CN110147980A

CN110147980A - Worksheet method and device

Info

Publication number: CN110147980A
Application number: CN201910266422.6A
Authority: CN
Inventors: 魏世康; 陈晨; 杨宇; 田正中; 兰杰; 吴洋; 刘泉
Original assignee: Koubei Shanghai Information Technology Co Ltd
Current assignee: Koubei Shanghai Information Technology Co Ltd
Priority date: 2019-04-03
Filing date: 2019-04-03
Publication date: 2019-08-20

Abstract

This application discloses a kind of worksheet methods, comprising: obtains problem log data and work order；Work order identification information is obtained from the first problem record data in problem log data, establishes the first corresponding relationship between the first work order that first problem records data and work order identification information is identified；Object entity mark is obtained from the Second Problem record data in problem log data, object entity mark is obtained from the second work order in work order, establishes the second corresponding relationship having between the Second Problem record data of same object entity identifier and the second work order；Obtain the characteristic information of the third problem log data in problem log data, obtain the characteristic information of the third work order in work order, according to the characteristic information of third problem log data, the characteristic information of third work order and corresponding relationship prediction model, the third corresponding relationship between third problem log data and third work order is established.Solves the problems, such as the problem of improving the efficiency of processing.

Description

Worksheet method and device

Technical field

This application involves technical field of information processing, and in particular to worksheet method, apparatus and electronic equipment.

Background technique

WorkForm System is the platform recorded, handled, tracked for the process demand of problem or event, is obtained in every profession and trade To extensive use, business scope is ensured in particular for O&M.For example, ensureing business scope in O&M, consumer, quotient are received After the problem of family or service monitoring system are found, there are efficacious prescriptions in customer service, guarantee, engineering department by work order stream of WorkForm System etc. Formula carries out quick response and processing for problem.So-called work order stream refers to the problem of establishing around WorkForm System processing and work order The system of circulation.The data for the different phase being related in work order stream are referred to as the upstream or downstream of work order stream, work order stream Upstream refer to customer service and interaction that user carries out through various channels and information the problem of output, take the form of problem log Data, also referred to as short biography；The downstream of work order stream is work order.Problem log data and work order are associated referred to as upstream and downstream pass Connection.

The upstream and downstream of work order stream is isolated completely at present, it is difficult to is associated, be led to issue handling inefficiency and number It is difficult according to analysis.Generating upstream and downstream the reason of isolating has three aspects: first is that, at present due to WorkForm System the problem of record data arrive Exist in the creation and the entire link of processing of work order and record nonstandard phenomenon, causes work order data to present semi-structured, for example, Solid data involved in work order is recorded in colloquial work order content, it is difficult to carry out effective data analysis.Second is that asking May also not have effective field in topic record data can be associated work order.Third is that problem log data and work order Type disunity causes to be difficult to carry out the practical problem unified Type division.Due to problem log data and work order In all there are problems that respective field can be to classifying, but the classification standard disunity of two kinds of data, and may non-format Change, therefore leads to the classification numerous and complicated of classification and lack unified measurement standard.

Problem log data and work order are isolated, and are made it difficult to be associated problem analysis, be will lead to part important information Missing, finally makes issue handling low efficiency.Therefore, how problem log data and work order to be associated, thus at raising The efficiency of reason problem is problem to be solved.

Summary of the invention

The application provides a kind of worksheet method, and solve problem log data and work order is associated with asking for low efficiency Topic.

The application provides a kind of worksheet method, comprising:

Obtain problem log data and work order；

Work order identification information is obtained from the first problem record data in described problem record data, establishes described first The first corresponding relationship between the first work order that problem log data and the work order identification information are identified；

Object entity mark is obtained from the Second Problem record data in described problem record data, from the work order The second work order in obtain object entity mark, establish have same object entity identifier Second Problem record data and second The second corresponding relationship between work order；

The characteristic information for obtaining the third problem log data in described problem record data, obtains the in the work order The characteristic information of three work orders, according to the characteristic information of the characteristic information of the third problem log data, the third work order with And corresponding relationship prediction model, establish the third corresponding relationship between the third problem log data and the third work order；

Wherein, the corresponding relationship prediction model is the characteristic information that data are recorded according to the first problem, described the The model that the characteristic information of one work order and first corresponding relationship training obtain, and/or, the corresponding relationship prediction model It is that the characteristic information of data, the characteristic information of second work order and the second corresponding pass are recorded according to the Second Problem The model that system's training obtains.

Optionally, the acquisition problem log data and work order, comprising:

Obtain primal problem record data and original work order；Data are recorded to the primal problem and original work order counts According to cleaning, described problem record data and the work order are obtained.

Optionally, further includes: the first work order that the work order identification information is identified is searched from the work order.

Optionally, further include following at least one operations:

It is identified as index with the object entity obtained from Second Problem record data, from second work order In the object entity mark of acquisition, lookup and the object entity mark obtained from Second Problem record data are identical right As entity identifier；

It is identified as index with the object entity obtained from second work order, from Second Problem record data In the object entity mark of acquisition, searches and identify identical object entity mark with the object entity obtained from second work order Know.

Optionally, further include following at least one operations:

With the first problem record characteristic information of data, the characteristic information of first work order and first pair described It should be related to for sample data, training according to the characteristic information of problem log data and the characteristic information forecasting problem of work order for remembering Record the corresponding relationship prediction model of the corresponding relationship of data and work order；

With the Second Problem record characteristic information of data, the characteristic information of second work order and second pair described It should be related to for sample data, training according to the characteristic information of problem log data and the characteristic information forecasting problem of work order for remembering Record the corresponding relationship prediction model of the corresponding relationship of data and work order.

Optionally, the corresponding relationship prediction model can export: the keyword and problem log data of the third work order Keywords matching probability, and/or, the probability of the categorical match of the classification and problem log data of the third work order.

Optionally, the characteristic information of the first problem record data is the keyword that the first problem records data And/or classification, the characteristic information of first work order are the keyword and/or classification of first work order, the Second Problem The characteristic information for recording data is the keyword and/or classification of Second Problem record data, the feature of second work order Information is the keyword and/or classification of second work order, and the characteristic information of the third problem log data is the third The keyword and/or classification of problem log data, the characteristic information of the third work order be the third work order keyword and/ Or classification.

Optionally, further includes: obtain problem information；Obtain the particular problem record data that record has the problem information； According at least one of first corresponding relationship, second corresponding relationship, the third corresponding relationship corresponding relationship, obtain Obtain specific work order corresponding with particular problem record data；It is obtained according to the specific work order for described problem information Processing progress information or processing result.

Optionally, the work order is problem work order or task work order.

The application also provides a kind of worksheet method, comprising:

Obtain problem log data；

Work order identification information is obtained from described problem record data；

According to the work order identification information, the work that described problem record data are identified with the work order identification information is established Corresponding relationship between list.

