affection computation portion 110, service quality determine generating unit 111, market content
Determine generating unit 112, prompt information generating unit 113, service side communication unit 114 and service side control unit 115.
Wherein, service side communication unit 114 is used to carry out between each composition part of service managing apparatus 1 and service is managed
The data communication between device 1 and other terminals is managed, service side control unit 115 contains for each of service managing apparatus 1
The computer program that composition part work is controlled.
Pair that dialog information acquisition unit 11 accesses community module and obtain corresponding community for periodically passing through administrator's account
Talk about message flow.
In the present embodiment, dialog information acquisition unit 11 can be obtained by service side communication unit 114 to the transmission of administrator terminal 4
Request is taken, to allow administrator terminal 4 to pass through in each community of administrator's account automatic sign in and obtain employee in each community
The conversation message generated is exchanged with user.
In the present embodiment, between the value of the regular set that dialog information acquisition unit 11 periodically obtains in the present embodiment is
Obtained every one week, in other embodiments the regular set can also be set according to actual conditions for every two days, interval January
Deng.Dialog information acquisition unit 11 has a time judging unit, for judging that current time is apart from the time that last time obtains
It is no to reach regular set.
In the present embodiment, conversation message stream is by dialogue all in (the present embodiment is one week) community in a period of time
Message composition, timestamp when every conversation message is sent by dialog text, the dialog text constitutes and the dialog text
Send account (employee's account, user account) etc..
Dialogue identification part 12 is used to engage in the dialogue to each dialog text in conversation message stream identification to obtain multiple
Session related with content type.
In the present embodiment, session is divided into session of taking the initiative in offering a hand, request service conversation and marketing meeting according to content type
Words.Wherein, session of taking the initiative in offering a hand be employee's active transmission service message after, employee user exchange generate session, request service meeting
Words be user send put question to message after, employee user exchange generate session, marketing session be employee send marketing message after, member
Work user exchanges the session generated.
In the present embodiment, dialogue identification part 12 identifies beginning (the i.e. clothes of employee's active transmission of above-mentioned three classes session first
The marketing message for the message, the enquirement message that user sends and employee's transmission of being engaged in, can pass through conventional classifier, such as simplicity
Bayes classifier Classification and Identification obtains), and context text cluster is carried out to dialog text based on time window, to obtain
Multiple sessions.Specifically, parameter and calculating formula involved in the above-mentioned process are as follows:
A), a context dialog text mjTime window W (mj) be；
In formula (1), tiFor dialog text miTimestamp, tjFor dialog text mjTimestamp, the value of D is 6 or 12 small
When.The dialog text m of all contextsjIt is in time window W (mj) in.
B), each dialog text m includes that there are three types of contexts, respectively author's context CA(m), context C is referred toC
(m) and time context CT(m)。
Author's context CA(m) belong to all dialog texts of the same author a by dialog text m and dialog text m
It constitutes:
In dialog text mjBelong to author's context CA(m) when, dialog text miWith dialog text mjBelong to the same session
Probability PA(dij) are as follows:
In formula (3), N indicates the probability density function of normal distribution, μaThe mean value of expression normpdf,
It is set as 0, σaIndicate the standard deviation for belonging to the normpdf of author's context, value can be from training set
Estimation.
Refer to context CC(m) by the every other dialog text of the dialog text m author a mentioned (for example, containing "@D
The dialog text of user ") and mention dialog text m author a every other dialog text constitute:
In formula (4), MaIndicate all dialog texts of author a.
In dialog text mjBelong to and refers to context CC(m) when, dialog text miWith dialog text mjBelong to the same session
Probability PC(dij) are as follows:
In formula (5), μcIt indicates the mean value of normpdf, be set as 0, σcExpression belongs to reference context
Normpdf standard deviation.
Time context CT(m) it is made of the every other sentence other than dialog text m:
CT(m)=M m (6)
In formula (6), M indicates all dialog texts.
In dialog text mjBelong to time context CT(m), dialog text miWith dialog text mjBelong to the same session
Probability PT(dij) are as follows:
In formula (7), μTIt indicates the mean value of normpdf, be set as 0, σTExpression belongs to time context
Normpdf standard deviation.
