CN108335110A - Chat message processing method and processing device - Google Patents

Chat message processing method and processing device Download PDF

Info

Publication number
CN108335110A
CN108335110A CN201710031872.8A CN201710031872A CN108335110A CN 108335110 A CN108335110 A CN 108335110A CN 201710031872 A CN201710031872 A CN 201710031872A CN 108335110 A CN108335110 A CN 108335110A
Authority
CN
China
Prior art keywords
conversation content
user
chat record
content entry
entry
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
CN201710031872.8A
Other languages
Chinese (zh)
Other versions
CN108335110B (en
Inventor
梁伟
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.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding 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 Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Priority to CN201710031872.8A priority Critical patent/CN108335110B/en
Publication of CN108335110A publication Critical patent/CN108335110A/en
Application granted granted Critical
Publication of CN108335110B publication Critical patent/CN108335110B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/01Customer relationship services
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35Clustering; Classification

Abstract

The embodiment of the present application discloses chat message processing method and processing device, wherein the method includes:Server-side determines that chat record content to be analyzed and preset classifying rules information, the chat record content include that first is input by user multiple and problem seeks advice from relevant conversation content entry;Using the classifying rules information, the problem classification belonging to the conversation content entry is determined;The conversation content entry of each problem category is counted and statistical result is provided.Pass through the embodiment of the present application, the conversation content entry that asks questions input by user to first it can carry out classification and statistical analysis in an automated manner, and statistical result can be supplied to second user or operation personnel etc., work without carrying out artificial statistical analysis by a large amount of operation personnel, to which statistical efficiency can be improved, and human cost and time cost can be reduced.

