CN110457473A - A kind of the problem of electric power customer service work order polymerization - Google Patents
A kind of the problem of electric power customer service work order polymerization Download PDFInfo
- Publication number
- CN110457473A CN110457473A CN201910643357.4A CN201910643357A CN110457473A CN 110457473 A CN110457473 A CN 110457473A CN 201910643357 A CN201910643357 A CN 201910643357A CN 110457473 A CN110457473 A CN 110457473A
- Authority
- CN
- China
- Prior art keywords
- work order
- text
- customer service
- polymerization
- electric power
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The present invention discloses a kind of the problem of electric power customer service work order polymerization, which comprises building customer service work order near synonym library carries out participle to work order text and near synonym are converted;After carrying out work order classification by default work order attribute to work order text, the polymerization for carrying out problem generates problem set, and the work order text and described problem are in mapping relations;In several described work order texts choose one of them be used as target work order text, and will with target work order text be in mapping relations the problem of be used as problem title, generation customer service work order issue list.Using the present invention, artificial subjective error is on the one hand reduced, on the other hand makes also to have on the working efficiency of customer service and is greatly promoted.
Description
Technical field
The present invention relates to electric power customer service work order case studies and administrative skill field, and in particular to a kind of electric power customer service
The problem of work order polymerization.
Background technique
Power grid customer service hotline service is divided into internal, external two kinds: internally, referring to and consults for the open information of interior employee
It askes, the quick service channel of problem escalation processing, it can be in the first time of day response internal user staff requirement, largely
It solves the puzzlement of worker at the production line, quickly propels work progress.Externally, refer to the information consultation for power customer, solve to use
Demand and problem of the family during electricity consumption improve the electricity consumption experience of vast Electricity customers.By dividing electric power customer service work order
Analysis research can preferably help manager to find deficiency or system defect in work.
Currently, the analysis of hotline service work order also rests on the artificial browsing of work order record, keyword retrieval, simply substantially
The word segmentation processing stage.Very strong subjective colo(u)r is had to the positioning of problem, is difficult to find one most in boundless and indistinct ground work order record
The problem of core, artificial problem analysis result will be more dispersed, ambiguous, will lose the meaning really solved the problems, such as.
Summary of the invention
To solve the above-mentioned problems, the present invention proposes a kind of the problem of electric power customer service work order polymerization, is able to solve people
The problem of work point analyses low efficiency, the deficiencies of subjectivity is strong, realizes work order text automatic mining, greatly improves working efficiency,
Also make analysis result more scientific.
Based on this, the problem of the present invention provides a kind of electric power customer service work orders polymerization, which comprises
Customer service work order near synonym library is constructed, participle is carried out to work order text and near synonym are converted;
After carrying out work order classification by default work order attribute to work order text, the polymerization for carrying out problem generates problem set, institute
Work order text and described problem are stated in mapping relations;
One of them is chosen in several described work order texts and is used as target work order text, and will be in target work order text
The problem of mapping relations, as problem title, generates customer service work order issue list.
Wherein, the customer service work order near synonym library includes: near synonym and target word, and near synonym and target word are closed in mapping
System.
Wherein, the work order text includes work order abstract, work order theme.
Wherein, the work order text carries out participle and near synonym conversion includes:
Word segmentation processing is carried out using Forward Maximum Method algorithm to the work order text, obtains word segmentation result;
The word segmentation result is matched with the near synonym in the customer service work order near synonym dictionary, obtains successful match
The near synonym;
Obtain the target word with the near synonym in mapping relations.
Wherein, described by default work order attribute includes: work order text said system, originating department, processing people.
Wherein, the polymerization of described problem includes:
Character type corpus will be obtained after work order word segmentation processing, map in numerical value vector space, each work order text
It is converted into numerical characteristics vector;
Based on work order numerical characteristics vector, the text similarity of work order two-by-two is calculated using COS distance;
The text similarity of the work order is greater than similarity threshold, and mutually independent set generates work order between work order
Collection.