C), the wordembedding term vector of dialog text m is expressed as v (m), uses dialog text miAbout three kinds
Text extension dialog text mi, then dialog text miExtension indicate mi ′Are as follows:
In formula (8), the meaning of parameters are as follows:
PA(dij) --- dialog text mjBelong to author's context CA(mi) when dialog text miWith dialog text mjBelong to same
The probability of one session；
PC(dij) --- dialog text mjBelong to and refers to context CC(mi) when dialog text miWith dialog text mjBelong to same
The probability of one session；
PT(dij) --- dialog text mjBelong to time context CT(mi) when dialog text miWith dialog text mjBelong to same
The probability of one session；
α --- the content of dialog text m weight shared in extension indicates；
λA--- the relative weighting of author's context；
λC--- refer to the relative weighting of context；
λT--- the relative weighting of time context, λA+λC+λT=1.
By above-mentioned formula (2) to (8), the expression of a sentence can be extended using three kinds of contexts, thus
Be conducive to dialogue identification.The effect of author's context belongs to the same dialogue in the message that adjacent time is delivered for the same author
A possibility that it is higher；Referring to context is to say ,@someone in message, then by the message of the people of@belong to same dialogue can
Energy property is higher；Time context refers to that a possibility that time upper adjacent message belongs to the same dialogue is higher.
D), text cluster further, is carried out to dialog text m, the beginning label by three classes dialogue is clustering prototype T, and
Calculate dialog text miWith dialog text mjText degree of approximation sim (mi, mj):
For remaining other dialog texts mj, calculate the similarity sim (m with clustering prototypej, T):
In formula, dialog text mjAffiliated cluster beIfThen dialog text mjAffiliated cluster beIfThen mjIt is not belonging to the dialogue to be identified herein, tthreshFor mjBelong toMinimum similarity degree threshold
Value.
The dialog text m that corresponding each clustering prototype T can be obtained by formula (4) and (5), successively according to each cluster
Affiliated dialog text m is combined to form each session by prototype T.
By above-mentioned formula (1) to (11), conversation message in each conversation message stream can be divided by talking with identification part 12
For multiple sessions.
The identification example of session is as shown in the table:
1 session of table identifies example
In table 1, label is conversation tag, and the conversation message of same label belongs to the same session, the time, sender, interior
Hold and respectively corresponds the timestamp of conversation message, sends account and dialog text.Wherein, conversation tag 1 is a marketing session,
Conversation tag 2,3,4 is all request service conversation.
Emotion divides calculation part 13 for carrying out the message that emotion point calculates corresponding each conversation message to each conversation message
Emotion point.
In the present embodiment, emotion divides calculation part 13 to use the double mode emotion identification method based on text and emoji expression,
This method passes through conventional text emotion dictionary (Chinese LIWC sentiment dictionary) and expression emotion dictionary (such as Novak et al.
Emoji sentiment dictionary " Novak, Petra Kralj, et al. " Sentiment of emojis. " PloS of proposition
One10.12 (2015): dialog text is compared e0144296. ") to complete sentiment analysis.
Specifically, emotion point calculation part 13 divides computing unit 132 by text participle unit 131, participle emotion and disappears
Breath emotion divides setup unit 133 to complete double mode emotion identification method.
When emotion is divided calculation part 13 to carry out emotion point to a conversation message and calculated, text participle unit 131 will talk with
The dialog text of message carries out participle to obtain multiple participles.The participle be divided into text participle (i.e. conventional text participle) with
And expression participle (i.e. the emoji expression of user's transmission).
Further, participle emotion divides computing unit 132 to carry out sentiment analysis to each participle according to the sentiment dictionary prestored
To obtain the emotion point of each participle, i.e., each participle is carried out comparing to obtain one by one by above-mentioned sentiment dictionary each
The emotion of participle point.Wherein, when a participle indicates positive emotion, the emotion of the participle (such as " laughing at ") is divided into 1, when one
Participle indicates that the emotion of the participle (such as " shedding tears ") when negative sense emotion is divided into -1, the participle when a participle does not have emotion
The emotion of (such as " ") is divided into 0.
After the emotion for obtaining all participles point, message emotion divides setup unit 133 according to participles all in conversation message
Emotion point and set the message emotion of the conversation message point.If in a conversation message all positive negative affects point and it is (net
Score) to be divided into 1 on the occasion of the emotion of the then conversation message, the emotion of the conversation message is divided into -1 if clear is divided into negative value, if
The emotion that clear is divided into 0 conversation message is divided into 0.