It is optionally, described to obtain work order identification information from described problem record data, comprising:

Obtain the predetermined create-rule of work order identification information；

According to the predetermined create-rule, the work order identification information is obtained from described problem record data.

The application also provides a kind of worksheet method, comprising:

Obtain problem log data and work order；

Object entity mark is obtained from described problem record data, and obtains object entity mark from the work order；

Establish the corresponding relationship having between the problem of same object entity identifier record data and work order.

Optionally, described to obtain object entity mark from described problem record data, and obtain from the work order pair As entity identifier, comprising:

Obtain the predetermined create-rule of object entity mark；According to the predetermined create-rule that object entity identifies, from described Object entity mark is obtained in problem log data, and object entity mark is obtained from the work order.

Optionally, the foundation has the corresponding pass between the problem of same object entity identifier record data and work order System, comprising:

Object entity is obtained according to the object entity mark obtained from described problem record data and from the work order Mark, determining has the problem of same object entity identifier record data and work order；

Corresponding relationship is established between the problem of same object entity identifier record data and work order having.

Optionally, the object entity mark and obtained from the work order that the basis is obtained from described problem record data Object entity mark is obtained, determining has the problem of same object entity identifier record data and work order, comprising: asks using from described Object entity mark the first object entity logo collection of construction obtained in topic record data, using being obtained from the work order pair As entity identifier constructs the second object entity logo collection；Determine the first object entity logo collection and second object Registration between set of entity identities；The first object entity set that the registration is met registration threshold condition is corresponding The problem of record data and the corresponding work order of the second object entity set, as having the problem of same object entity identifier Record data and work order.

Optionally, described that there is the corresponding pass of foundation between the problem of same object entity identifier record data and work order System, comprising: obtain the generation time of described problem record data；Obtain the generation time of the work order；Remembered according to described problem The generation time for recording the generation time and the work order of data determines the generation time of described problem record data and the work order Interval；It is recorded the problem of time interval meets time interval threshold condition between data and work order and establishes corresponding relationship.

Optionally, the object entity mark and obtained from the work order that the basis is obtained from described problem record data Object entity mark is obtained, determining has the problem of same object entity identifier record data and work order, comprising: according to default association Priority, determining has the problem of same object entity identifier record data and work order.

Optionally, described according to default associated priority, determine that the problem of having same object entity identifier records data And work order, comprising:

Determine the similarity between the text of described problem record data and the text of the work order；

Work order is ranked up according to similarity, successively whether decision problem record data are real with same object with work order Body mark, acquisition have the problem of same object entity identifier record data and work order.

Obtain the information of the founder of creation described problem record data；Obtain the work order created by same creator；Sentence Whether disconnected described problem record data with the work order have same object entity identifier, and obtaining has same object entity identifier The problem of record data and work order.

Optionally, further includes: if described problem record data and the work order do not have same object entity identifier, The work order of non-equal founder's creation is obtained, determining has the problem of same object entity identifier record data and work order.

The application also provides a kind of worksheet method, comprising:

Obtain problem log data and work order；

The characteristic information for obtaining described problem record data, obtains the characteristic information of the work order, is remembered according to described problem The characteristic information of data, the characteristic information and corresponding relationship prediction model of the work order are recorded, described problem record data are established With the corresponding relationship between the work order；

Wherein, the corresponding relationship prediction model is the spy that data are recorded according to the problem of establishing corresponding relationship with work order Reference breath, established with problem log data corresponding relationship work order characteristic information and established corresponding relationship it is trained The model arrived.

Optionally, the characteristic information is classification；Correspondingly, the method, further includes: predicted according to the corresponding relationship Model determines the probability of the classification of work order and the categorical match of problem log data；Matched probability is met into the first probability threshold The work order of value condition is as the work order with described problem record data with corresponding relationship.

Optionally, the characteristic information is keyword；Correspondingly, the method, further includes: pre- according to the corresponding relationship Model is surveyed, determines the probability of the keyword of work order and the Keywords matching of problem log data；Matched probability is met second The work order of probability threshold value condition, as the work order with described problem record data with corresponding relationship.

The application also provides a kind of work order flow data processing device, comprising:

Problem log and work order acquiring unit, for obtaining problem log data and work order；

Work order identifies associative cell, for obtaining work order from the first problem record data in described problem record data Identification information establishes first between the first work order that the first problem records data and the work order identification information is identified Corresponding relationship；

Object entity identifies associative cell, for obtaining from the Second Problem record data in described problem record data Object entity mark obtains object entity mark from the second work order in the work order, and establishing has same object entity mark The second corresponding relationship between the Second Problem record data of knowledge and the second work order；

Characteristic information associative cell, for obtaining the feature letter of the third problem log data in described problem record data Breath, obtains the characteristic information of the third work order in the work order, according to the characteristic information of the third problem log data, described The characteristic information and corresponding relationship prediction model of third work order, establish the third problem log data and the third work order Between third corresponding relationship；

Problem log obtaining unit, for obtaining problem log data；

Work order identifies obtaining unit, for obtaining work order identification information from described problem record data；

Work order identifies associative cell, for according to the work order identification information, establish described problem record data with it is described The corresponding relationship between work order that work order identification information is identified.

Problem log and work order obtaining unit, for obtaining problem log data and work order；

Object entity identifies obtaining unit, for obtaining object entity mark from described problem record data, and from institute It states and obtains object entity mark in work order；

Object entity identifies associative cell, has the problem of same object entity identifier record data and work order for establishing Between corresponding relationship.

Characteristic information associative cell obtains the spy of the work order for obtaining the characteristic information of described problem record data Reference breath records the characteristic information of data, the characteristic information and corresponding relationship prediction model of the work order according to described problem, Establish the corresponding relationship between described problem record data and the work order；

Compared with prior art, the application has the following advantages:

A kind of worksheet method provided by the present application records data and work order identification information institute by establishing first problem The first corresponding relationship between first work order of mark；Establish have same object entity identifier Second Problem record data and The second corresponding relationship between second work order；According to the characteristic information of third problem log data, the characteristic information of third work order And corresponding relationship prediction model, the third established between the third problem log data and the third work order is corresponding to close System；It, can be according to object so as to which first problem record data and the first work order are associated according to work order identification information Second Problem is recorded data and the second work order and associated by entity identifier, can be according to characteristic information by third problem log number It associates according to third work order, solves the problems, such as the association low efficiency of problem log data and work order.

The application also provides a kind of worksheet method, by obtaining work order identification information, root from problem log data According to the work order identification information, pair between the work order that described problem records data and the work order identification information is identified is established It should be related to, so as to which problem log data associate with work order according to work order identification information, solve comprising work order mark The problem of knowing information records the related question of data and work order.

The application also provides a kind of worksheet method, by obtaining object entity mark from problem log data, and Object entity mark is obtained from the work order, foundation has between the problem of same object entity identifier record data and work order Corresponding relationship, so as to according to object entity mark problem log data are associated with work order, solve include pair The related question of data and work order is recorded as the problem of entity identifier.

