CN107194617A - A kind of app software engineers soft skill categorizing system and method - Google Patents
A kind of app software engineers soft skill categorizing system and method Download PDFInfo
- Publication number
- CN107194617A CN107194617A CN201710544531.0A CN201710544531A CN107194617A CN 107194617 A CN107194617 A CN 107194617A CN 201710544531 A CN201710544531 A CN 201710544531A CN 107194617 A CN107194617 A CN 107194617A
- Authority
- CN
- China
- Prior art keywords
- short sentence
- soft skill
- soft
- skill
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Educational Administration (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Development Economics (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Game Theory and Decision Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
Abstract
The present invention relates to a kind of app software engineers soft skill categorizing system and method, the soft skill extracted for classifying from app PHP's wanted advertisements.The present invention extracts soft skill in being required from the recruitment of software engineer's technical ability, Data Collection and cleaning are carried out to wanted advertisement information first, then pretreatment is carried out to wherein information and extracts short sentence and word, the weight of each word is identified with TF IDF methods, afterwards short sentence calculating formula of similarity is defined according to the term weighing in each short sentence, finally clustered by reference pair soft skill of short sentence similarity, keyword is extracted as classifying rules structural classification device from cluster result, and most mobile software development engineer soft skill is categorized as 13 classes at last.
Description
Technical field
The present invention proposes a kind of app software engineers soft skill categorizing system and method, will be moved using hierarchical clustering algorithm
The soft skill demand classification of dynamic exploitation software engineer is 13 classes.
Background technology
In general, the technical ability of developer is divided into two aspects:Hard technical ability and soft skill.Hard technical ability is that a people should
The technical requirements and knowledge possessed, it is used for performing task;They include people should special executive plan task theory
Basis and practical experience.Soft skill is non-technical skills, incorporates psychological phenomena, and such as social interaction's ability is linked up, innovation and association
Make.Developer is typically considered technician, therefore, and their technical capability is emphasised in real work and research.
But software development is mental intensive industry, and it is based on team unity, so the soft skill of developer, such as links up and assist
Make etc., it should also be paid close attention to by same degree.In the research developed both at home and abroad to traditional software at present, existing part research is taken off
Show which soft skill has very high requirement in software development, and proposed software engineer under global software development background and answered
The soft skill having requirement, but the importance ranking of those soft skills is ignored.And these researchs and traditional soft
Part exploitation is relevant, and traditional software exploitation has the difference in development scheme and way to manage with Mobile solution software development, therefore
The soft skill demand proposed in traditional software development environment may be not suitable for Mobile solution software development environment.In addition, these are soft
The classification of technical ability is not comprehensive enough, it is impossible to the multi-faceted classification for embodying soft skill.
Soft skill is divided into 9 classifications, including communication skill, interpersonal relationships technical ability, analysis and solution in existing research
Problem ability, team unity spirit, organizational capacity, Fast Learning ability, capacity of working on one's own, innovation ability and adaptability.
But without being related to responsibility consciousness and active sex chromosome mosaicism, and this be also modern Mobile solution software development staff must not
The soft skill that can lack.Soft skill text is typical unstructured data, and traditional analysis method is to use manual analysis method,
Inefficiency, and supported without data, the experience and subjective consciousness of people is depended primarily on, reliability is relatively low.
The content of the invention
It is of the present invention to solve the problems, such as:Soft skill research contents vacancy in Mobile solution software development is filled up, proposes that a kind of app is soft
Part engineer's soft skill categorizing system and method, realized using Text Mining Technology and hierarchical clustering algorithm soft skill extract and
Classification feature, this be directed to the widely used method of unstructured data, can with rapid extraction user information interested,
The low problem of inefficiency, accuracy rate that Traditional Man extracts soft skill method is solved, while using a large amount of text datas to rely on, increasing
Soft skill has been added to extract the reliability of classification.