The sample calculation of one message emotion point is as shown in the table:
2 message emotion of table divides example
In table 2, " [shedding tears] " is an emoji expression, is identified as indicating negative emotion in expression emotion dictionary, i.e.,
Corresponding participle emotion is divided into -1.
User feeling Department of Statistics 14 is for counting the message feelings that each user account corresponds to conversation message whithin a period of time
The user feeling point that sense is allocated as each user.
It is for a period of time one month in the present embodiment, the corresponding user account of each user feeling point is in one month
The message emotion of all conversation messages point.
User's position determination unit 15 determines whether user is negative user according to be averaged emotion point of user.
In the present embodiment, message emotion point in user feeling point is lower than preset negative position by user's position determination unit 15
The conduct of scoring (such as value -0.5) is negatively made a speech, and determines all negative speeches in the user account currently determined
Whether the accounting of all speech quantity is higher than preset ratio (such as higher than 80%), determines that the user is negative uses if being higher than
Family.
Negative user storage unit 16 is stored the user account of negative user as negative account.
Negative user search part 17 is used for after emotion divides calculation part 13 to calculate message emotion point, retrieval negative user storage
Portion is to obtain negative account.
Emotion divides re-computation portion 18 by the corresponding message emotion point of negative account according to a preset distribution weight weight
New message emotion point is newly calculated.The value range of the distribution weight is 0.1~0.9.
In the present embodiment, the value for distributing weight is 0.8, therefore, is carried out in the message emotion to a negative account point
When re-computation, message emotion branch is directly multiplied by distribution weight 0.8, to form the lower new message emotion of a numerical value
Point, the calculating for the customer impact emotion point for avoiding some speeches very negative.
In other embodiments, distribution weight can also be weight corresponding with multiple grades, and negative account is according to more
A ranking score is matched, and emotion divides the higher negative account of 18 In Grade of re-computation portion to distribute lower distribution weight, thus more accurate
Reduce influence of the negative user to emotion point in ground.
Session affection computation portion 19 is used to disappear according to the conversation message for corresponding to user account in each session and accordingly
Emotion point is ceased, the service conversation emotion point of corresponding each service conversation and the marketing session feelings of corresponding each marketing session are calculated
Sense point.
In the present embodiment, service conversation emotion point is identical as the calculation method of marketing session emotion point, is calculated one
The average value of the message emotion point of all conversation messages by user's speech, i.e., the total session emotion of each session point in section session
Cust_SentiitAre as follows:
In formula, C indicates that the sum of each session, J indicate the number of the user, Cust_SentiijtcIt is the t weeks
In the dialog procedure c of an interior employee i and user j, the message emotion point of the emotion satisfaction of user, wherein i represents a zooid
Work, j indicate that a consumer, t represent one week, and c represents some session in one week.
Employee's affection computation portion 110 be used for according to the message emotion point of the conversation messages of each all transmissions of employee's account according to
The secondary employee for the calculating each employee's account emotion that be averaged scores.
In the present embodiment, employee is averaged what emotion scoring divided for the message emotion of all conversation messages of each employee speech
Average value.In other embodiments, which, which can also reject when calculating in employee's speech, belongs to marketing
The speech of session, thus according to other conversation messages obtain employee be averaged emotion scoring.
Service quality determine generating unit 111 be used for according to the mean value of each employee's account corresponding service conversation emotion point with
And employee's emotion scoring that is averaged determines that the service quality of each employee determines information to generate corresponding service quality.
In the present embodiment, for each employee's account, it is all that service quality determines that generating unit 111 is corresponded to according to the account
The emotion that the average value (i.e. average service session emotion point) and employee that service conversation emotion is divided is averaged, which scores, determines the employee's
Service quality.
In the present embodiment, service quality determines information for service quality scoring, and service quality determines that generating unit 111 is generating
When service quality scores, directly the average service session emotion of the same employee's account of correspondence point and the employee emotion that be averaged are scored
It is averaged, and scores as the service quality.
In other embodiments, service quality determines that information can also be quality classification information, such as when average service meeting
Quality classification information is height when words emotion point and employee are averaged emotion scoring more than certain numerical value, when average service session feelings
Quality classification information is low when sense point and employee are averaged emotion scoring below certain numerical value, when average service session emotion point
And employee be averaged emotion scoring when being other situations quality classification information be general.
Market content determines that generating unit 112 is used to whether determine each marketing session according to each marketing session emotion point
It is unwelcome to determine information to generate corresponding market content.