Description

Chat message processing method and processing device
Technical field
This application involves Internet technical fields, more particularly to chat message processing method and processing device.
Background technology
With the fast development of the technologies such as the constantly improve and traditional communication of E-commerce transaction platform, mobile communication, More and more people obtain required commodity in daily life by way of shopping online.In E-commerce transaction platform In system, it may include system platform operator, consumer or buyer user (can be described as the first user, can by client or The information of the web page browsing businessman either business object of seller user's publication is simultaneously bought etc.), businessman or seller user (can Referred to as second user can be issued its business object sold by the platform, and provide the detail information about business object Deng).
In the operational process of E-commerce transaction platform, when the first user encounters certain technical problems, traffic issues etc., It can be exchanged with the technical staff etc. of the contact staff of second user or system platform to seek to solve by instant messaging application Certainly scheme finds clicking related chain for example, when the first user wants a certain business object in purchase second user shop Always there is a phenomenon where dodging to move back when connecing, then above-mentioned phenomenon can be fed back to the related technical personnel of system platform, so that above-mentioned phenomenon It is addressed;For another example, when the first user wants a certain business object in purchase second user shop, it is desirable to its price energy It is enough more preferential again, then it can be linked up with the contact staff in second user shop, with ask whether can be by the business object Price turn down some, etc..
Currently, some E-commerce transaction platforms are wished to the chat record content generated in above-mentioned chat process according to certain A little demands are counted (for example, the problem of the first user is proposed carries out statistic of classification etc.), to be optimized to platform, usually It is that artificial statistical analysis is carried out to obtain statistical result, due to chat record content packet to chat record content by operation personnel The data volume contained is very big, and therefore, it is necessary to call a large amount of operation personnel to take a substantial amount of time can just obtain accordingly Statistical result, statistical efficiency is relatively low, and human cost and time cost are higher.
In short, how to improve the statistics of the chat record content to being generated in above-mentioned E-commerce transaction platform operational process Efficiency becomes the technical issues of needing those skilled in the art to solve.
Invention content
This application provides chat message processing method and processing device, can be improved in E-commerce transaction platform operational process The statistical efficiency of the chat record content of generation saves human cost and time cost.
This application provides following schemes:
A kind of chat record information processing method, including:
Server-side determines chat record content to be analyzed and preset classifying rules information, the chat record content Include that first is input by user multiple and problem seeks advice from relevant conversation content entry;
Using the classifying rules information, the problem classification belonging to the conversation content entry is determined;
The conversation content entry of each problem category is counted, and statistical result is provided.
A kind of chat record information processing method, including:
Client determines chat record file to be analyzed;
According to the chat record file to be analyzed, analysis request is generated, and be submitted to server-side, by the server-side It is raw to determine that chat record content to be analyzed and preset classifying rules information determine institute using the classifying rules information The problem classification belonging to conversation content entry is stated, the conversation content entry of each problem category is counted, and statistics knot is provided Fruit.
A kind of chat record information processing unit is applied to server-side, including:
Content and information determination unit, for determining chat record content to be analyzed and preset classifying rules letter Breath, the chat record content include that first is input by user multiple and problem seeks advice from relevant conversation content entry;
Problem category determination unit is determined for utilizing the classifying rules information belonging to the conversation content entry Problem category;
Statistic unit counts for the conversation content entry to each problem category, and provides statistical result.
A kind of chat record information processing unit is applied to client, including:
Chat record document determining unit, for determining chat record file to be analyzed;
Analysis request submits unit, for according to the chat record file to be analyzed, generating analysis request, and submit To server-side, is given birth to by the server-side and determine chat record content to be analyzed and preset classifying rules information, utilize institute Classifying rules information is stated, determines the problem classification belonging to the conversation content entry, to the conversation content entry of each problem category It is counted, and statistical result is provided.
According to specific embodiment provided by the present application, this application discloses following technique effects:
By the embodiment of the present application, it can first determine that chat record content to be analyzed is (input by user including first It is multiple to seek advice from relevant conversation content entry with problem) and preset classifying rules information, recycle the classifying rules to believe Breath, determines the problem classification belonging to the conversation content entry, is then counted to the conversation content entry of each problem category And statistical result is provided, with this, can by way of automation the conversation content entry that asks questions input by user to first Classification and statistical analysis are carried out, and statistical result can be supplied to second user or operation personnel etc., without by a large amount of Operation personnel carry out the work of artificial statistical analysis, to which statistical efficiency can be improved, and can reduce human cost and time at This.
Certainly, any product for implementing the application does not necessarily require achieving all the advantages described above at the same time.
Description of the drawings
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, below will be to institute in embodiment Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the application Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings Obtain other attached drawings.
Fig. 1 is system block diagram provided by the embodiments of the present application;
Fig. 2 is the flow chart of first method provided by the embodiments of the present application;
Fig. 3 is the flow chart of second method provided by the embodiments of the present application;
Fig. 4 is the schematic diagram of first device provided by the embodiments of the present application;
Fig. 5 is the schematic diagram of second device provided by the embodiments of the present application.
Specific implementation mode
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete Site preparation describes, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on Embodiment in the application, the every other embodiment that those of ordinary skill in the art are obtained belong to the application protection Range.
The efficiency counted is asked questions to the first user in chat record content in order to improve, shown in Fig. 1, this Application embodiment provides a kind of chat record information processing system, the system towards user can be second user (such as quotient Family or seller user etc.), can also be system operation personnel, that is to say, that can be that second user or system operation personnel carry It, can be by its installing terminal equipment and using the client, to divide specified chat record file for client Analysis, checks specific statistical result, and can be improved and optimize to business object or to system etc. according to statistical result.Separately Outside, which can also include server-side, and in specific implementation, chat record text to be analyzed can be specified by client Part, server-side can first determine chat record content to be analyzed (including the first input by user multiple and problem consulting Relevant conversation content entry) and preset classifying rules information, the classifying rules information is recycled, determines the dialogue Then problem classification belonging to content item is counted and is provided statistical result to the conversation content entry of each problem category, With this, the conversation content entry that asks questions input by user to first classification and statistical can be carried out by way of automation Statistical result, and can be supplied to second user or operation personnel etc. by analysis, without by a large amount of operation personnel into pedestrian The work of work statistical analysis to which statistical efficiency can be improved, and can substantially reduce human cost and time cost.
It describes in detail below to concrete implementation mode.
Embodiment one
Referring to Fig. 2, which from the angle of server-side, provides a kind of chat record information processing method first, It may include steps of:
S201, server-side determine chat record content to be analyzed and preset classifying rules information.
Wherein, it may include that the first user (for example, consumer-user or buyer user etc.) is defeated in the chat record content The multiple and problem entered seeks advice from relevant conversation content entry.
In specific implementation, according to the difference of application scenarios, the chat record content can be different.For example, in face Into the application scenarios of second user, the chat record content to be analyzed may include that the first user chats with second user The chat record content generated during it, it is described to may include with the relevant conversation content entry of problem consulting:With second user The problem of in terms of associated business object, seeks advice from relevant conversation content entry.Wherein, so-called business object can be second The shop object of user or merchandise items etc..In practical applications, the first user is during browsing or purchase, such as There are queries to certain business object for fruit, then can by the interface provided in the page or independent instant messaging tools etc., with The Customer Service people of second user initiates dialogue, and the first user can input text so that its problem to be described, and second uses The Customer Service people at family is answered.In such dialog procedure, so that it may to generate chat record content.