The technology of the present invention solution:A kind of app software engineers soft skill categorizing system, including Data Collection and cleaning
Module, full text pretreatment module and set up class Modules;Wherein:
Data Collection and cleaning modul, realize the function that data and data filtering screening are collected from data source recruitment website,
Finally give the text data for including mobile software development personnel's skill requirement.Last set keyword is constructed first:It is mobile
Exploitation, Android/Android engineer, IOS engineer;The webpage capture instrument searched for and write using Python is obtained
Numerous wanted advertisements.In the wanted advertisement issued on recruitment website mainly include four aspect contents, be respectively position title,
Position temptation, job description and job requirement, only crawl the wanted advertisement containing defined keyword in position title, because
These advertisements are just closely related with Mobile Development;Because not including job requirement in the wanted advertisement of part, and the present invention mainly makes
Data are exactly job requirement, so carrying out search operation in data set, the advertisement not comprising job requirement are found out, by it
Deleted from data set;Finally, the text data set of job requirement has been obtained including, used in being pre-processed in full in next module
Continue to analyze;
Full text pretreatment module, obtains Data Collection and the text data set of cleaning modul, the job requirement from data set
Soft skill word is screened in sentence.First, the text between job requirement subtitle and work address is extracted from advertisement, this
Part text is to include the job requirement of mobile software development soft skill and hard technical ability;Then, enter according to programming language keyword
Row filter operation, deletes the sentence for including such programming language keyword, so as to obtain pure soft skill sentence;Followed by
Cutting operation, short sentence is divided into using comma, fullstop and branch as cut-point by sentence;Finally short sentence is carried out using participle instrument
Participle, obtains soft skill term data collection, and the short sentence that this module is obtained and term data collection are used as soft skill sort module
Direct data set is called for it.
Class Modules are set up, the soft skill short sentence and term data collection obtained according to full text pretreatment module is set up accordingly
Soft skill is classified and the extracting rule from classification results.First, soft skill word is calculated in data set according to TF-IDF methods
Weight, the more high then weight of the word frequency of occurrences is lower;Then, calculate soft skill short sentence between semantic similarity, this stage according to
Calculated according to term weighing;Next hierarchical clustering algorithm is used, cluster operation is carried out to soft skill short sentence, soft skill is obtained
Classification results;Finally, structural classification device, from classification results extracting rule as grader class representative.
A kind of app software engineers soft skill sorting technique, step is as follows:
(1) initial data used is extracted from online recruitment website pull hook net, and one group is constructed first and is searched
Rope keyword:Mobile Development, Android/Android engineer, IOS engineer.The webpage searched for and write using Python
Gripping tool obtains the numerous wanted advertisements that institute's definition of keywords is included in position title.Next sieved from wanted advertisement
The data text not comprising job requirement is selected, is abandoned, remaining data collection is handed over to full text pretreatment module and used.
(2) text that the present invention is extracted in the job requirement of each advertisement is stored into database.In view of each record
In potentially include the requirement of more than one soft skill, according to three kinds of punctuation marks, i.e. comma, branch and fullstop contract these long sentences
Short is short sentence, to ensure that every short sentence is related to the skill set requirements of minimum;The hard technical ability in filtering job requirement text, is used afterwards
The name of programming language such as C, JAVA, Python and PHP etc. are filtered as keyword;Finally, using participle instrument JieBa
Participle is carried out to acquired short sentence.JieBa participles instrument realizes that efficient word figure is scanned using Trie Tree structures, passes through
The maximum probability path of dynamic programming method search terms segmentation, and Unrecorded word is carried out preferably using HMM model
Automatic identification.By using JieBa participle instruments, we finally give the word of each short sentence, sentence collection and word collection for building
Vertical class Modules are called.
(3) by obtaining the data set of full text pretreatment module, each wanted advertisement soft skill requirement is obtained and each
The related short sentence of the noun word of sentence;Then, according to the similarity of word in different sentences, cluster point is carried out to these short sentences
Analysis.First, it is proposed that a kind of method for calculating term weighing.According to TF-IDF methods, the number of times that word occurs is more, then word
Weight it is lower;Afterwards, if a short sentence a is included and another short sentence b identical words, then they have very high phase
Like property, the similitude of short sentence is calculated according to term weighing.In order to be clustered, there is provided a threshold value 0.8, according to multiple clusters
The result of experiment is selected;If short sentence degree of membership is more than 0.8, it may be determined that it belongs to such, because there may be several
Classification, so calculating the degree of membership of this short sentence in each classification, and is found that maximum membership degree;If maximum is less than threshold
Value, then generate a new category, this short sentence by be new category first element.Otherwise, degree of membership maximum will be attributed to relative
The classification answered.Because only considered the frequency information of same word in clustering algorithm, and different terms semantic information is not considered
Influence, so, some classifications according to the manual merger of semantic similarity;Finally, in order to obtain by one group of word or short sentence
The final soft skill classification of Rule Expression, constructs a grader;For each soft skill classification, from previous step obtain it is every
Keyword is extracted in short sentence in individual classification as the rule of grader.