Wherein, if the work order pair of work order intersection is not present with the work order in addition to the work order collection for the work order collection, respectively
Auto-polymerization is problematic, corresponding two work orders of a problem;
If the work order collection is with the work order in addition to the work order collection, there are the work orders pair of work order intersection, to there are work order friendships
The work order of the work order pair of collection takes union, and auto-polymerization is problematic respectively, corresponding two work orders of a problem.
If the text similarity is not more than similarity threshold, the corresponding work order of a problem.
Wherein, one of them is chosen in several described work order texts as target work order text includes:
If the work order that described problem is included is Gr, wherein r=1,2 ... R, R are positive integer, calculate the two of work order text
It is as follows to obtain corresponding similarity matrix for two similarities:
Calculate each work order and the other any work order similarities of described problem and:
Select GSrMaximum work order text is as target work order text.
Wherein, if the GSrMaximum work order text then appoints and takes one of them work order text as target work there are multiple
Single text.
Wherein, the customer service work order problem includes: problem list, the corresponding relationship of problem and the work order text.
The present invention compensate for manual analysis can by the very limited deficiency of technological means, instead of manually to browse,
Keyword retrieval, statistical analysis etc. are inefficient can working method.
By constructing customer service work order near synonym library, participle is carried out to work order text and near synonym are converted;It is logical to work order text
After crossing default work order attribute progress work order classification, the polymerization for carrying out problem generates problem set, the work order text and described problem
In mapping relations;One of them is chosen in several described work order texts and is used as target work order text, and will be with target work order text
The problem of this is in mapping relations generates customer service work order issue list as problem title.In this manner it is achieved that asking work order text
The real-time crawl of topic.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
The flow chart of the problem of Fig. 1 is electric power customer service work order provided in an embodiment of the present invention polymerization.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The flow chart of the problem of Fig. 1 is electric power customer service work order provided in an embodiment of the present invention polymerization, the method packet
It includes:
S101, building customer service work order near synonym library, carry out participle to work order text and near synonym are converted;
Customer service work order near synonym library is constructed, the customer service work order near synonym library refers to that one has work order and describes the close of characteristic
Adopted vocabulary inventory, is divided near synonym, target word.Wherein target word refers to a kind of near synonym of certain being matched in work order text
It converges, need to uniformly be converted into an identical target vocabulary.There are one-to-one or one-to-many relationships near synonym for target word.
The research object for selecting work order title to polymerize as work order problem.Here work order title can be according to actually adopting several feelings
Condition describes most accurate, succinct text to problem with work order abstract, work order theme etc. and substitutes.
The near synonym library is imported into participle dictionary, and uses Forward Maximum Method algorithm, work order text is segmented
Processing.
The Forward Maximum Method algorithm is to take m character of Chinese sentence to be slit as matching field from left to right, and m is
Longest entry number in big machine dictionary.
It searches big machine dictionary and is matched.If successful match, go out this matching field as a word segmentation
Come.
If matching is unsuccessful, the last character of this matching field is removed, remaining character string is as new
It with field, is matched again, above procedure is repeated, until being syncopated as all words.
In conjunction with the near synonym library and work order word segmentation result, the nearly justice conversion of work order text is carried out.By work order word segmentation result with
Near synonym near synonym library are matched, if it exists the vocabulary of successful match, then retrieve the near synonym phase with successful match
Corresponding target word carries out faithful permutation.
S102, after carrying out work order classification by default work order attribute to work order text, the polymerization generation problem of problem is carried out
Collection, the work order text and described problem are in mapping relations;
Work order is divided into different classes according to default work order attribute, work order attribute includes: said system, affiliated subsystem
System, said module, affiliated menu, sponsor unit, originating department, promoter, processing unit, processing department, processing people, type,
Urgency level.Due to different unit customer service workform management difference, local work order attribute can be had a certain difference, here work order
Attribute is not limited to the range of attributes enumerated, but all properties including actual acquired data.Selection carries out just work order
The attribute of classification is walked, it can be according to place analysis demand personalization selection.For example, selection said system, then customer service work order is just pressed
Lighting system attribute is divided into the work order subclass of not homologous ray.