In the application scenarios of system-oriented operation personnel, the chat record content to be analyzed may include the first user The chat record content generated in chat process is carried out with system operation personnel, it is described to seek advice from relevant conversation content item with problem Mesh may include and the problem of technical aspect seeks advice from relevant conversation content entry.In practical applications, sales platform can pass through Various ways provide the entrance that can be engaged in the dialogue with system operation personnel for the first user, if the first user is using sale During platform, the problem of finding some technical aspects, for example, when opening certain page, there is a problem of that refresh rate is slow, Alternatively, when executing certain operation, program automatically exits from problem, etc., then can be entered by this entrance and system The interface that operation personnel engages in the dialogue, to oneself needing the problem of seeking advice to be described in the edit box at interface, system operation Personnel can carry out replying upon receipt, etc..In this way, for this kind of dialogue, chat record content can also be generated.
In practical applications, chat record content can be recorded by instant communication server, alternatively, in each use The client at family can also locally carry out recording, etc. in terminal device.It in the embodiment of the present application, can be in advance from server Chat record content is downloaded, as data source to be analyzed, alternatively, if terminal device locally also saves chat record Content then directly locally can also read chat record content from terminal device.Wherein, since chat record content is to talk with The record generated during double hair dialogues, therefore, chat record content is typically to be made of the conversation content entry of a rule, When preserving chat record, form can be preserved there are many specific, for example, preserving etc. in the form of a file.Wherein, right In this with chat record content existing for document form, this document is commonly known as chat record file, wherein saving more The information such as a conversation content entry and the corresponding speaker identification of each entry, time limit of speech.Therefore, implement in the application In example, chat record file to be analyzed can be first determined, then therefrom extraction first is input by user relevant right with problem consulting Content item is talked about, with the chat record content that determination is to be analyzed.Wherein, chat record file can be in advance under server It carries, or preserve in terminal device local in advance, in short, when being analyzed, the chat that can be prepared in advance is remembered File is recorded, as the input data of specific statistical analysis module.
It should be noted that since chat record content has this attribute of generation time, theoretically, from second user From when issuing the business objects such as its shop or commodity, as long as second user has the Customer Service people of oneself, it is possible to meeting The chat record between the first user is generated, also it is possible to which new chat record can be generated constantly.First user transports with system Chat record content between battalion personnel be also have the characteristics that it is same, that is, chat record content may be with the time not Break and is increasing.And in the embodiment of the present application, can be that analysis system is carried out to the history chat record content in designated time period Meter, such as, it is possible to specify nearest one month chat record content is analyzed and counted, alternatively, to nearest chat in one week Record content analyzes and counts, alternatively, can be with the initial time of designated analysis and end time, e.g., to 1 day to 4 April The chat record generated between months 30 days analyze, etc..The above configuration information can be the user by system Voluntarily it is arranged according to actual demand.
It is multi-party due to being generally included in the chat record file if the data source of user's input is chat record file The information in face, including the sender information of each conversation content entry, transmission time information etc., also, which includes right The conversation content entry of both sides is talked about, and only needs to carry out the problems in first user's enquirement process text in the embodiment of the present application Analysis statistics, accordingly, with respect to dialogues such as the answer of second user Customer Service people or system operation personnel in dialog procedure Content item, and the bipartite greeting of dialogue, conclusion etc., actually belong to useless data, therefore, specific When realization, the first input by user multiple and relevant conversation content of problem consulting can be extracted first from chat record file Entry.
In specific implementation, in order to achieve the purpose that said extracted, various ways may be used to realize, for example, can be with Above-mentioned hash is rejected first, in this way, it is remaining be exactly needed for the embodiment of the present application first it is input by user with ask Topic seeks advice from relevant conversation content entry.Specifically when rejecting hash, it is possibility to have a variety of realization methods, for example, wherein Under a kind of mode, second user Customer Service people or the conversation content item of system operation personnel input can be rejected first Mesh.When specific implementation, since chat record file includes the sender information of each conversation content entry, and second user is objective The quantity of family attendant and system operation personnel all have relative immobility, for example, certain second user may provide six Customer Service people, etc., also, the mark such as the account of these personnel, ID can be known in advance for user, Therefore, in order to achieve the purpose that reject this partial data, user can pre-set attendant's list information, the service Staff list is used to record the identification information (such as User ID etc.) of the attendant to engage in the dialogue with the first user, wherein clothes Business personnel may include the attendant of second user, system operation personnel etc..In this way, due in the chat record file, It may include dialog text content and speaker identification information (such as User ID etc.), it therefore, can be by the chat record file In, speaker identification conversation content entry corresponding with the entry that the identification information recorded in attendant's list matches It is deleted, that is, the conversation content that the Customer Service people of second user is inputted is deleted, alternatively, by system operation personnel The conversation content of input is deleted.
After the conversation content for inputting attendant is deleted, remaining conversation content can be considered that first is input by user Therefore conversation content can determine that described first is input by user relevant with problem consulting according to the remaining conversation content Conversation content entry.
As it was noted above, can may also have the conversation contents such as some greetings, conclusion in conversation content entry, such as Therefore first user and the conversation content entry (for example, " hello ", " ", etc.) of the mutual greeting of attendant will take After the conversation content of business personnel input is deleted, in the remaining conversation content, some may be will also include and seeked advice from problem Unrelated hash,.In order to which those hashes are rejected, it is possibility to have a variety of realization methods.For example, one of which is specific Realization method under, hash can be pre-set and reject rule, rule is rejected using the preset hash, will be upper State and seek advice from unrelated dialogue entries with problem in remaining conversation content and deleted, obtain described first it is input by user with ask Topic seeks advice from relevant conversation content entry, to reduce the data volume in subsequent statistical analysis step, saves overhead, improves system Count efficiency.
When specific implementation, it can pre-set and reject regular corresponding regular expression with hash, for example can be: “'\s*\S+\s*=+=s*Instant message s*=+=[s S] *=+=' ", can in remaining conversation content into The matching of the capable regular expression, the dialogue entries to match with the regular expression then can be identified as hash, and can incite somebody to action Those dialogue entries are rejected, and the purpose that unrelated dialogue entries are deleted will be seeked advice from problem to realize.
Certainly, in other realization methods, the relevant hashes such as above-mentioned and greeting can also be first rejected, then reject clothes The conversation content entry, etc. of business personnel input, is no longer described in detail one by one here.
S202 determines the problem classification belonging to the conversation content entry using the classifying rules information.
After determining the first relevant conversation content entry input by user with problem consulting, so that it may with to these dialogues Content item is classified, to determine the problem classification belonging to each conversation content entry.It, can be in order to reach the purpose Classifying rules is pre-set, is classified according to the rule.Specific classifying rules can there are many, in a kind of specific implementation side In formula, the classifying rules information may include that the associated key word information of each problem category, the keyword such as can be " to move back Goods ", " reimbursement ", " delivery " etc., each problem category key word information associated with it can be preserved in the form of file etc., example Such as, the content specifically preserved in file can be as shown in table 1:
Table 1
Serial number Problem category Keyword
1 Classification 1 Keyword A
2 Classification 2 Keyword B
…… …… ……
In the case where providing above-mentioned key word file, can first judge in the conversation content entry with the presence or absence of certain pass Keyword, if there is certain keyword, then can according to the keyword, by such as by inquiry table 1 in a manner of determine the dialogue Problem classification belonging to content item.
In another specific implementation, the classifying rules may also include the associated regular expressions of each problem category Formula.
When specific implementation, it can would generally be used previously according to the first user is summarized the experience out when seeking advice from each problem category Which type of expression way carries out particular problem description, and it is special to extract corresponding text formatting further according to those expression ways Sign, the text formatting feature that text then can be described according to particular problem under each problem category generate corresponding regular expression.