In the present embodiment, market content determines that information is divided into unwelcome content and general content, and market content determines
Generating unit 112 will be known as marketing session of the marketing session emotion point lower than -0.2 and be determined as unwelcome content, and by remaining
It is determined as general content.
Prompt information generating unit 113 is used to generate corresponding service quality prompt information according to service quality judgement information,
And determine that information generates corresponding market content and determines prompt information according to market content.
In the present embodiment, prompt information generating unit 113 is when generating service quality prompt information, according to scheduled sentence lattice
Service quality scoring, such as " your service quality scoring is XX " is added in formula.In addition, when service quality scoring is lower than preset value
(such as -0.2), prompt information generating unit 113 can also additionally generate reminding employees in service quality prompt information and need to be improved
Information of service, such as " your service quality is poor, it is noted that is promoted " etc..
In other embodiments, if service quality determines that information is quality classification information, then prompt information generating unit 113
When generating quality of service information, determine that the quality classification information of information generates corresponding service quality and prompts according to service quality
Information, such as when quality classification information is low, generation " your service quality is poor, it is noted that is promoted " etc. is for prompting employee
Pay attention to the prompt information improved service quality.
In the present embodiment, prompt information generating unit 113 is when generating market content judgement prompt information, only in market content
When determining that information is unwelcome content, generates market content corresponding with marketing session and determine prompt information, such as " marketing
Content is unwelcome, it is proposed that pays attention to later " etc..
In the present embodiment, the service quality prompt information that prompt information generating unit 113 generates is opposite with each employee's account
It answers, after generation, service quality prompt information is just sent respectively to accordingly by service side communication unit 114 according to each employee's account
Employee's terminal.Market content determines that prompt information is equally corresponding with employee's account, after generation, services side communication unit 114
Market content is determined into prompt information and corresponding market content (i.e. the marketing message that employee sends in marketing session) basis
Each employee's account is sent respectively to corresponding employee's terminal 2.
In the present embodiment, service side communication unit 114 passes through administrator terminal when sending prompt information and market content
4 are sent to employee's terminal 2.
Fig. 3 is the structural block diagram of employee's terminal in the embodiment of the present invention.
As shown in figure 3, employee's terminal 2 has employee side picture storage unit 21, employee side input display unit 22, employee side logical
Letter portion 23 and employee side control unit 24.
Wherein, employee side communication unit 23 be used to carry out between each composition part of employee's terminal 2 and employee's terminal 2 with
Data communication between other terminals, employee side control unit 24 contain for each composition part to employee's terminal 2 work into
The computer program of row control.
Employee side picture storage unit 21 is stored with community and shows that picture and prompt information show picture.In the present embodiment,
Community shows that picture and prompt information show that picture is all the display picture of community module.
Community information display screen is used to show when employee logins community and shows the dialogue in community in this screen
Message allows employee to check.In the present embodiment, the editor that employee can also carry out message by the picture sends operation.
Prompt information shows that picture is used to show when employee's terminal receives prompt information and shows phase in this screen
The prompt information answered.In the present embodiment, prompt information shows that picture is the session that employee's account and administrator's account form two-by-two
Interface, prompt information are shown in the form of conversation message in the session interface.
Employee side input display unit 22 is for showing above-mentioned picture, so that employee be allowed to carry out corresponding people by these pictures
Machine interaction.
In other embodiments, employee's terminal 2 can also have a reminding module, for when receiving prompt information
Prompt information is checked by displaying the prompt box, generating the modes reminding employees such as prompt tone.
Fig. 4 is the structural block diagram of user terminal in the embodiment of the present invention.
As shown in figure 4, user terminal 3 has user side picture storage unit 31, user side input display unit 32, user side logical
Letter portion 33 and user side control unit 34.
Wherein, user side communication unit 33 be used to carry out between each composition part of user terminal 3 and user terminal 3 with
Data communication between other terminals, user side control unit 34 contain for each composition part to user terminal 3 work into
The computer program of row control.
User side picture storage unit 31 is stored with community and shows picture.In the present embodiment, community shows that picture is community mould
The display picture of block.
Community information display screen is used to show when user logins community and shows the dialogue in community in this screen
Message allows user to check.In the present embodiment, the editor that user can also carry out message by the picture sends operation.
User side input display unit 32 is for showing above-mentioned picture, so that employee be allowed to carry out corresponding people by these pictures
Machine interaction.