It maps in numerical value vector space, work order participle and treated character type corpus each work order text
It is converted into numerical characteristics vector.The vector space model used is TF-IDF.
If work order to be processed has N number of, work order text is n by participle and treated part of speech sum, by vector sky
Between after model conversion, its feature weight can be calculated in the participle vocabulary text of each work order:
Wij(i=1,2 ..., N;J=1,2 ..., n)
And then the numerical characteristics matrix of full dose work order text is calculated, it may be expressed as:
Wherein WijIndicate the feature weight of j-th of the vocabulary of i-th of work order text, calculation is as follows:
Wij=Tij*ID Fij
Wherein TijIndicate the word frequency of j-th of the vocabulary of i-th of work order text, IDFijIndicate the jth of i-th of work order text
The inverse document frequency of a vocabulary, corresponding calculation are as follows:
word_freqijIndicate the quantity that j-th of the vocabulary of i-th of work order text occurs in work order i;word_numij
Indicate the vocabulary total amount of i-th of work order text;Doc_num indicates work order total amount, i.e. quantity N;doc_fregijIndicate occur i-th
The number of documents of j-th of vocabulary of a work order text.
The numerical characteristics vector of so each work order text may be expressed as: Wi=(Wi1,Wi2,…,Win)。
If Wpj(j=1,2 ... n), Wqj(j=1,2 ... n) be any two work order Text eigenvector, then work order p,
The Text similarity computing mode of work order q is as follows:
It is possible thereby to which the similarity matrix that full dose work order two-by-two is calculated is as follows:
Text similarity threshold epsilon is set, finds work order of the similarity greater than threshold value to Dk(k=1,2 ..., m;m≤n).
If DkIn any pair of work order be Gp、Gq, then it should meet following condition:
Condition 1:Gp、GqFor two mutually independent work orders;
Condition 2:Gp、GqWork order text similarity be greater than threshold epsilon.
The problem of based on work order text similarity threshold decision auto-polymerization, specifically include following three kinds of situations:
Situation 1: not in DkInterior work order itself independently forms a problem.In this case, the corresponding work of each problem
It is single.
Situation 2:DkIn with itself other than any other work order to the work order pair that work order intersection is all not present, it is automatic poly- respectively
Composition problem.In this case, corresponding two work orders of each problem.
Situation 3:DkIn there are the work orders pair of intersection, then to there are the work orders of the work order pair of intersection to take union, and concentrate
All work order auto-polymerizations it is problematic.In this case, the corresponding more than two work orders of each problem.
After the auto-polymerization of work order problem, work order problem set Q can be formedo(o=1,2 ..., M;M≤n).
One of them is chosen in S103, several described work order texts and is used as target work order text, and will be with target work order
The problem of text is in mapping relations generates customer service work order issue list as problem title.
If the work order that certain problem is included is Gr(r=1,2 ..., R, R≤n) calculates two two-phases of all work orders of the problem
Like degree, it is as follows to obtain corresponding similarity matrix:
Calculate each work order and the other any work order similarities of the problem and:
Select GSrMaximum work order takes its title as the title of the problem.If GSrMaximum work order is there are multiple, then
Appointing takes one of work order title as problem title.
Described problem inventory contains the corresponding relationship of problem list and problem and original work order.
For ease of understanding, a kind of the problem of electric power customer service work order provided by the present invention, will be gathered with specific example below
Conjunction method expansion detailed description.
(S1): the building proprietary near synonym library of customer service work order carries out participle to work order title and near synonym is converted;
The proprietary near synonym library of electric power customer service work order is constructed, includes near synonym and target word, few examples such as table 1.