For example, the first user is when " return of goods " are wanted in expression, it will usually which using " I wants to have moved back XX ", " I wants XX to move back ", " I is not intended to XX " etc. therefore can be according to extracting corresponding text lattice in those expression ways about " return of goods " Then formula feature generates and corresponding regular expression under " return of goods " this problem category according to those text formatting features.By In, for consulting same problem classification under expression way may be one or more, therefore, each problem category can be associated with At least one regular expression.
When specific implementation, each problem category regular expression associated with it can be also saved in key word file, this Sample, the content preserved in key word file can be as shown in table 2:
Table 2
It therefore, can be by inquiry table 2 according to the associated problem category of the keyword, to determine associated regular expression. Wherein, the form of regular expression such as can be:" ' n d { 4 } years d the d months d d days d d:\d\d:\d\d\s*(([^:\ s]+):.*' & $ keyword& " .*) ", wherein " keyword " corresponds to the keyword.
Specifically when determining the problem classification belonging to conversation content entry, the form of traversal may be used, judge the dialogue Whether content item hits certain regular expression, if the conversation content entry hit regular expression, can determine that this is right Words content item belongs to the corresponding problem category of the regular expression.
Certainly, in other specific implementations, there can also be other modes to determine belonging to conversation content entry The problem of classification, I will not elaborate.
S203 counts the conversation content entry of each problem category, and provides statistical result.
In practical applications, it can be counted according to actual needs and obtain corresponding statistical result, for example can determine each Corresponding first user of conversation content entry for including under problem category, and count each problem category corresponding first and use Amount amount, the first number of users, etc. for such as proposing first number of users of " return of goods ", proposing " page, which dodges, to move back ", so as at certain When corresponding first user of one problem category reaches predetermined quantity, which is solved in time, with to the business in shop Object, system platform function etc. be improved, optimize.
For another example, it is total in the first user for participating in puing question to that the first number of users for including under each problem category can be counted Accounting in number, as propose " reimbursement " the first number of users participate in put question to the first total number of users in accounting, propose " page There is mess code in face " accounting, etc. of first number of users in the first total number of users for participating in puing question to, so as to in shop Business object, function of system platform etc. are improved, optimize.
In practical applications, in the application scenarios towards second user, which can be supplied to above-mentioned statistical result The second user client, so as to the second user using the statistical result to the business object there are the problem of into Row improves, to realize the purpose optimized to the business object in shop;In the application scenarios of system-oriented operation personnel, The statistical result can be supplied to the system operation personnel, so that the system operation personnel utilize the statistical result pair The technical aspect there are the problem of be improved, to realize the purpose that is optimized to system.
By the embodiment of the present application, can by way of automation the conversation content that asks questions input by user to first Entry carries out the determination of problem category and carries out statistical analysis, and statistical result can be supplied to second user or operation personnel Deng, the work without carrying out artificial statistical analysis by a large amount of operation personnel, to which statistical efficiency can be improved, and can be big It is big to reduce human cost and time cost.
Further, since the first user be based on its purchase business object there are the problem of and send out consulting, therefore should Chat record information processing system can not be all that second user offer is more thin according to business object when towards second user The statistical result of change, in favor of second user understand in time the specific business object in its shop there are the problem of, and can basis Problem condition carries out the operations such as more targeted improvement, optimization to specific business object.
In specific implementation, since the problem of the first user sends out seeks advice from, typically or recently bought with it Certain business object it is relevant, also, the historical transactional information of the first user would generally be recorded in sales platform, also It is to say, often there is certain relevance between chat record content and historical transaction record.Therefore, this association can be utilized Property, the specific associated business object of conversation content entry is determined, then again from business object, to specific problem category It analyzes and counts.When specific implementation, the period belonging to the chat record to be analyzed can be first determined, then extract described The historical transaction record information that one user generates within the correspondence period, then can according to the historical transaction record information, Determine that described first is input by user and problem seeks advice from the relevant associated business object of conversation content entry, further according to associated The difference of business object, by described first it is input by user seek advice from relevant conversation content entry with problem and be divided into multiple groups, That is, as unit of each business object, conversation content entry is grouped.
Then, the classifying rules can be utilized respectively with the group of the relevant conversation content entry of each business object Information classifies the conversation content entry, and respectively in the group of each business object relative dialog content item, The conversation content entry of each problem category is counted, and provides and is directed to and the relevant conversation content entry of each business object Statistical result.With this, which can provide the system for each business object according to the difference of business object for second user Meter is as a result, based on the statistical result more refined, and more conducively second user understands and asked existing for the business object in its shop Topic, and can be improved business object according to problem condition, optimizes, can lifting system the degree of automation.
Further, due to second user selling operation object during would generally include pre-sales and sell latter two rank Section, therefore, which can not be all also the second use according to the sales stage when towards second user Family provides the statistical result that further refines, be more conducive to second user understand business object in its shop there are the problem of, And business object can be improved according to problem condition, optimized.
Specifically, may include that represent second user chats with first user in preset attendant's list It first sales stage attendant's list and second sales stage attendant's list, in the present embodiment, the first pin It is alternatively referred to as pre-sales to sell the stage, and the second sales stage is alternatively referred to as after sale.
Based on this, can by described first it is input by user seek advice from relevant conversation content entry with problem and be grouped, with It obtains with first sales stage attendant's chat process, first input by user seeks advice from relevant conversation content item with problem Mesh, and with second sales stage attendant's chat process, first input by user seeks advice from relevant dialogue with problem Hold entry, that is to say, that by the first input by user and relevant conversation content entry of problem consulting, be divided into two groups, one group is Seek advice from relevant conversation content entry with problem with being generated in second user pre-sales service personnel's chat process, another group be with What is generated in second user after-sale service personnel's chat process seeks advice from relevant conversation content entry with problem.
Then, respectively in the above-mentioned conversation content with the conversation content entry of pre-sales service personnel and with after-sale service personnel In entry, using the classifying rules information, the conversation content entry is classified, and respectively in above-mentioned and pre-sales service The conversation content entry of personnel and in the conversation content entry of after-sale service personnel, to the conversation content entry of each problem category It is counted, and is provided for pair with the statistical result of the conversation content entry of pre-sales service personnel and with after-sale service personnel The statistical result of content item is talked about, with this, which can provide business object for second user and exist according to the difference of sales stage It is pre-sales and the problem of be respectively present after sale, based on the statistical result further refined, second user can be more conducive to and understand its shop Business object in paving there are the problem of, and business object can be improved according to problem condition, optimized, further promoted The degree of automation of system.
Embodiment two
The embodiment second is that with embodiment one correspondingly, from client (may include second user client or system Operation personnel's client etc.) angle, provide a kind of chat record information processing method, referring to Fig. 3, this method may include Following steps:
S301:Client determines chat record file to be analyzed;
S302:According to the chat record file to be analyzed, analysis request is generated, and be submitted to server-side, by described Server-side life determines chat record content to be analyzed and preset classifying rules information, using the classifying rules information, It determines the problem classification belonging to the conversation content entry, the conversation content entry of each problem category is counted, and provides Statistical result.
It is due to the embodiment second is that corresponding with embodiment one, relevant specific implementation may refer to aforementioned reality The introduction in example one is applied, which is not described herein again.
Corresponding with the chat record information processing method that above-described embodiment one provides, the embodiment of the present application also provides one Kind chat record information processing unit, referring to Fig. 4, which may include that content and information determination unit 401, problem category are true Order member 402 and statistic unit 403, wherein:
Content and information determination unit 401 can be used for determining chat record content to be analyzed and preset classification gauge Then information, the chat record content include that first is input by user multiple and problem seeks advice from relevant conversation content entry.
Problem category determination unit 402 can be used for utilizing the classifying rules information, determine the conversation content entry institute The problem of category classification.
Statistic unit 403 can be used for counting the conversation content entry of each problem category, and provide statistical result.
In specific implementation, the content and information determination unit 401, it may include:
Chat record file determination subelement can be used for determining chat record file to be analyzed;
Conversation content entry extract subelement, can be used for from the chat record file extract first it is input by user with Problem seeks advice from relevant conversation content entry.
Wherein, the chat record file may include dialog text content and speaker identification information, be based on this, described Conversation content entry extracts subelement, can be specifically used for:
Determine that preset attendant's list information, attendant's list engage in the dialogue for recording with the first user Attendant identification information;
By in the chat record file, speaker identification and the identification information phase recorded in attendant's list The corresponding conversation content of entry matched is deleted;
Determine that described first is input by user and problem seeks advice from relevant conversation content entry according to remaining conversation content.
Further, the conversation content entry extracts subelement, can also be specifically used for:
Rule is rejected using preset hash, unrelated dialogue entries will be seeked advice from remaining conversation content with problem It deletes, obtains the described first input by user and relevant conversation content entry of problem consulting.
Wherein, the classifying rules information may include:Each associated key word information of problem category is based on this, described to ask Classification determination unit 402 is inscribed, can be specifically used for:
Judge to whether there is certain keyword in the conversation content entry;
If it is, according to the keyword, the problem classification belonging to the conversation content entry is determined.
Wherein, the classifying rules further includes the associated regular expression of each problem category, the regular expression according to Particular problem describes the text formatting feature generation of text under each problem category, is based on this, described problem classification determination unit 402, it can be specifically used for:
According to the associated problem category of the keyword, associated regular expression is determined;
Judge whether the conversation content entry hits the associated regular expression of the problem category, if it is, determining The conversation content entry belongs to the problem category.
In specific implementation, the statistic unit 403 can be specifically used for:
Determine corresponding first user of the conversation content entry for including under each problem category;
Count corresponding first number of users of each problem category.
Further, the statistic unit 403 can also be specifically used for:
Count accounting of the first number of users for including under each problem category in the first total number of users for participating in puing question to.
In practical applications, preset attendant's list may include representing second user and first user into The first sales stage attendant's list and second sales stage attendant's list of row chat,
Described device may also include:
First grouped element is used for the described first input by user and relevant conversation content entry progress of problem consulting Grouping is obtained with first sales stage attendant's chat process, and first input by user seeks advice from relevant dialogue with problem Content item, and with second sales stage attendant's chat process, first is input by user relevant with problem consulting Conversation content entry;
Based on this, described problem classification determination unit 402, it may also be used for:
Respectively in each group, using the classifying rules information, the conversation content entry is classified;
The statistic unit 403, it may also be used for:
Respectively in each group, the conversation content entry of each problem category is counted, and statistical result is provided.
In the application scenarios in face of second user, the chat record content to be analyzed includes the first user and second User carries out the chat record content generated in chat process, described to include with the relevant conversation content of problem consulting:With second The problem of in terms of the business object of user-association, seeks advice from relevant conversation content,
Based on this, described device may also include:
First statistical result provides unit, can be used for the statistical result being supplied to the second user client, with Toilet state second user using the statistical result in terms of the business object there are the problem of be improved.
Further, described device may also include:
Period determination unit, for determining the period belonging to the chat record to be analyzed;
Historical transaction record information extraction unit, the history generated during that corresponding time period for extracting first user Transaction record information;
Business object determination unit, for according to the historical transaction record information, determining that described first is input by user The relevant associated business object of conversation content entry is seeked advice from problem;
Second packet unit, for the difference according to associated business object, by the described first input by user and problem It seeks advice from relevant conversation content entry and is divided into multiple groups;
Based on this, described problem classification determination unit 402, it may also be used for:
Respectively in each group, using the classifying rules information, the conversation content entry is classified;
The statistic unit 403, it may also be used for:
Respectively in each group, the conversation content entry of each problem category is counted, and statistical result is provided.
In the application scenarios of system-oriented platform side technical staff, the chat record content to be analyzed includes first User carries out the chat record content generated in chat process with platform side technical staff, described to seek advice from relevant dialogue with problem Content includes:Relevant conversation content is seeked advice from the problem of technical aspect,
Based on this, described device may also include:
Second statistical result provides unit, for the statistical result to be supplied to the platform side technical staff, so as to The platform side technical staff using the statistical result to the technical aspect there are the problem of be improved.
Corresponding with embodiment two, the embodiment of the present application also provides a kind of chat record information processing units, referring to figure 5, which is applied to client, can specifically include:
Chat record document determining unit 501, for determining chat record file to be analyzed;
Analysis request submits unit 502, for according to the chat record file to be analyzed, generating analysis request, and It is submitted to server-side, is given birth to by the server-side and determines chat record content to be analyzed and preset classifying rules information, profit With the classifying rules information, the problem classification belonging to the conversation content entry is determined, to the conversation content of each problem category Entry is counted, and provides statistical result.
Chat message processing unit provided by the embodiments of the present application can first determine chat record content to be analyzed (wherein Including the first input by user multiple and relevant conversation content entry of problem consulting) and preset classifying rules information, then Using the classifying rules information, the problem classification belonging to the conversation content entry is determined, then to pair of each problem category Words content item is counted and is provided statistical result, can be by way of automation to the first consulting input by user with this Statistical result can be supplied to second user or operation personnel etc. by the conversation content entry of problem into classification and statistical analysis, Work without carrying out artificial statistical analysis by a large amount of operation personnel, to which statistical efficiency can be improved, and can be significantly Reduce human cost and time cost.
As seen through the above description of the embodiments, those skilled in the art can be understood that the application can It is realized by the mode of software plus required general hardware platform.Based on this understanding, the technical solution essence of the application On in other words the part that contributes to existing technology can be expressed in the form of software products, the computer software product It can be stored in a storage medium, such as ROM/RAM, magnetic disc, CD, including some instructions are used so that a computer equipment (can be personal computer, server either network equipment etc.) executes the certain of each embodiment of the application or embodiment Method described in part.
Each embodiment in this specification is described in a progressive manner, identical similar portion between each embodiment Point just to refer each other, and each embodiment focuses on the differences from other embodiments.Especially for system or For system embodiment, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to method The part of embodiment illustrates.System and system embodiment described above is only schematical, wherein the conduct The unit that separating component illustrates may or may not be physically separated, the component shown as unit can be or Person may not be physical unit, you can be located at a place, or may be distributed over multiple network units.It can root According to actual need that some or all of module therein is selected to achieve the purpose of the solution of this embodiment.Ordinary skill Personnel are without creative efforts, you can to understand and implement.
Above to chat message processing method and processing device provided herein, it is described in detail, it is used herein The principle and implementation of this application are described for specific case, and the explanation of above example is only intended to help to understand The present processes and its core concept;Meanwhile for those of ordinary skill in the art, according to the thought of the application, having There will be changes in body embodiment and application range.In conclusion the content of the present specification should not be construed as to the application Limitation.