The application also provides a kind of worksheet method, by obtaining the characteristic information of problem log data, obtains work order Characteristic information, the characteristic informations of data, the characteristic information of the work order and corresponding relationship are recorded according to described problem and predicted Model, establish described problem record data and the work order between corresponding relationship, so as to according to characteristic information to problem Record data and work order are associated, and are solved the problems, such as that problem log data and work order are associated, are asked to improve processing The efficiency of topic.

Detailed description of the invention

Fig. 1 is the work order flow diagram of the guarantee business of certain existing WorkForm System；

Fig. 2 is existing problem log data and the respective content schematic diagram of work order；

Fig. 3 is a kind of process flow diagram for worksheet method that the application first embodiment provides；

Fig. 4 is to identify related question record data and work order flow chart according to work order in the application first embodiment；

Fig. 5 is to identify related question record data and work order flow chart according to object entity in the application first embodiment；

Fig. 6 is the generating process schematic diagram of classification corresponding relationship model in the application first embodiment；

Fig. 7 is the generating process schematic diagram of keyword relational model in the application first embodiment；

Fig. 8 is the process signal for establishing third corresponding relationship in the application first embodiment using corresponding relationship prediction model Figure；

Fig. 9 is the various dimensions third correspondence diagram established in the application first embodiment；

Figure 10 is the flow chart for the actual deployment scheme that the application first embodiment provides；

Figure 11 is a kind of process flow diagram for worksheet method that the application second embodiment provides；

Figure 12 is a kind of process flow diagram for worksheet method that the application 3rd embodiment provides；

Figure 13 is a kind of process flow diagram for worksheet method that the application fourth embodiment provides；

Figure 14 is a kind of work sheet processing device schematic diagram that the 5th embodiment of the application provides；

Figure 15 is a kind of work sheet processing device schematic diagram that the application sixth embodiment provides；

Figure 16 is a kind of work sheet processing device schematic diagram that the 7th embodiment of the application provides；

Figure 17 is a kind of work sheet processing device schematic diagram that the 8th embodiment of the application provides；

Figure 18 is a kind of electronic equipment schematic diagram provided by the present application.

Specific embodiment

Many details are explained in the following description in order to fully understand the application.But the application can be with Much it is different from other way described herein to implement, those skilled in the art can be without prejudice to the application intension the case where Under do similar popularization, therefore the application is not limited by following public specific implementation.

The application provides worksheet method, apparatus and electronic equipment.It carries out one by one in the following embodiments detailed Explanation.

The application first embodiment provides a kind of worksheet method, can ensure that business scope deployment is implemented in O&M, It can be used for other business scopes that task or event are handled in a manner of work order stream.In the application, work order can be record and ask Topic or work order the problem of event, are also possible to the task work order of task or work plan that record issues.With O&M in the application It is illustrated for the worksheet of guarantee business scope.

For ease of understanding, work order process is introduced so that O&M ensures certain WorkForm System of business as an example first.It please join Examine Fig. 1, Fig. 1 shows the work order process that business is ensured under the line of certain WorkForm System, includes: in figure

After C-terminal (i.e. consumer end or terminal user end) 101 goes wrong, user can seek advice from the intelligent answer of certain platform System can go to the process flow of phone customer service 103 if problem is unresolved, and it is (interactive that user enters intelligence IVR after making a phone call Voice answer-back), if problem is still unresolved, jump to artificial customer service；

After the end B (i.e. enterprise Clien or merchant end) 102 goes wrong, trade company can seek advice from intelligence questions system or straight Making and receiving calls seeks advice from artificial customer service, goes to the process flow of phone customer service 103；

In the process flow of phone customer service 103, a line customer service receive customer complaint phone feedback customer issue, to problem into Row record generates problem log data (or being short biography), and solves customer issue using existing permission；If customer issue is Knotty problem, then a line customer service is submitted work order and is handled, if the one line customer service of problem is difficult to independently solve, work order is turned to send to Upgrade service ensures 104 process flow；

Upgrade service ensures in 104 process flow that two wires support personnel is further processed 103 turns of works sent of phone customer service It is single, if two wires support personnel fails to handle the customer service problem in the work order, work order is turned to the processing stream for sending to technical guarantee 105 Journey is handled by three line technology support personnels；

In the process flow of technical guarantee 105, after three line technology support personnels solve customer issue, work order is by auto-returned To work order, submitter is handled；

Work order submitter (a line customer service) terminates work order and the client that wires back；In addition, work order submitter can also in practical application To be that the personnel that other can create work order are directed to the work order of problem creation, for example, the interaction problems of peripheral product discovery can also be with Propose work order.

It can be seen that problem log data from above-mentioned scene and work order information be the different phase generation in work order process, It is lack of standardization due to recording, and the multiplicity of data involved in work order process, lead to problem log data and work order data may be in It is now semi-structured data, such as may colloquial style situations such as describing solid data.So-called semi-structured data, refers to according to number According to structuring degree, data between structural data and unstructured data.Structural data, which refers to have, to be defined Corresponding relationship, can be indicated and store using relevant database, show as the data of two dimensional form, such as name, year Age etc.；Unstructured data then without fixed structure, such as picture, video etc., needs entirety to be stored and analyzed.Between structure Change the semi-structured data between data and unstructured data, not only comprising can be with the data of formatted storage, but also including can be with The self-description information for being difficult to structuring for layering.It is semi-structured so that problem log data and work order are difficult to associate. Referring to FIG. 2, the content that problem log data and work order include is set forth in the figure, short biography information 201 is a problem note The information for including in record data, work order information 202 are a kind of information that work order includes, short biography information 201 and work order information 202 Each entry lacks unified format and measurement standard.

The problem of work order process, records data and work order isolates phenomenon, will lead to worksheet inefficiency.This The worksheet method provided is provided, the related question of problem log data and work order is solved.

The worksheet method that the application first embodiment provides is illustrated below in conjunction with Fig. 3 to Figure 10.

Worksheet method shown in Fig. 3, comprising: step S301 to step S304.

Step S301 obtains problem log data and work order.

In the present embodiment, the acquisition problem log data and work order, comprising:

Obtain primal problem record data and original work order；

Data are recorded to the primal problem and original work order carries out data cleansing, obtain described problem record data and institute State work order.

The data cleansing, refers to the mistake of identification primal problem record data and original work order, including checks data one Cause property, the cleaning operation such as invalid value and missing values.For example, obtaining primal problem record from online database or online log The primal problem is recorded data and original work order carries out data cleansing, the data after cleaning are remembered by data and original work order It records in off-line data table.For another example, problem log data and work order data are filtered, retain related question record data and The valid data such as entity ID, work order ID, keyword that work order uses.For another example, it is identified from partly-structured data field With extract entity ID, the valid data such as keyword.