The 1 proprietary near synonym library of electric power customer service work order of table
Selected section example work order, include work order ID, title, details such as table 2:
2 work order example of table
After nearly adopted library imports participle model, for the work order word segmentation result such as table 3 of table 2:
3 work order text of table participle
Near synonym transformation result such as table 4 after participle:
The nearly justice conversion of 4 word segmentation result of table
(S2): after carrying out preliminarily work order classification by preset attribute again, the auto-polymerization of problem is carried out to work order text;
The attribute for work order preliminary classification selected in this example is said system, therefore preceding 7 work orders can be divided
It is divided into marketing system subclass for terminal system subclass, the last one work order, then carries out the auto-polymerization of work order problem respectively.
5 WorkForm System attribute of table
For purposes of illustration only, following the problem of mainly using terminal system work order, polymerize deployment algorithm result explanation.
It is converted into numerical characteristics vector with work order word segmentation processing result of the vector space model to terminal system, as a result such as
Under:
6 work order numerical characteristics vector of table
The work order similarity matrix two-by-two being calculated by cosine similarity is indicated in table 6 from top to bottom with X1~X7
Work order.
7 work order similarity matrix of table
X1 | X2 | X3 | X4 | X5 | X6 | X7 | |
X1 | 1 | 0.9117 | 0.5403 | 0.5403 | 0.4399 | 0.5403 | 0.1297 |
X2 | 0.9117 | 1 | 0.5066 | 0.5066 | 0.3197 | 0.5066 | 0.0934 |
X3 | 0.5403 | 0.5066 | 1 | 1 | 0.9599 | 1 | 0.3712 |
X4 | 0.5403 | 0.5066 | 1 | 1 | 0.9599 | 1 | 0.3712 |
X5 | 0.4399 | 0.3197 | 0.9599 | 0.9599 | 1 | 0.9599 | 0.3066 |
X6 | 0.5403 | 0.5066 | 1 | 1 | 0.9599 | 1 | 0.3712 |
X7 | 0.1297 | 0.0934 | 0.3712 | 0.3712 | 0.3066 | 0.3712 | 1 |
The text similarity threshold value that work order problem polymerize tentatively is set as 0.6, then meeting the work order of threshold value to having:
(X1, X2), (X3, X4), (X3, X5), (X3, X6), (X4, X5), (X4, X6)
Wherein X7 is not in above-mentioned work order pair, therefore X7 is individually polymerized to problem Q1.
(X1, X2) and other work orders are not to having a work order intersection, therefore the problematic Q2 of X1, X2 auto-polymerization.
(X3, X4), (X3, X5), (X3, X6), (X4, X5), between (X4, X6) work order pair there are work order intersection, therefore X3,
The problematic Q3 of X4, X5, X6 auto-polymerization.
Similarly, marketing system work order is polymerized to problem Q4 automatically
(S3): after formation problem, best heading is put forward from the original work order that problem is related to as problem title;
By taking Q3 as an example, the process that title automatically generates is described the problem.
The work order similarity matrix of Q3 is obtained first
8 Q3 work order similarity matrix of table
X3 | X4 | X5 | X6 | |
X3 | 1 | 1 | 0.9599 | 1 |
X4 | 1 | 1 | 0.9599 | 1 |
X5 | 0.9599 | 0.9599 | 1 | 0.9599 |
X6 | 1 | 1 | 0.9599 | 1 |
Calculate the similarity and GS of each work order Yu the other any work orders of Q3r:
9 Q3 work order GS of tablerValue
Wherein GSrBeing worth maximum work order is X3, X4, X6, therefore selects work order X3 title: " printer uses abnormal " is as Q3
The problem of title.
Similarly title the problem of available Q1, Q2, Q4.
(S4): eventually forming customer service work order issue list.
The final corresponding problem list of issue list and work order details table are as shown in table 10, table 11.
10 problem list of table
11 work order details of table
The method of the invention, first acquisition work order basic information simplify work in conjunction with nearly adopted converting text word segmentation processing
Text expression diversification when single typing improves the precision of problem polymerization so that the work order text similarity of same problem is higher.
It is based further on work order text similarity theoretical basis problem of implementation auto-polymerization overall process.From embodiment it is found that this method section
It learns, precisely, effectively, workform management department can be helped to find most distinct issues instantly in time, and be deeper problem
Research provides basis.