Claims (15)

1. a kind of chat record information processing method, which is characterized in that including:
Server-side determines chat record content to be analyzed and preset classifying rules information, is wrapped in the chat record content Include the first input by user multiple and relevant conversation content entry of problem consulting;
Using the classifying rules information, the problem classification belonging to the conversation content entry is determined;
The conversation content entry of each problem category is counted, and statistical result is provided.
2. according to the method described in claim 1, it is characterized in that, determination chat record content to be analyzed, including:
Determine chat record file to be analyzed;
Extraction first is input by user from the chat record file seeks advice from relevant conversation content entry with problem.
3. according to the method described in claim 2, it is characterized in that, the chat record file include dialog text content with And speaker identification information, the extraction first from the chat record file are input by user multiple related to problem consulting Conversation content entry, including:
Determine that preset attendant's list information, attendant's list are used to record the clothes to engage in the dialogue with the first user The identification information of business personnel;
The identification information in the chat record file, recorded in speaker identification and attendant's list is matched The corresponding conversation content of entry is deleted;
Determine that described first is input by user and problem seeks advice from relevant conversation content entry according to remaining conversation content.
4. according to the method described in claim 3, it is characterized in that, described determine that described first uses according to remaining conversation content Family input seeks advice from relevant conversation content entry with problem, including:
Rule is rejected using preset hash, unrelated dialogue entries will be seeked advice from remaining conversation content with problem and deleted It removes, obtains the described first input by user and relevant conversation content entry of problem consulting.
5. according to the method described in claim 3, it is characterized in that, preset attendant's list includes representing the second use The first sales stage attendant's list and second sales stage attendant's name that family is chatted with first user Single, the method further includes:
By described first it is input by user seek advice from relevant conversation content entry with problem and be grouped, obtain and the first sale rank In section attendant's chat process, the first input by user and relevant conversation content entry of problem consulting, and sold with second It sells in stage attendant's chat process, the first input by user and relevant conversation content entry of problem consulting;
It is described to utilize the classifying rules information, determine the problem classification belonging to the conversation content entry, including:
Respectively in each group, using the classifying rules information, the problem classification belonging to the conversation content entry is determined;
The conversation content entry to each problem category counts, and provides statistical result, including:
Respectively in each group, the conversation content entry of each problem category is counted, and statistical result is provided.
6. according to the method described in claim 1, it is characterized in that, the chat record content to be analyzed includes the first user The chat record content generated in chat process is carried out with second user, it is described to include with the relevant conversation content of problem consulting: The problem of in terms of the associated business object of second user, seeks advice from relevant conversation content, and the method further includes:
The statistical result is supplied to the second user client, so that the second user utilizes the statistical result pair In terms of the business object there are the problem of be improved.
7. according to the method described in claim 6, it is characterized in that, further including:
Determine the period belonging to the chat record to be analyzed;
Extract the historical transaction record information that first user generates during that corresponding time period;
According to the historical transaction record information, determine that described first is input by user and problem seeks advice from relevant conversation content item The associated business object of mesh;
According to the difference of associated business object, by the described first input by user and relevant conversation content entry of problem consulting It is divided into multiple groups;
It is described to utilize the classifying rules information, determine the problem classification belonging to the conversation content entry, including:
Respectively in each group, using the classifying rules information, the problem classification belonging to the conversation content entry is determined;
The conversation content entry to each problem category counts, and provides statistical result, including:
Respectively in each group, the conversation content entry of each problem category is counted, and statistical result is provided.
8. according to the method described in claim 1, it is characterized in that, the chat record content to be analyzed includes the first user The chat record content generated in chat process is carried out with platform side technical staff, it is described to seek advice from relevant conversation content with problem Including:Relevant conversation content is seeked advice from the problem of technical aspect, the method further includes:
The statistical result is supplied to the platform side technical staff, so that the platform side technical staff utilizes the statistics As a result to the technical aspect there are the problem of be improved.
9. according to claim 1 to 8 any one of them method, which is characterized in that the classifying rules information includes:Each problem The key word information of category associations;
It is described to utilize the classifying rules information, determine the problem classification belonging to the conversation content entry, including:
Judge to whether there is certain keyword in the conversation content entry;
If it is, according to the keyword, the problem classification belonging to the conversation content entry is determined.
10. wanting the method described in 9 according to right, which is characterized in that the classifying rules further includes that each problem category is associated just Then expression formula, the text formatting feature that the regular expression describes text according to particular problem under each problem category generate;
It is described that problem classification belonging to the conversation content entry is determined according to the keyword, including:
According to the associated problem category of the keyword, associated regular expression is determined;
Judge whether the conversation content entry hits the associated regular expression of the problem category, if it is, determining that this is right Words content item belongs to the problem category.
11. according to claim 1 to 8 any one of them method, which is characterized in that the conversation content to each problem category Entry is counted, and provides statistical result, including:
Determine corresponding first user of the conversation content entry for including under each problem category;
Count corresponding first number of users of each problem category.
12. according to the method described in claim 9, it is characterized in that, the conversation content entry to each problem category carries out Statistics, and statistical result is provided, further include:
Count accounting of the first number of users for including under surely each problem category in the first total number of users for participating in puing question to.
13. a kind of chat record information processing method, which is characterized in that including:
Client determines chat record file to be analyzed;
According to the chat record file to be analyzed, analysis request is generated, and be submitted to server-side, is given birth to by the server-side true Fixed chat record content to be analyzed and preset classifying rules information, using the classifying rules information, it is described right to determine The problem classification belonging to content item is talked about, the conversation content entry of each problem category is counted, and statistical result is provided.
14. a kind of chat record information processing unit, which is characterized in that it is applied to server-side, including:
Content and information determination unit, for determining chat record content to be analyzed and preset classifying rules information, institute It includes that first is input by user multiple and problem seeks advice from relevant conversation content entry to state chat record content;
Problem category determination unit determines the problem belonging to the conversation content entry for utilizing the classifying rules information Classification;
Statistic unit counts for the conversation content entry to each problem category, and provides statistical result.
15. a kind of chat record information processing unit, which is characterized in that it is applied to client, including:
Chat record document determining unit, for determining chat record file to be analyzed;
Analysis request submits unit, for according to the chat record file to be analyzed, generating analysis request, and be submitted to clothes It is engaged in end, being given birth to by the server-side and determining chat record content to be analyzed and preset classifying rules information, utilize described point Rule-like information determines the problem classification belonging to the conversation content entry, is carried out to the conversation content entry of each problem category Statistics, and statistical result is provided.
CN201710031872.8A 2017-01-17 2017-01-17 Chat information processing method and device Active CN108335110B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710031872.8A CN108335110B (en) 2017-01-17 2017-01-17 Chat information processing method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710031872.8A CN108335110B (en) 2017-01-17 2017-01-17 Chat information processing method and device