In practical application, problem log data and work order are the relationships of multi-to-multi, i.e. a problem log data corresponding one Item or a plurality of work order, the corresponding one or more problem log data of a work order.When problem is more complicated, need to disassemble into more A different types of work order carries out separating processing, then will appear the situation that a problem log data correspond to a plurality of work order.For example, One line customer service receives food and drink trade company A phone feedback and the infull problem of vegetable occurs, and POS machine operation also goes wrong, that One line customer service proposes two work orders, and a worksheet of ordering is ordered problem, an ISV (Independent Software Verdor, independent software vendor) worksheet POS problem.For another example, same customer service receives multiple phone feedbacks and asks in one day Topic, the meeting a plurality of problem log data of corresponding record propose a work order if these problems are substantially identical, thus will appear The situation of the corresponding work order of a plurality of problem log data.For another example, same trade company, multiple telephone questionnaire Resolving probiems progress, visitor Clothes can record data correlation to same work order the problem of telephony recording that correspondence is each.These are gradually established in the next steps to ask Corresponding relationship between topic record data and work order.

Step S302 obtains work order identification information from the first problem record data in described problem record data, builds Found the first corresponding relationship between the first work order that the first problem records data and the work order identification information is identified.

The work order identification information is used for unique identification work order.It is lack of standardization and semi-structured due to problem log data, Work order identification information manual record during creating or submitting work order by contact staff, therefore some problem log data packets Identification information containing work order, some problem log data do not include work order identification information.The problem of including work order identification information, records Data can be identified by work order and be directly linked to work order；And the problem of not including work order identification information, records data then no and work Single associated clear corresponding relationship, it is difficult to be associated with work order, need to be further processed in subsequent step.

The first problem records data to record data the problem of including work order identification information.It is identified comprising same work order The first problem record data of information can be one or more problem log data.It may include in one problem log data One or more work order identification informations, therefore first work order mutually should be one or more work orders.

In the present embodiment, need to identify and extract from problem log data work order identification information, such as from problem log Work order identification information is identified in the content of text of data, to obtain first problem record data.Specifically, passing through following processing Obtain work order identification information:

After successfully extracting work order identification information, the work order identification information is identified is searched from the work order One work order.To which first problem record data be associated with the first work order according to work order identification information.According to certain work order system The experimental data of system, content of text the problem of 60% in record data, such as the corresponding content of memo field, contain work order Mark, therefore work order can be associated with by record data the problem of processing 60% of this step.Referring to FIG. 4, being shown in figure Problem log data correspond to the process flow of work order according to work order mark.Short biography information 401 is problem log data in figure, from Work order ID, i.e. work order identification information are extracted in short biography content of text in short biography information, and unique work is corresponded to according to work order ID List, i.e. work order information 402.

Step S303 obtains object entity mark from the Second Problem record data in described problem record data, from Object entity mark is obtained in the second work order in the work order, establishing, there is the Second Problem of same object entity identifier to record The second corresponding relationship between data and the second work order.

The object entity mark, refers to the identification information for being used to indicate object entity.For example, the object entity mark Know, can be following any identification informations: user subject mark, order entity mark, preferential entity identifier.For example, user is real Body mark can be trade company ID；Order entity mark can be order ID or vegetable ID；Preferential entity identifier can be movable ID, Certificate ID.

In the present embodiment, needs to identify and extract from problem log data and work order respectively object entity mark, extract Recording data to the problem of object entity mark is that Second Problem records data, and the work order for extracting object entity mark is second Work order.Second Problem record data comprising same object entity identifier can be one or more problem log data；Include Second work order of same object entity identifier can be one or more work order.

Specifically, establishing the Second Problem record data and the second work with same object entity identifier using following processing The second corresponding relationship between list:

Obtain the predetermined create-rule of object entity mark；

According to the predetermined create-rule that object entity identifies, object entity mark is obtained from described problem record data, Object entity mark is obtained from the work order；

Establish the corresponding relationship having between the Second Problem record data of same object entity identifier and the second work order.

Wherein, described to establish pair having between the Second Problem record data of same object entity identifier and the second work order In the processing that should be related to, including following at least one processing modes:

Mode one, with from the Second Problem record data object entity that obtains be identified as index, from described the In the object entity mark obtained in two work orders, searches and identified with the object entity obtained from Second Problem record data Identical object entity mark；

Mode two is identified as index with the object entity obtained from second work order, remembers from the Second Problem In the object entity mark obtained in record data, lookup and the object entity mark obtained from second work order are identical right As entity identifier.

In practical application, since same target entity information is likely to appear in a plurality of problem log data or a plurality of work Dan Zhong, therefore object entity mark cannot be used for unique identification problem log data, it can not unique identification work order.In order to improve Second Problem records the accuracy of data and the second work order corresponding relationship, the specific embodiment provided in the present embodiment In, further, being determined using following processing has the problem of same object entity identifier record data and work order:

Using obtained from problem log data object entity mark construction the first object entity logo collection, using from Object entity mark the second object entity logo collection of construction is obtained in work order；

Determine the registration between the first object entity logo collection and the second object entity logo collection；

The registration is met to the corresponding problem log data of the first object entity set of registration threshold condition, with And the second corresponding work order of object entity set, as having the problem of same object entity identifier record data and work order.

In the another embodiment provided in the present embodiment, by following processing with same object entity mark The problem of knowledge, records and establishes corresponding relationship between data and work order:

Obtain the generation time of described problem record data；

Obtain the generation time of the work order；

The generation time that the generation time and the work order of data is recorded according to described problem determines that described problem records number According to the generation time interval with the work order；

It is recorded the problem of time interval meets time interval threshold condition between data and work order and establishes corresponding relationship.

In practical application, can also be determined in conjunction with the registration and the time interval problem log data and work order it Between corresponding relationship.For example, the generation time interval of problem log data and work order is shorter, and the object entity mark being overlapped More, the degree of association of problem log data and work order is higher, and corresponding relationship is more accurate.

In the another embodiment provided in the present embodiment, according to default associated priority, determine with identical The problem of object entity identifies record data and work order, specifically, when establishing the corresponding relationship of problem log data and work order, The accuracy of corresponding relationship is improved using other feature priority match such as founder or text similarity.For example, according to problem The similarity between the content of text of the content of text and work order in data is recorded, is preferentially asked to reaching similarity threshold condition Topic record data and work order are associated, and specifically include following processing:

For another example, preferentially to same creator the problem of record data and work order be associated, specifically include following processing:

Obtain the information of the founder of creation described problem record data；

Obtain the work order created by same creator；

Judge described problem record data whether with the work order have same object entity identifier, obtain have it is identical right Data and work order are recorded as the problem of entity identifier；

If described problem records data and the work order does not have same object entity identifier, non-equal creation is obtained The work order of person's creation, determining has the problem of same object entity identifier record data and work order.