Above-mentioned, although the foregoing specific embodiments of the present invention is described with reference to the accompanying drawings, not protects model to the present invention
The limitation enclosed, those skilled in the art should understand that, based on the technical solutions of the present invention, those skilled in the art are not
Need to make the creative labor the various modifications or changes that can be made still within protection scope of the present invention.
Claims (10)
1. a kind of the problem of electric power customer service work order polymerization characterized by comprising
Customer service work order near synonym library is constructed, participle is carried out to work order text and near synonym are converted;
After carrying out work order classification by default work order attribute to work order text, the polymerization for carrying out problem generates problem set, the work
Single text and described problem are in mapping relations;
One of them is chosen in several described work order texts and is used as target work order text, and will be with target work order text in mapping
The problem of relationship, as problem title, generates customer service work order issue list.
2. the problem of electric power customer service work order as described in claim 1 polymerization, which is characterized in that the customer service work order is closely adopted
Dictionary includes: near synonym and target word, and near synonym and target word are in mapping relations.
3. the problem of electric power customer service work order as described in claim 1 polymerization, which is characterized in that the work order text includes
Work order abstract, work order theme.
4. the problem of electric power customer service work order as claimed in claim 1 or 2 polymerization, which is characterized in that the work order text
It carries out participle and near synonym conversion includes:
Word segmentation processing is carried out using Forward Maximum Method algorithm to the work order text, obtains word segmentation result;
The word segmentation result is matched with the near synonym in the customer service work order near synonym dictionary, obtains the institute of successful match
State near synonym;
Obtain the target word with the near synonym in mapping relations.
5. the problem of electric power customer service work order as described in claim 1 polymerization, which is characterized in that described by presetting work order
Attribute includes: work order text said system, originating department, processing people.
6. the problem of electric power customer service work order as described in claim 1 polymerization, which is characterized in that the polymerization packet of described problem
It includes:
Character type corpus will be obtained after work order word segmentation processing, map in numerical value vector space, each work order text conversion
At numerical characteristics vector;
Based on work order numerical characteristics vector, the text similarity of work order two-by-two is calculated using COS distance;
The text similarity of the work order is greater than similarity threshold, and mutually independent set generates work order collection between work order.
7. the problem of electric power customer service work order as described in claim 1 polymerization, which is characterized in that if the work order collection with remove
The work order pair of work order intersection is not present in work order except the work order collection, and auto-polymerization is problematic respectively, a problem corresponding two
A work order;
If the work order collection is with the work order in addition to the work order collection, there are the work orders pair of work order intersection, to there are work order intersections
The work order of work order pair takes union, and auto-polymerization is problematic respectively, corresponding two work orders of a problem;
If the text similarity is not more than similarity threshold, the corresponding work order of a problem.
8. the problem of electric power customer service work order as described in claim 1 polymerization, which is characterized in that several work orders text
One of them is chosen in this as target work order text includes:
If the work order that described problem is included is Gr, wherein r=1,2 ... R, R are positive integer, calculate two two-phases of work order text
Like degree, it is as follows to obtain corresponding similarity matrix:
[SR1 SR2 … SRR]
Calculate each work order and the other any work order similarities of described problem and:
Select GSrMaximum work order text is as target work order text.
9. the problem of electric power customer service work order as claimed in claim 8 polymerization, which is characterized in that if the GSrMaximum work
Single text then appoints and takes one of them work order text as target work order text there are multiple.