Publications (2)

Publication Number Publication Date
CN108335110A true CN108335110A (en) 2018-07-27
CN108335110B CN108335110B (en) 2022-04-12

Family

ID=62922640

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710031872.8A Active CN108335110B (en) 2017-01-17 2017-01-17 Chat information processing method and device

Country Status (1)

Country Link
CN (1) CN108335110B (en)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109189902A (en) * 2018-08-09 2019-01-11 珠海格力电器股份有限公司 The method and device of automatic-answering back device consulting
CN109214854A (en) * 2018-07-10 2019-01-15 淘度客电子商务股份有限公司 A kind of self-help service for user method applied to electric business platform
CN109241256A (en) * 2018-08-20 2019-01-18 百度在线网络技术(北京)有限公司 Dialog process method, apparatus, computer equipment and readable storage medium storing program for executing
CN109858021A (en) * 2019-01-02 2019-06-07 平安科技(深圳)有限公司 Traffic issues statistical method, device, computer equipment and its storage medium
CN109903134A (en) * 2019-02-26 2019-06-18 北京多点在线科技有限公司 A kind of order automatic generation method and system, storage medium
CN110427473A (en) * 2019-08-02 2019-11-08 泰康保险集团股份有限公司 Data processing method, device, equipment and storage medium
CN111062728A (en) * 2018-10-17 2020-04-24 阿里巴巴集团控股有限公司 Queuing optimization method and device for manual online consultation
CN111192082A (en) * 2019-12-26 2020-05-22 广东美的白色家电技术创新中心有限公司 Product selling point analysis method, terminal equipment and computer readable storage medium
CN111698143A (en) * 2019-03-14 2020-09-22 阿里巴巴集团控股有限公司 Information processing method, information display method and device
CN112434197A (en) * 2021-01-27 2021-03-02 博智安全科技股份有限公司 Reverse extraction method, device, equipment and storage medium of text content
CN114706969A (en) * 2022-05-31 2022-07-05 深圳追一科技有限公司 Attention content acquisition method and device, electronic equipment and storage medium

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1770798A (en) * 2004-10-28 2006-05-10 中兴通讯股份有限公司 Method for traffic data gathering and analysis statistic
CN101178721A (en) * 2007-10-12 2008-05-14 北京拓尔思信息技术有限公司 Method for classifying and managing useful poser information in forum
US20090248671A1 (en) * 2008-03-28 2009-10-01 Daisuke Maruyama Information classification system, information processing apparatus, information classification method and program
CN103167172A (en) * 2013-02-08 2013-06-19 广州三星通信技术研究有限公司 Integration method and system for variety of chat records
CN103873583A (en) * 2014-03-24 2014-06-18 北京聚思信息咨询有限公司 Method and system for analyzing behaviors of internet users based on cloud platform
CN104281607A (en) * 2013-07-08 2015-01-14 上海锐英软件技术有限公司 Microblog hot topic analyzing method
CN104361003A (en) * 2014-10-10 2015-02-18 金硕澳门离岸商业服务有限公司 Method and device for classified displaying of chat records
CN104731895A (en) * 2015-03-18 2015-06-24 北京京东尚科信息技术有限公司 Auto-answer method and device
CN105740382A (en) * 2016-01-27 2016-07-06 中山大学 Aspect classification method for short comment texts
CN106095972A (en) * 2016-06-17 2016-11-09 联动优势科技有限公司 A kind of information classification approach and device