By this step process, Second Problem is recorded by data by object entity mark and is associated with the second work order. According to the experimental data of certain WorkForm System, work order on record data correlation, further passes through the problem of 60% in step S302 Problem log data can be increased to 80% by 60% with being associated with for work order by this step.Referring to FIG. 5, showing in figure by right As problem log data are corresponded to the process flow of work order by entity identifier, short biography information 501 is problem log data in figure, from The object entity mark (entity ID shown in figure) extracted in short biography content of text in short biography information includes: user subject ID, order entity ID, preferential entity ID；Object entity mark (the figure extracted from the work order content of text of work order information 502 Shown in entity ID) include: user subject ID, order entity ID；And in short biography information 501 and in work order information 502 User subject ID is identical with order entity ID, then establishes the corresponding relationship of short biography information 501 and work order information 502.

Step S304 obtains the characteristic information of the third problem log data in described problem record data, described in acquisition The characteristic information of third work order in work order, according to the characteristic information of the third problem log data, the third work order Characteristic information and corresponding relationship prediction model establish the third between the third problem log data and the third work order Corresponding relationship；

This step establishes the third between third problem log data and third work order according to same or similar characteristic information Corresponding relationship.In the present embodiment, the characteristic information can be classification and/or keyword.Correspondingly, the first problem record The characteristic information of data is the keyword and/or classification that the first problem records data, the characteristic information of first work order Characteristic information for the keyword and/or classification of first work order, the Second Problem record data is the Second Problem The keyword and/or classification of data are recorded, the characteristic information of second work order is the keyword and/or class of second work order Not, the characteristic information of the third problem log data is the keyword and/or classification of the third problem log data, described The characteristic information of third work order is the keyword and/or classification of the third work order.

In practical application, after above-mentioned steps are handled, subproblem record data and work order establish corresponding relationship, but It is still to there is subproblem to record the not associated upper work order of data.In a specific embodiment in the embodiment of the present application, in use The problem of establishing corresponding relationship in step record data and the work order training corresponding relationship prediction model are stated, are specifically included down State at least one training method:

Training method one, with the characteristic information of the first problem record characteristic information of data, first work order with And first corresponding relationship is sample data, training according to the characteristic information of problem log data and the feature of work order for believing Cease the corresponding relationship prediction model of the corresponding relationship of forecasting problem record data and work order；

Training method two, with the characteristic information of the Second Problem record characteristic information of data, second work order with And second corresponding relationship is sample data, training according to the characteristic information of problem log data and the feature of work order for believing Cease the corresponding relationship prediction model of the corresponding relationship of forecasting problem record data and work order.

Machine learning algorithm such as TopK and naive Bayesian scheduling algorithm can be used in practical application, training obtains described Corresponding relationship prediction model.For the more application scenarios of problem log data, it is pre- to present embodiments provide a kind of corresponding relationship Survey model, for problem log data to be corresponded to work order classification, corresponding relationship prediction model under the scene, comprising having been established The characteristic information and the characteristic information of the problem of corresponding relationship record data (short biography) and work order belong to specific work order classification Conditional probability.For recording data A the problem of needing to be associated with work order classification.The spy extracted from problem log data A Reference breath includes category feature B, keyword feature C, inputs as the corresponding relationship prediction model, calculates category feature B The probability of each work order classification is corresponded to keyword feature C, if the probability distribution obtained and the conditional probability of work order classification a Distribution can match, then establish corresponding relationship between problem log data A and work order type a, thus by problem log data with The related question of work order is converted to classification problem.Similar, the present embodiment also provides another corresponding relationship prediction model, It include to record the characteristic information of data (short biography) and work order the problem of corresponding relationship has had been established and the characteristic information belongs to The conditional probability of problem log data category, for work order to be corresponded to problem log data category.

In a specific embodiment provided in this embodiment, corresponding relationship prediction model is to record number using first problem According to classification, the classification of the first work order and the first corresponding relationship be the obtained classification corresponding relationship model of sample data training, For establishing third corresponding relationship according to the classification of third problem log data and the classification of third work order.Referring to FIG. 6, in figure Show the generating process schematic diagram of classification corresponding relationship model.Short biography classification is the classification extracted from short biography information, work order Classification is the classification extracted from work order, calculates classification according to short biography (i.e. problem log data) and the corresponding relationship of work order Conditional probability, using machine learning algorithm such as TopK and naive Bayesian scheduling algorithm, training obtains classification corresponding relationship model.

In a specific embodiment provided in this embodiment, corresponding relationship prediction model is to record number using first problem According to keyword, the first work order keyword and the second corresponding relationship be the obtained keyword relationship mould of sample data training Type, for establishing third corresponding relationship according to the keyword of third problem log data and the keyword of third work order.It please refers to Fig. 7 shows the generating process schematic diagram of keyword relational model in figure.From short biography information extraction short biography keyword, from work order Information extraction work order keyword, the conditional probability of keyword is calculated according to the corresponding relationship of short biography information and work order information, is adopted With machine learning algorithm such as TopK and naive Bayesian scheduling algorithm, training obtains keyword relational model.

It is available by the corresponding relationship prediction model in the present embodiment: the keyword of the third work order with ask The probability of the Keywords matching of topic record data, and/or, the classification of the third work order and the categorical match of problem log data Probability.Correspondingly, in this step further include: using the probability of categorical match meet the work order of the first probability threshold value condition as with The third problem log data have the third work order of corresponding relationship；And/or the probability of Keywords matching is met second generally The work order of rate threshold condition is as the third work order with the third problem log data with corresponding relationship.Include identical or phase It can be one or more problem log data like the third problem log data of characteristic information, believe comprising same or similar feature The third work order of breath is also possible to one or more work order.

In the present embodiment, the corresponding relationship prediction model that training obtains in an application scenarios includes classification corresponding relationship mould Type and keyword relational model two parts, using corresponding relationship prediction model establish third problem log data and third work order it Between third corresponding relationship.Please refer to Fig. 8 and Fig. 9.Fig. 8, which is shown, establishes third corresponding relationship using corresponding relationship prediction model Process schematic, include: to extract short biography classification and short biography keyword from short biography information 801 in figure, never associated work order letter Breath 802 extracts work order classifications and work order keyword, using short biography classification and work order classification as the input of classification corresponding relationship model, Study obtains the classification corresponding relationship of short biography and work order；Using short biography keyword and work order keyword as keyword relational model Input, study obtain the keyword corresponding relationship of short biography and work order.Fig. 9, which is shown, establishes third corresponding relationship as various dimensions correspondence Association, association latitude include the classification corresponding relationship and content corresponding relationship between short biography information and work order information, wherein in Holding corresponding relationship is the corresponding relationship established according to keyword.