Fig. 5 is the structural block diagram of administrator terminal in the embodiment of the present invention.
As shown in figure 5, administrator terminal 4, which has, logins acquisition unit 41, management-side communication unit 42 and management-side control unit
43。
Wherein, management-side communication unit 42 is used between each composition part of the person's of being managed terminal 4 and administrator is whole
Data communication between end 4 and other terminals, management-side control unit 43 contain for each composition part to administrator terminal 4
Work the computer program controlled.
In the present embodiment, the acquisition request of the transmission of dialog information acquisition unit 11, management are received in management-side communication unit 42
Side control unit 43 logins acquisition unit 41 with regard to control and logins each community by community module and administrator's account and successively will be every
Conversation message all in the past period further controls management-side communication unit as a conversation message stream in a community
All conversation message stream is sent to service managing apparatus 1 by 42.
In the present embodiment, the value of preset time period is identical as the regular set of dialog information acquisition unit 11, that is, logins
Acquisition unit 41 obtains all conversation messages in one week.
Fig. 6 is the flow chart of service quality optimization management process in the embodiment of the present invention.
As shown in fig. 6, service quality optimization management process includes the following steps:
Step S1, dialog information acquisition unit 11 access community module by administrator's account and obtain the dialogue of corresponding community
Message flow, subsequently into step S2；
Step S2, dialogue identification part 12 successively carry out each dialog text in the conversation message stream obtained in step S1
Dialogue identification is to obtain multiple sessions of each community, subsequently into step S3；
Step S3, each conversation message in conversation message stream that emotion divides calculation part 13 successively to obtain to step S1 carry out
Emotion point calculates, and obtains corresponding to the message emotion point of each conversation message, subsequently into step S4；
Step S4, negative user search part 17 retrieve negative user storage unit 16 to obtaining negative account, subsequently into
Step S5；
Step S5, the corresponding message emotion point basis point of the negative account that emotion point re-computation portion 18 obtains step S4
It recalculates to obtain new message emotion point with weight, subsequently into step S6；
Step S6, session affection computation portion 19 is according to the conversation message for corresponding to user account in each session and accordingly
Message emotion point calculates the service conversation emotion point of corresponding each service conversation and the marketing session of corresponding each marketing session
Emotion point, subsequently into step S7；
Step S7, employee's affection computation portion 110 is according to the message emotion of the conversation message of each all transmissions of employee's account
Point mean value successively calculate each employee's account employee be averaged emotion score, subsequently into step S8；
Step S8, service quality determine generating unit 111 according to the flat of all service conversation emotions of each employee's account point
Mean value determines that the service quality of each employee determines information to generate corresponding service quality, subsequently into step S9；
Step S9, market content determine that generating unit 112 determines each marketing session according to each marketing session emotion point
It is whether unwelcome to generate corresponding market content judgement information, subsequently into step S10；
Step S10, prompt information generating unit 113 determine that information generates accordingly according to the service quality generated in step S8
Service quality prompt information, and determine that information generates corresponding market content judgement and mentions according to the market content generated in step S9
Show information, subsequently into step S11；
Step S11, service side communication unit 114 propose the service quality prompt information that step S10 is generated, market content judgement
Show that information and corresponding market content are sent to corresponding employee's terminal according to employee's account, subsequently into end state.
In the present embodiment, the time judging unit judgement of dialog information acquisition unit 11 reaches regular set, will be again
It obtains the conversation message stream of each community and executes above-mentioned steps again.
By the above process, organization can periodically preferably to the service quality of employee be monitored management and
Reminding improves.
Embodiment action and effect
According to the service quality optimization management system provided in this embodiment based on community system, due to having dialogue identification
Portion, thus the conversation message stream that dialog information acquisition unit is obtained from community can be carried out identification to obtain service conversation with
And marketing session.Also calculation part is divided to carry out emotion recognition, therefore meeting to each conversation message in conversation message stream with emotion
Words affection computation portion can further calculate the session emotion point of each session according to the message emotion of these conversation messages point, make
Service quality determines that generating unit determines the service quality of employee according to the session emotion of service conversation point, market content is made to determine life
It is not popular with users that market content point determined whether there is according to the session emotion of marketing session at portion, by by these data-pushings
To employee, it will be able to so that employee is recognized the service effectiveness of itself and the preference of user, to pointedly improve itself
Service quality.The present invention excavates the user feeling in conversation message by arrangement and emotion recognition to community message, from
And play the role of monitoring to service quality of the employee when servicing user by community, it is more advantageous to enterprise and employee's root
Service quality is improved according to obtained information, the professional ability of the final satisfaction for improving consumer and employee.