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1770798A (en) * 2004-10-28 2006-05-10 中兴通讯股份有限公司 Method for traffic data gathering and analysis statistic
CN101178721A (en) * 2007-10-12 2008-05-14 北京拓尔思信息技术有限公司 Method for classifying and managing useful poser information in forum
US20090248671A1 (en) * 2008-03-28 2009-10-01 Daisuke Maruyama Information classification system, information processing apparatus, information classification method and program
CN103167172A (en) * 2013-02-08 2013-06-19 广州三星通信技术研究有限公司 Integration method and system for variety of chat records
CN104281607A (en) * 2013-07-08 2015-01-14 上海锐英软件技术有限公司 Microblog hot topic analyzing method
CN103873583A (en) * 2014-03-24 2014-06-18 北京聚思信息咨询有限公司 Method and system for analyzing behaviors of internet users based on cloud platform
CN104361003A (en) * 2014-10-10 2015-02-18 金硕澳门离岸商业服务有限公司 Method and device for classified displaying of chat records
CN104731895A (en) * 2015-03-18 2015-06-24 北京京东尚科信息技术有限公司 Auto-answer method and device
CN105740382A (en) * 2016-01-27 2016-07-06 中山大学 Aspect classification method for short comment texts
CN106095972A (en) * 2016-06-17 2016-11-09 联动优势科技有限公司 A kind of information classification approach and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
上海宸特电子科技有限公司: "旺旺聊天记录分析", 《旺旺聊天记录分析 *

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109214854A (en) * 2018-07-10 2019-01-15 淘度客电子商务股份有限公司 A kind of self-help service for user method applied to electric business platform
CN109189902A (en) * 2018-08-09 2019-01-11 珠海格力电器股份有限公司 The method and device of automatic-answering back device consulting
CN109241256B (en) * 2018-08-20 2022-09-27 百度在线网络技术(北京)有限公司 Dialogue processing method and device, computer equipment and readable storage medium
CN109241256A (en) * 2018-08-20 2019-01-18 百度在线网络技术(北京)有限公司 Dialog process method, apparatus, computer equipment and readable storage medium storing program for executing
CN111062728A (en) * 2018-10-17 2020-04-24 阿里巴巴集团控股有限公司 Queuing optimization method and device for manual online consultation
CN111062728B (en) * 2018-10-17 2023-06-30 阿里巴巴集团控股有限公司 Queuing optimization method and device for manual online consultation
CN109858021A (en) * 2019-01-02 2019-06-07 平安科技(深圳)有限公司 Traffic issues statistical method, device, computer equipment and its storage medium
CN109858021B (en) * 2019-01-02 2023-11-14 平安科技(深圳)有限公司 Service problem statistics method, device, computer equipment and storage medium thereof
CN109903134A (en) * 2019-02-26 2019-06-18 北京多点在线科技有限公司 A kind of order automatic generation method and system, storage medium
CN111698143A (en) * 2019-03-14 2020-09-22 阿里巴巴集团控股有限公司 Information processing method, information display method and device
CN111698143B (en) * 2019-03-14 2022-12-16 阿里巴巴集团控股有限公司 Information processing method, information display method and device
CN110427473A (en) * 2019-08-02 2019-11-08 泰康保险集团股份有限公司 Data processing method, device, equipment and storage medium
CN111192082A (en) * 2019-12-26 2020-05-22 广东美的白色家电技术创新中心有限公司 Product selling point analysis method, terminal equipment and computer readable storage medium
CN111192082B (en) * 2019-12-26 2024-03-26 广东美的白色家电技术创新中心有限公司 Product selling point analysis method, terminal equipment and computer readable storage medium
CN112434197A (en) * 2021-01-27 2021-03-02 博智安全科技股份有限公司 Reverse extraction method, device, equipment and storage medium of text content
CN114706969A (en) * 2022-05-31 2022-07-05 深圳追一科技有限公司 Attention content acquisition method and device, electronic equipment and storage medium

Also Published As

Publication number Publication date
CN108335110B (en) 2022-04-12

Similar Documents

Publication Publication Date Title
CN108335110A (en) Chat message processing method and processing device
US20200258091A1 (en) System and method for providing a social customer care system
US20110125550A1 (en) Method for determining customer value and potential from social media and other public data sources
CN107430618A (en) Realize the system and method interacted with host computer device progress user speech
AU2012207503A1 (en) Method and system of selecting responders
CN108228559A (en) A kind of human-computer interaction realization method and system for customer service
CN107464143A (en) Network commodity marketing system and method
CN101276450A (en) Information processing device, information processing method and information processing program
CN110276639A (en) A kind of network marketing system and method
Chen et al. A study of cross-border E-commerce research trends: Based on knowledge mapping and literature analysis
Effah The rise and fall of a dot-com pioneer in a developing country
JP6908463B2 (en) Management server, management method and program
CN114254161A (en) Graph database-based social network platform user behavior analysis method and system
CN109726978A (en) A kind of creation service platform
CN108830610A (en) A kind of contract automatic generation method based on online sale, social software
CN108280727A (en) A kind of differential pin unified platform system
CN107273384A (en) The determination method and apparatus of crowd's attribute
Nurhayati et al. The importance of website in business promotion
CN110033326A (en) A kind of enterprise network marketing platform
Yang et al. Institutional Entrepreneurship and Acquiring Legitimacy of Social Commerce Platform
TW201322159A (en) Online real-time question answering system
Wu et al. A Social Media Based Profiling Approach for Potential Churning Customers: An Example for Telecom Industry
TWI616761B (en) Information matching method and system applied to e-commerce website
TW201243755A (en) WOM monitoring system
CN117743554A (en) Community operation method and system based on AI large language model

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
REG Reference to a national code

Ref country code: HK

Ref legal event code: DE

Ref document number: 1258190

Country of ref document: HK

GR01 Patent grant
GR01 Patent grant