By this step process, third problem log data are associated with third work order by characteristic information.According to The experimental data of certain WorkForm System, by step S302 and step S303 place 80% the problem of record data correlation on work It is single, this 80% data is further used as training sample and trains corresponding relationship prediction model, uses institute in this step State corresponding relationship prediction model can will be not associated on third problem log data and third work order between establish corresponding relationship.Please The flow chart of an actual deployment scheme is shown with reference to Figure 10, Figure 10, includes following processing steps in figure:

S1001 carries out data cleansing to problem log data and work order；

S1002, decision problem, which records, whether there is work order ID in data；If it is, turning S1002-1；If it is not, then turning S1002-2；

S1002-1, work order mark association；Identified by work order, such as work order ID, establish first problem record data and First corresponding relationship of the first work order；

S1002-2, object entity mark association；Second Problem is established by object entity mark and records data and second Second corresponding relationship of work order；

S1003 obtains the corresponding relationship of associated problem log data and work order；Including the first corresponding relationship and second Corresponding relationship；

S1004 carries out data cleansing to class relations and keyword relationship；

S1005 obtains corresponding relationship prediction model；The corresponding relationship prediction model include classification corresponding relationship model and Two parts of keyword relational model；

S1006, related question record data and work order；Not associated problem is established using corresponding relationship prediction model to remember Record the corresponding relationship between data and work order.

In the present embodiment, further, after establishing the corresponding relationship between problem log data and work order, further include Following processing:

Obtain problem information；

Obtain the particular problem record data that record has the problem information；

According at least one of first corresponding relationship, second corresponding relationship, the third corresponding relationship pair It should be related to, obtain specific work order corresponding with particular problem record data；

The processing progress information or processing result for being directed to described problem information are obtained according to the specific work order.To have Help the processing progress of promotion problem.

On that basi of the above embodiments, the application second embodiment provides a kind of worksheet method.

The worksheet method that the application second embodiment provides is illustrated below in conjunction with Figure 11.Due to above-mentioned reality Based on applying example, therefore the corresponding explanation of above-described embodiment is partially referred to accordingly.

Worksheet method shown in Figure 11, comprising: step S1101 to step S1103.

Step S1101 obtains problem log data.

In the present embodiment, the acquisition problem log data, comprising: obtain primal problem and record data；To described original Problem log data carry out data cleansing, obtain described problem record data.

The data cleansing, refers to the mistake of identification primal problem record data and original work order, including checks data one Cause property, the cleaning operation such as invalid value and missing values.For example, obtaining primal problem record from online database or online log Primal problem record data are carried out data cleansing, the data after cleaning are recorded in off-line data table by data.

In practical application, problem log data and work order are the relationships of multi-to-multi, i.e. a problem log data corresponding one Item or a plurality of work order, the corresponding one or more problem log data of a work order.These problems are gradually established in the next steps Record the corresponding relationship between data and work order.

Step S1102 obtains work order identification information from described problem record data.

In the present embodiment, need to identify and extract from problem log data work order identification information, such as from problem log Work order identification information is identified in the content of text of data, specifically includes following processing:

Step S1103 establishes described problem record data and the work order identification information according to the work order identification information The corresponding relationship between work order identified.

In the present embodiment, after successfully extracting work order identification information, the work order mark letter is searched from the work order Identified work order is ceased, so that problem log data associate with work order according to work order identification information.According to certain work order system The experimental data of system, content of text the problem of 60% in record data, such as the corresponding content of memo field, contain work order Mark, therefore work order can be associated with by record data the problem of processing 60% of this step.

On that basi of the above embodiments, the application 3rd embodiment provides a kind of worksheet method.

The worksheet method that the application 3rd embodiment provides is illustrated below in conjunction with Figure 12.Due to above-mentioned reality Based on applying example, therefore the corresponding explanation of above-described embodiment is partially referred to accordingly.

Worksheet method shown in Figure 12, comprising: step S1201 to step S1203.

Step S1201 obtains problem log data and work order.

Obtain primal problem record data and original work order；

The data cleansing, refers to the mistake of identification primal problem record data and original work order, including checks data one Cause property, the cleaning operation such as invalid value and missing values.For example, obtaining primal problem record from online database or online log The primal problem is recorded data and original work order carries out data cleansing, the data after cleaning are remembered by data and original work order It records in off-line data table.

Step S1202 obtains object entity mark from described problem record data, and obtains object from the work order Entity identifier.

In the present embodiment, need to identify and extract from problem log data and work order respectively object entity mark, specifically Including following processing:

Obtain the predetermined create-rule of object entity mark；

According to the predetermined create-rule that object entity identifies, object entity mark is obtained from described problem record data, Object entity mark is obtained from the work order.

Step S1203 establishes the corresponding relationship having between the problem of same object entity identifier record data and work order.

In the present embodiment, the foundation has the correspondence between the problem of same object entity identifier record data and work order Relationship, comprising:

In practical application, since same target entity information is likely to appear in a plurality of problem log data or a plurality of work Dan Zhong, therefore object entity mark cannot be used for unique identification problem log data, it can not unique identification work order.In order to improve The accuracy of problem log data and work order corresponding relationship, in the specific embodiment provided in the present embodiment, under It states processing and determines that there is the problem of same object entity identifier record data and work order:

Using object entity mark the first object entity logo collection of construction obtained from described problem record data, make The second object entity logo collection is constructed with object entity mark is obtained from the work order；

Obtain the generation time of described problem record data；

Obtain the generation time of the work order；

Obtain the work order created by same creator；

On that basi of the above embodiments, the application fourth embodiment provides a kind of worksheet method.

The worksheet method that the application fourth embodiment provides is illustrated below in conjunction with Figure 13.Due to above-mentioned reality Based on applying example, therefore the corresponding explanation of above-described embodiment is partially referred to accordingly.

Worksheet method shown in Figure 13, comprising: step S1301 to step S1302.

Step S1301 obtains problem log data and work order.

Obtain primal problem record data and original work order；

Step S1302 obtains the characteristic information of described problem record data, obtains the characteristic information of the work order, according to Described problem records the characteristic information and corresponding relationship prediction model of the characteristic information of data, the work order, asks described in foundation Corresponding relationship between topic record data and the work order；

This step establishes the corresponding relationship between problem log data and work order according to same or similar characteristic information.This reality It applies in example, the characteristic information can be classification and/or keyword.

It further include using record data and work order the problem of corresponding relationship has had been established, using machine in the embodiment of the present application Device learning algorithm such as TopK and naive Bayesian scheduling algorithm, training obtain corresponding relationship prediction model.One of them is specific real It applies in mode, corresponding relationship prediction model is classification, the classification of work order that data are recorded using the problem of corresponding relationship has been established And corresponding relationship is the classification corresponding relationship model that sample data training obtains, it is not associated for being established according to category feature Corresponding relationship between problem log data and work order.In another embodiment, corresponding relationship prediction model is to use The problem of corresponding relationship has been established records the keyword of data, the keyword of work order and the corresponding relationship as sample data instruction The keyword relational model got, the key for being established as according to keyword between associated problem log data and work order Word corresponding relationship.

In a specific embodiment provided in this embodiment, work order can be determined by the corresponding relationship prediction model Classification and problem log data categorical match probability；The work order that matched probability meets the first probability threshold value condition is made To record the work order that data have corresponding relationship with described problem.