In embodiment, due to user feeling Department of Statistics and user's position determination unit, therefore can be from pair of community
User for a long time containing negative emotions is analyzed in words message, to store its user account to this by negative user storage unit
Class negative user is monitored.It is not only convenient for enterprise and pointedly carries out stronger monitoring and decision, prevent such
The long-term negative emotions of user impact emotion scoring.
It, can also be to service state of the employee in the service of offer due to employee's affection computation portion in embodiment
Degree is monitored, to be more advantageous to the service quality scoring for evaluating each employee, also enterprise can be helped to recognize each zooid
Performance of work when for user service.
In embodiment, due to combining text emotion dictionary and expression emotion dictionary, text can not only be carried out
Judgment by emotion, the judgement that emotion can also be carried out to expression, can more accurately determine the user emotion in community, more
The satisfaction that consumer is held convenient for enterprise and employee, to promote its service quality.
Above-described embodiment is only used for the specific embodiment illustrated the present invention, and the present invention is not limited to the above embodiments
Description range.
In embodiment, user feeling Department of Statistics counting user emotion point, and by user's position determination unit according to user's feelings
Sense point determines negative user.In other embodiments, user feeling Department of Statistics can also be to each user account in a period of time
Interior speech quantity is counted, thus allow user's position determination unit according to the speech quantity and user feeling point simultaneously to
Family is determined, to keep the judgement of negative user more accurate.For example, can be in above-described embodiment with user's position determination unit
On the basis of determining user feeling point, only speech quantity is higher than certain amount (such as speech quantity necessarily is greater than 20 etc.)
User just can be determined that as negative user.
In embodiment, service quality prompt information determines that information is raw for generating service quality prompt information, market content
Prompt information is determined at market content.In other embodiments, service quality prompt information and market content determine information also
It can be sent in the server of the mechanisms such as enterprise and be stored by service side communication unit, consequently facilitating enterprise is according to these information
It is studied.
In embodiment, administrator terminal is independent computer.In other embodiments, each composition of administrator terminal
Component can also be a part of service manager server, that is, logins a part of acquisition unit as service manager server, manages
Manage service side communication unit and service side control of the function of side communication unit and management-side control unit by service manager server
It completes in portion.
Claims (7)
1. a kind of service quality optimization management system based on community system characterized by comprising
Service managing apparatus；
At least one employee's terminal, is provided with community module corresponding with the community system, and each employee's terminal has
One is used to access employee's account of the community module, and each employee's account is corresponding with a community, employee's account
Number for allowing employee to login the community by employee's account and provide a user question and answer marketing service information；
Multiple user terminals are provided with the community module, have the user account for accessing the community module, for allowing
The user logins the community by the user account and obtains the question and answer marketing service information that the employee provides；
One administrator terminal is provided with the community module, has administrator's account for accessing all community modules
Number, the service managing apparatus logins the community by administrator's account and periodically obtains the text of the dialogue in the community
This,
Wherein, there is the service managing apparatus dialog information acquisition unit, dialogue identification part, emotion to divide calculation part, session emotion
Calculation part, service quality determine that generating unit, market content determine generating unit, prompt information generating unit and service side communication unit,
The dialog information acquisition unit periodically accesses the community module by administrator's account and obtains corresponding each institute
The conversation message stream stating community and being made of a plurality of conversation message,
It is more to obtain that the dialogue identification part engages in the dialogue identification to conversation message described in each item in the conversation message stream
It is a to content type in relation to and include multiple corresponding conversation messages session, the session is divided into service conversation and marketing
Session,
The emotion point calculation part successively the conversation message described in each item carry out emotion point calculate and obtain corresponding to it is right described in each item
The message emotion point of message is talked about,
Session affection computation portion is according to the conversation message and phase for corresponding to the user account in each session
The message emotion answered point calculates the service conversation emotion point of corresponding each service conversation and corresponds to each battalion
The marketing session emotion point of session is sold,
The service quality determines generating unit being averaged according to all service conversation emotions of each employee's account point
Value determines that the service quality of each employee determines information to generate corresponding service quality,
The market content determines that generating unit is according to what each marketing session emotion point determined each marketing session
It is no unwelcome to generate corresponding market content judgement information,
The prompt information generating unit determines that information generates corresponding service quality prompt information, and root according to the service quality
Determine that information generates corresponding market content and determines prompt information according to the market content,
The service quality prompt information and the market content are determined prompt information according to institute by the service side communication unit
It states employee's account and is sent to corresponding employee's terminal.