10. the problem of electric power customer service work order as described in claim 1 polymerization, which is characterized in that the customer service work order is asked
Topic includes: problem list, the corresponding relationship of problem and the work order text.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910643357.4A CN110457473A (en) | 2019-07-16 | 2019-07-16 | A kind of the problem of electric power customer service work order polymerization |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910643357.4A CN110457473A (en) | 2019-07-16 | 2019-07-16 | A kind of the problem of electric power customer service work order polymerization |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110457473A true CN110457473A (en) | 2019-11-15 |
Family
ID=68481320
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910643357.4A Pending CN110457473A (en) | 2019-07-16 | 2019-07-16 | A kind of the problem of electric power customer service work order polymerization |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110457473A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110955777A (en) * | 2019-12-11 | 2020-04-03 | 国网河北省电力有限公司电力科学研究院 | Method for improving customer service quality based on 95598 work order |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8935318B1 (en) * | 2011-03-28 | 2015-01-13 | Google Inc. | Opportunistic job processing in a distributed computer environment |
CN105335496A (en) * | 2015-10-22 | 2016-02-17 | 国网山东省电力公司电力科学研究院 | Customer service repeated call treatment method based on cosine similarity text mining algorithm |
CN105468677A (en) * | 2015-11-13 | 2016-04-06 | 国家计算机网络与信息安全管理中心 | Log clustering method based on graph structure |
CN108876452A (en) * | 2018-06-12 | 2018-11-23 | 广东电网有限责任公司 | Electricity customers demand information acquisition methods, device and electronic equipment |
-
2019
- 2019-07-16 CN CN201910643357.4A patent/CN110457473A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8935318B1 (en) * | 2011-03-28 | 2015-01-13 | Google Inc. | Opportunistic job processing in a distributed computer environment |
CN105335496A (en) * | 2015-10-22 | 2016-02-17 | 国网山东省电力公司电力科学研究院 | Customer service repeated call treatment method based on cosine similarity text mining algorithm |
CN105468677A (en) * | 2015-11-13 | 2016-04-06 | 国家计算机网络与信息安全管理中心 | Log clustering method based on graph structure |
CN108876452A (en) * | 2018-06-12 | 2018-11-23 | 广东电网有限责任公司 | Electricity customers demand information acquisition methods, device and electronic equipment |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110955777A (en) * | 2019-12-11 | 2020-04-03 | 国网河北省电力有限公司电力科学研究院 | Method for improving customer service quality based on 95598 work order |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109376352B (en) | Patent text modeling method based on word2vec and semantic similarity | |
US8341112B2 (en) | Annotation by search | |
Ni et al. | Short text clustering by finding core terms | |
CN102508909B (en) | Image retrieval method based on multiple intelligent algorithms and image fusion technology | |
CN106547864B (en) | A kind of Personalized search based on query expansion | |
Zhang et al. | Ontology-based semantic retrieval for engineering domain knowledge | |
CN108549731A (en) | A kind of knowledge mapping construction method based on ontology model | |
CN103577394B (en) | A kind of machine translation method based on even numbers group searching tree and device | |
CN104391908B (en) | Multiple key indexing means based on local sensitivity Hash on a kind of figure | |
US20110191335A1 (en) | Method and system for conducting legal research using clustering analytics | |
US8583669B2 (en) | Query suggestion for efficient legal E-discovery | |
CN107665217A (en) | A kind of vocabulary processing method and system for searching service | |
CN109408578A (en) | One kind being directed to isomerous environment monitoring data fusion method | |
CN104915388B (en) | It is a kind of that method is recommended based on spectral clustering and the book labels of mass-rent technology | |
CN107832319A (en) | A kind of heuristic enquiry expanding method based on semantic relationship network | |
CN110457473A (en) | A kind of the problem of electric power customer service work order polymerization | |
CN104778205B (en) | A kind of mobile application sequence and clustering method based on Heterogeneous Information network | |
Hanyurwimfura et al. | A centroid and relationship based clustering for organizing | |
CN113722460B (en) | Index data warehousing method, device, equipment and storage medium | |
CN114428834B (en) | Retrieval method, retrieval device, electronic equipment and storage medium | |
CN108509449B (en) | Information processing method and server | |
CN115937881A (en) | Method for automatically identifying content of knowledge graph construction standard form | |
CN108846094A (en) | A method of based on index in classification interaction | |
CN105868406A (en) | Multi-database based patent retrieval system | |
CN109344261B (en) | Common word and common introduction analysis-based knowledge graph analysis system for primary and secondary education |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191115 |
|
RJ01 | Rejection of invention patent application after publication |