In another embodiment provided in this embodiment, according to the corresponding relationship prediction model, work order is determined Keyword and problem log data Keywords matching probability；Matched probability is met to the work of the second probability threshold value condition It is single, as the work order with described problem record data with corresponding relationship.

In the present embodiment, the corresponding relationship prediction model that training obtains in an application scenarios includes classification corresponding relationship mould Type and keyword relational model two parts, establish the correspondence between problem log data and work order using corresponding relationship prediction model Relationship is various dimensions corresponding relationship, for example, corresponding latitude includes classification corresponding relationship and content corresponding relationship, wherein content pair It should be related to the corresponding relationship to establish according to keyword.

Corresponding with a kind of worksheet method that the application first embodiment provides, the 5th embodiment of the application also mentions A kind of work sheet processing device is supplied.

As shown in figure 14, the work sheet processing device, comprising: problem log and work order acquiring unit 1401, work order mark Associative cell 1402, object entity identify associative cell 1403, characteristic information associative cell 1404.

Problem log and work order acquiring unit 1401, for obtaining problem log data and work order；

Work order identifies associative cell 1402, for obtaining from the first problem record data in described problem record data Work order identification information is established between the first work order that the first problem records data and the work order identification information is identified First corresponding relationship；

Object entity identifies associative cell 1403, for from the Second Problem record data in described problem record data Object entity mark is obtained, object entity mark is obtained from the second work order in the work order, establishing has same object real The second corresponding relationship between the Second Problem record data of body mark and the second work order；

Characteristic information associative cell 1404, for obtaining the spy of the third problem log data in described problem record data Reference breath, obtain the characteristic information of the third work order in the work order, according to the characteristic information of the third problem log data, The characteristic information and corresponding relationship prediction model of the third work order, establish the third problem log data and the third Third corresponding relationship between work order；

Optionally, described problem record and work order acquiring unit 1401, are specifically used for:

Obtain primal problem record data and original work order；

Optionally, the work order identifies associative cell 1402, is also used to: the work order mark letter is searched from the work order Cease the first identified work order.

Optionally, the object entity identifies associative cell 1403, also particularly useful for following at least one operations:

Optionally, the characteristic information associative cell 1404 is also used to following at least one operations:

Optionally, described device further includes tracking cell, and the tracking cell is used for:

Obtain problem information；

The processing progress information or processing result for being directed to described problem information are obtained according to the specific work order.

Optionally, the work order is problem work order or task work order.

Corresponding with a kind of worksheet method that the application first embodiment provides, the application sixth embodiment also mentions A kind of work sheet processing device is supplied.

As shown in figure 15, the work sheet processing device, comprising:

Problem log obtaining unit 1501, for obtaining problem log data；

Work order identifies obtaining unit 1502, for obtaining work order identification information from described problem record data；

Work order identifies associative cell 1503, for according to the work order identification information, establish described problem record data with The corresponding relationship between work order that the work order identification information is identified.

Optionally, the work order mark obtaining unit 1502 is specifically used for: obtaining the predetermined generation rule of work order identification information Then；According to the predetermined create-rule, the work order identification information is obtained from described problem record data.

Corresponding with a kind of worksheet method that the application first embodiment provides, the 7th embodiment of the application also mentions A kind of work sheet processing device is supplied.

As shown in figure 16, the work sheet processing device, comprising: problem log and work order obtaining unit 1601, object entity Identify obtaining unit 1602, object entity identifies associative cell 1603.

Problem log and work order obtaining unit 1601, for obtaining problem log data and work order；

Object entity identifies obtaining unit 1602, for obtaining object entity mark from described problem record data, and Object entity mark is obtained from the work order；

Object entity identify associative cell 1603, for establish have the problem of same object entity identifier record data with Corresponding relationship between work order.

Optionally, the object entity mark obtaining unit 1602 is specifically used for:

Obtain the predetermined create-rule of object entity mark；

Optionally, the object entity mark associative cell 1603 is specifically used for:

Optionally, the object entity mark associative cell 1603 includes that registration is associated with subelement, and the registration closes Connection subelement is used for:

Optionally, the object entity identifies associative cell 1603, including interval association subelement, interval association Unit is used for:

Obtain the generation time of described problem record data；

Obtain the generation time of the work order；

Optionally, the object entity mark associative cell 1603 includes priority unit, the priority unit For: according to default associated priority, determining has the problem of same object entity identifier record data and work order.

Optionally, the priority unit is specifically used for:

Obtain the work order created by same creator；

Judge described problem record data whether with the work order have same object entity identifier, obtain have it is identical right Data and work order are recorded as the problem of entity identifier.

Optionally, the priority unit is specifically used for:

Corresponding with a kind of worksheet method that the application first embodiment provides, the 8th embodiment of the application also mentions A kind of work sheet processing device is supplied.

As shown in figure 17, the work sheet processing device, comprising: problem log and work order obtaining unit 1701, characteristic information Associative cell 1702.

Problem log and work order obtaining unit 1701, for obtaining problem log data and work order；

Characteristic information associative cell 1702 obtains the work order for obtaining the characteristic information of described problem record data Characteristic information, the characteristic informations of data, the characteristic information of the work order and corresponding relationship are recorded according to described problem and predicted Model establishes the corresponding relationship between described problem record data and the work order；

Optionally, the characteristic information is classification；Correspondingly, the characteristic information associative cell 1702 is specifically used for: root According to the corresponding relationship prediction model, the probability of the classification of work order and the categorical match of problem log data is determined；

The work order that matched probability is met the first probability threshold value condition is corresponding as having with described problem record data The work order of relationship.

Optionally, the characteristic information is keyword；Correspondingly, the characteristic information associative cell 1702 is specifically used for: According to the corresponding relationship prediction model, the probability of the keyword of work order and the Keywords matching of problem log data is determined；

The work order that matched probability is met to the second probability threshold value condition, it is corresponding as having with described problem record data The work order of relationship.

Corresponding with a kind of worksheet method that the application first embodiment provides, the 9th embodiment of the application also mentions A kind of electronic equipment is supplied.

As shown in figure 18, the electronic equipment includes:

Processor 1801；And

Memory 1802；For storing computer executable instructions, the processor 1802 is used for the memory 1801 Execute the computer executable instructions:

Obtain problem log data and work order；

Optionally, the processor 1802 is also used to execute following computer executable instructions:

Obtain primal problem record data and original work order；

Optionally, the processor 1802 is also used to execute following computer executable instructions: searching from the work order The first work order that the work order identification information is identified.

Optionally, the processor 1802 is also used to execute following at least one computer executable instructions:

Optionally, the processor 1802 is also used to execute following computer executable instructions: calculating the third work order Keyword and problem log data Keywords matching probability, and/or, the classification of the third work order and problem log number According to categorical match probability.

Obtain problem information；

Optionally, the work order is problem work order or task work order.