2. the service quality optimization management system according to claim 1 based on community system, it is characterised in that:
Wherein, the service managing apparatus also have user feeling Department of Statistics, user's position determination unit, negative user storage unit,
Negative user search part and emotion divide re-computation portion,
The user feeling Department of Statistics counts the message that each user account corresponds to the conversation message whithin a period of time
The user feeling point that emotion is allocated as each user,
User's position determination unit determines that the user determines whether negative user according to the user feeling point,
The negative user storage unit stores the user account for being determined as the negative user as negative account,
After the emotion divides calculation part to calculate the message emotion point, the negative user search part retrieves the negative user
Storage unit to obtaining negative account,
The emotion divides re-computation portion by the corresponding message emotion point of the negative account according to a preset distribution
Weight recalculates to obtain new message emotion point, and the value range of the distribution weight is 0.1~0.9.
3. the service quality optimization management system according to claim 1 based on community system, it is characterised in that:
Wherein, the service managing apparatus also has employee's affection computation portion,
Employee's affection computation portion is according to the message feelings of the conversation message of each all transmissions of employee's account
Sense point mean value successively calculate each employee's account employee be averaged emotion score,
Service quality determines generating unit in the average value according to all service conversation emotions of each employee's account point
When determining the service quality of each employee, the service of each employee is determined also according to the employee emotion scoring that is averaged
Quality simultaneously generates corresponding service quality judgement information.
4. the service quality optimization management system according to claim 3 based on community system, it is characterised in that:
Wherein, the service quality determines that information scores for service quality,
The service quality determines generating unit in the flat of all service conversation emotions point for determining each employee's account
Mean value and the corresponding employee be averaged emotion scoring when, the service quality scoring is generated according to the average value of the two.
5. the service quality optimization management system according to claim 1 based on community system, it is characterised in that:
Wherein, the conversation message includes at least dialog text,
The emotion point calculation part includes:
Text participle unit, the dialog text of the conversation message described in each item segments respectively；
Participle emotion divides computing unit, and it is each to obtain to carry out sentiment analysis to each participle according to the sentiment dictionary prestored
The emotion of a participle point；
Message emotion divides setup unit, and according to the emotion of the participles all in each conversation message point, this is right with setting
Talk about the message emotion point of message.
6. the service quality optimization management system according to claim 5 based on community system, it is characterised in that:
Wherein, the sentiment dictionary includes text emotion dictionary and expression emotion dictionary.
7. the service quality optimization management system according to claim 1 based on community system, it is characterised in that:
Wherein, session affection computation portion is calculating the service conversation emotion point and the marketing session emotion timesharing,
The total session emotion of each session divides Cust_SentiitAre as follows:
In formula, C indicates that the sum of each session, J indicate the number of the user, Cust_SentiijtcIt is in the t weeks one
The message emotion of the user point in the dialog procedure c of secondary employee i and user j, wherein i represents an employee, and j indicates one
A consumer, t are represented one week, and c represents some session in one week.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910664645.8A CN110414823B (en) | 2019-07-23 | 2019-07-23 | Community system-based service quality optimization management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910664645.8A CN110414823B (en) | 2019-07-23 | 2019-07-23 | Community system-based service quality optimization management system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110414823A true CN110414823A (en) | 2019-11-05 |
| CN110414823B CN110414823B (en) | 2023-04-07 |
Family
ID=68362516
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910664645.8A Active CN110414823B (en) | 2019-07-23 | 2019-07-23 | Community system-based service quality optimization management system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110414823B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006202132A (en) * | 2005-01-21 | 2006-08-03 | Weave Kk | Marketing method and marketing system |