Corresponding with a kind of worksheet method that the application second embodiment provides, the tenth embodiment of the application also mentions A kind of electronic equipment is supplied, schematic diagram is similar to Figure 18.

Tenth embodiment provide electronic equipment include:

Processor；And

Memory；The memory is for storing computer executable instructions, and the processor is for executing the calculating Machine executable instruction:

Obtain problem log data；

Optionally, the processor is also used to execute following computer executable instructions:

Corresponding with a kind of worksheet method that the application 3rd embodiment provides, the 11st embodiment of the application is also A kind of electronic equipment is provided, schematic diagram is similar to Figure 18.

11st embodiment provide electronic equipment include:

Processor；And

Obtain problem log data and work order；

Obtain the predetermined create-rule of object entity mark；

Optionally, the processor is also used to execute following computer executable instructions: obtaining described problem record data The generation time；Obtain the generation time of the work order；Time and the work order are generated according to what described problem recorded data The time is generated, determines the generation time interval of described problem record data and the work order；Meet time interval in time interval The problem of threshold condition, records and establishes corresponding relationship between data and work order.

Optionally, the processor is also used to execute following computer executable instructions: according to default associated priority, really Surely there is the problem of same object entity identifier record data and work order.

Obtain the work order created by same creator；

Optionally, the processor is also used to execute following computer executable instructions: if described problem records data Do not have same object entity identifier with the work order, then obtains the work order of non-equal founder's creation, determine with identical right Data and work order are recorded as the problem of entity identifier.

Corresponding with a kind of worksheet method that the application fourth embodiment provides, the 12nd embodiment of the application is also A kind of electronic equipment is provided, schematic diagram is similar to Figure 18.

12nd embodiment provide electronic equipment include:

Processor；And

Obtain problem log data and work order；

Optionally, the characteristic information is classification；It can be performed correspondingly, the processor is also used to execute following computers Instruction: according to the corresponding relationship prediction model, the probability of the classification of work order and the categorical match of problem log data is determined；

Optionally, the characteristic information is keyword；Correspondingly, the processor be also used to execute following computers can It executes instruction: according to the corresponding relationship prediction model, determining the keyword of work order and the Keywords matching of problem log data Probability；

In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net Network interface and memory.

Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium Example.

1, computer-readable medium can be by any side including permanent and non-permanent, removable and non-removable media Method or technology realize that information stores.Information can be computer readable instructions, data structure, the module of program or other numbers According to.The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory techniques, CD-ROM are read-only Memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or Other magnetic storage devices or any other non-transmission medium, can be used for storage can be accessed by a computing device information.According to Herein defines, and computer-readable medium does not include non-temporary computer readable media (transitory media), such as modulates Data-signal and carrier wave.

2, it will be understood by those skilled in the art that embodiments herein can provide as the production of method, system or computer program Product.Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the application Form.It can be used moreover, the application can be used in the computer that one or more wherein includes computer usable program code The computer program product implemented on storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) Form.

Although the application is disclosed as above with preferred embodiment, it is not for limiting the application, any this field skill Art personnel are not departing from spirit and scope, can make possible variation and modification, therefore the guarantor of the application Shield range should be subject to the range that the claim of this application defined.

Claims

1. a kind of worksheet method characterized by comprising

Obtain problem log data and work order；

Work order identification information is obtained from the first problem record data in described problem record data, establishes the first problem The first corresponding relationship between the first work order that record data and the work order identification information are identified；

Object entity mark is obtained from the Second Problem record data in described problem record data, from the in the work order Object entity mark is obtained in two work orders, establishes the Second Problem record data and the second work order with same object entity identifier Between the second corresponding relationship；

The characteristic information for obtaining the third problem log data in described problem record data, obtains the third work in the work order Single characteristic information, according to the characteristic information of the third problem log data, the characteristic information of the third work order and right Relationship Prediction model is answered, the third corresponding relationship between the third problem log data and the third work order is established；

Wherein, the corresponding relationship prediction model is the characteristic information that data are recorded according to the first problem, first work The model that single characteristic information and first corresponding relationship training obtain, and/or, the corresponding relationship prediction model is root According to the characteristic information of Second Problem record data, the characteristic information of second work order and second corresponding relationship instruction The model got.

2. the method according to claim 1, wherein the acquisition problem log data and work order, comprising:

Obtain primal problem record data and original work order；

Data are recorded to the primal problem and original work order carries out data cleansing, obtain described problem record data and the work It is single.

3. the method according to claim 1, wherein further include:

The first work order that the work order identification information is identified is searched from the work order.

4. the method according to claim 1, wherein further including following at least one operations:

It is identified as index with the object entity obtained from Second Problem record data, is obtained from second work order Object entity mark in, search that identify identical object real for the object entity that obtains with from Second Problem record data Body mark；

It is identified as index with the object entity obtained from second work order, is obtained from Second Problem record data Object entity mark in, search and identify identical object entity with the object entity that obtains from second work order and identify.

5. the method according to claim 1, wherein further including following at least one operations:

With the characteristic information of first problem record data, the characteristic information of first work order and the first corresponding pass System is sample data, and training is for recording number according to the characteristic information of problem log data and the characteristic information forecasting problem of work order According to the corresponding relationship prediction model of the corresponding relationship with work order；

With the characteristic information of Second Problem record data, the characteristic information of second work order and the second corresponding pass System is sample data, and training is for recording number according to the characteristic information of problem log data and the characteristic information forecasting problem of work order According to the corresponding relationship prediction model of the corresponding relationship with work order.

6. according to the method described in claim 5, it is characterized in that, the corresponding relationship prediction model can export: the third The probability of the Keywords matching of the keyword and problem log data of work order, and/or, the classification and problem of the third work order are remembered Record the probability of the categorical match of data.

7. the method according to claim 1, wherein the characteristic information of first problem record data is described First problem records the keyword and/or classification of data, and the characteristic information of first work order is the key of first work order Word and/or classification, the characteristic information of Second Problem record data be the Second Problem record data keyword and/or Classification, the characteristic information of second work order are the keyword and/or classification of second work order, the third problem log number According to characteristic information be the third problem log data keyword and/or classification, the characteristic information of the third work order is The keyword and/or classification of the third work order.

8. a kind of worksheet method characterized by comprising

Obtain problem log data；

According to the work order identification information, establish described problem record data and work order that the work order identification information is identified it Between corresponding relationship.

9. a kind of worksheet method characterized by comprising

Obtain problem log data and work order；

10. a kind of worksheet method characterized by comprising

Obtain problem log data and work order；

The characteristic information for obtaining described problem record data, obtains the characteristic information of the work order, records number according to described problem According to characteristic information, the work order characteristic information and corresponding relationship prediction model, establish described problem record data and institute State the corresponding relationship between work order；

Wherein, the corresponding relationship prediction model is the feature letter according to the problem of establishing corresponding relationship with work order record data Breath, established with problem log data corresponding relationship work order characteristic information and the training of established corresponding relationship obtain Model.