| CN101937534A (en) * | 2010-09-17 | 2011-01-05 | 上海沐盈信息科技有限公司 | Electronic marketing management system and management method |
| CN105321098A (en) * | 2014-07-29 | 2016-02-10 | 张家豪 | Community marketing system and operation method thereof |
| CN105634927A (en) * | 2016-01-29 | 2016-06-01 | 广西中烟工业有限责任公司 | System and method for promoting brands by converging terminal clients through wechat communities |
| US20160364754A1 (en) * | 2015-06-11 | 2016-12-15 | International Business Machines Corporation | Personalized marketing by deriving the sentiments from telephonic and textual conversation over a mobile device |
| CN106779357A (en) * | 2016-12-02 | 2017-05-31 | 普奥云信息科技(北京)有限公司 | Evaluation of employee management system |
| CN107317744A (en) * | 2017-05-20 | 2017-11-03 | 深圳市前海安测信息技术有限公司 | Health service topsis system and method based on wechat group |
| US10003560B1 (en) * | 2012-08-31 | 2018-06-19 | Sprinklr, Inc. | Method and system for correlating social media conversations |
| CN108241736A (en) * | 2016-12-23 | 2018-07-03 | 林伟斌 | Multi-network community management system and management method thereof |
-
2019
- 2019-07-23 CN CN201910664645.8A patent/CN110414823B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006202132A (en) * | 2005-01-21 | 2006-08-03 | Weave Kk | Marketing method and marketing system |
| CN101937534A (en) * | 2010-09-17 | 2011-01-05 | 上海沐盈信息科技有限公司 | Electronic marketing management system and management method |
| US10003560B1 (en) * | 2012-08-31 | 2018-06-19 | Sprinklr, Inc. | Method and system for correlating social media conversations |
| CN105321098A (en) * | 2014-07-29 | 2016-02-10 | 张家豪 | Community marketing system and operation method thereof |
| US20160364754A1 (en) * | 2015-06-11 | 2016-12-15 | International Business Machines Corporation | Personalized marketing by deriving the sentiments from telephonic and textual conversation over a mobile device |
| CN105634927A (en) * | 2016-01-29 | 2016-06-01 | 广西中烟工业有限责任公司 | System and method for promoting brands by converging terminal clients through wechat communities |
| CN106779357A (en) * | 2016-12-02 | 2017-05-31 | 普奥云信息科技(北京)有限公司 | Evaluation of employee management system |
| CN108241736A (en) * | 2016-12-23 | 2018-07-03 | 林伟斌 | Multi-network community management system and management method thereof |
| CN107317744A (en) * | 2017-05-20 | 2017-11-03 | 深圳市前海安测信息技术有限公司 | Health service topsis system and method based on wechat group |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110414823B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108764649A (en) | Insurance sales method for real-time monitoring, device, equipment and storage medium | |
| US20140280621A1 (en) | Conversation analysis of asynchronous decentralized media | |
| CN111489095B (en) | Risk user management method, apparatus, computer device and storage medium | |
| CN110210762A (en) | A kind of employee performance dynamic wire examination method based on big data analysis | |
| CN110570215A (en) | Intelligent customer service system | |
| CN104008182A (en) | Measuring method of social network communication influence and measure system thereof | |
| CN111461636A (en) | Virtual robot-based government affair service platform and application | |
| CN108197963A (en) | A kind of intelligent customer service manages system | |
| CN110427620A (en) | Service quality optimization management system based on community system | |
| CN110414999A (en) | A kind of service quality administrative system based on community system | |
| CN110414823A (en) | Service quality optimization management system based on community system | |
| CN115168603B (en) | Automatic feedback response method, device and storage medium for color ring back tone service process | |
| CN111047362A (en) | Statistical management method and system for use activity of intelligent sound box | |
| CN114418366B (en) | Data processing method and device for intelligent cloud interview | |
| Makoza et al. | How gender is embedded in the national ICT policy of Malawi | |
| CN106817710A (en) | The localization method and device of a kind of network problem | |
| CN116167587A (en) | Automatic customer service distribution method and storage system | |
| CN116189346A (en) | Intelligent queuing and calling system for digital science and technology deep integration hall management and marketing scene | |
| CN109859026A (en) | A kind of Audit data management system | |
| WO2021073258A1 (en) | Task follow-up method, apparatus and device based on emotion analysis, and storage medium | |
| CN112929502B (en) | Voice recognition method and system based on electric power customer service | |
| US20230222449A1 (en) | Online Interview Method and System | |
| Pandey et al. | If you build it, will they come? Exploring narratives that shape the internet in Nepal | |
| CN107742220A (en) | Generation method, big data system and the storage medium of industry white paper | |
| CN113516572A (en) | Online learning statistics and evaluation system |
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 |