The advantage of the invention is that:
(1) in the existing research on mobile software application development, not on developer soft skill content,
The present invention innovatively proposes a kind of Mobile solution PHP soft skill categorizing system and method, and according to acquisition
Recruitment information be therefrom extracted 13 soft skill classifications, enterprise can be analyzed from the frequency of occurrence of cluster result to each soft skill
Attention degree.
(2) hierarchical clustering algorithm is applied among wanted advertisement by the present invention, passes through the job requirement in wanted advertisement point
Analyse technical ability needed for applicant.There is provided a kind of thinking interdisciplinary, clustering algorithm is applied in wanted advertisement text, by this
One analytical information interested.
Brief description of the drawings
Fig. 1 is implementation process figure of the present invention;
Fig. 2 is the processing illustration of data full text pretreatment module in the present invention;
Fig. 3 is sets up the flow chart and DFD of class Modules in the present invention.
Embodiment
As shown in figure 1, a kind of app software engineers soft skill categorizing system of the invention, including three modules:Data Collection
With cleaning, pre-process in full and set up classification, specific technical scheme is described as follows:
1. the collection and cleaning of data
Initial data used herein is extracted from online recruitment website pull hook net.One group is constructed first to search
Rope keyword:Mobile Development, Android/Android engineer, IOS engineer.The webpage searched for and write using Python
Gripping tool is obtained in many wanted advertisements, the position title of these advertisements comprising defined keyword.Following root
Constituted according to ad content and the text data not comprising job requirement is filtered out from wanted advertisement, abandoned.
2. pre-process in full
The text that the present invention is extracted in the job requirement of each advertisement is stored into database.In view of can in each record
The requirement of more than one soft skill can be included, these long sentences are shorten to according to three kinds of punctuation marks (comma, branch and fullstop)
Short sentence, to ensure that each short sentence is related to the skill set requirements of minimum.The hard technical ability in filtering job requirement text, uses programming afterwards
The name of language such as C, JAVA, Python and PHP etc. are filtered as keyword;Finally, using participle instrument JieBa to
The short sentence of acquisition carries out participle.JieBa participles instrument realizes that efficient word figure is scanned using Trie Tree structures, passes through dynamic
The maximum probability path of planing method search terms segmentation, and it is preferably automatic to the progress of Unrecorded word using HMM model
Identification.By using JieBa participle instruments, we finally give the word of each short sentence, and sentence collection and word collection are for setting up class
Other module is called.This process processing illustration is as shown in Figure 2.
3. build soft skill classification
By above-mentioned data prediction, the noun word phase of each wanted advertisement soft skill requirement and each sentence is obtained
The short sentence of pass.Then, according to the similarity of word in different sentences, clustering is carried out to these short sentences.
First, it is proposed that a kind of method for calculating term weighing.According to TF-IDF methods, the number of times that word occurs is more,
Then the weight of word is lower.According to TF-IDF methods, weight calculation equation is as follows:
Wherein, wiRepresent short sentence SwiIn word, S represents short sentence collection, SwiRepresenting one group includes wiShort sentence, count (S)
Represent word quantity sum, count (S in short sentence collection Swi) represent short sentence SwiIn word quantity sum, IDFwiRepresent wiPower
Weight.
Afterwards, if short sentence a is included and another short sentence b identical words, then they have very high similar
Property.A variable Sim (a, b) is defined to represent two short sentences a, b similarity.According to the contribution of each word, similarity meter
Calculate formula as follows:
Wherein wiAnd ziRepresent respectively word in short sentence a and short sentence b common factor and and any one element for concentrating, therefore on
Molecular moiety in formula is stated to represent to each element weights summation in the common factor of word in short sentence a and short sentence b;Denominator is represented to short
In sentence a and short sentence b word and concentrate each element weights to sum.
According to the similitude of short sentence, they are clustered to obtain the preliminary classification of soft skill, the process is changed by circulation
What in generation, completed.From first to last short sentence, judge that each sentence belongs to existing classification or new category, the category is
Based on similitude dynamic construction.
Class is defined to represent an existing category set, ClassjFor wherein jth class;And siRepresent one group of short sentence
A short sentence in candidate, it is not classified.Therefore, a membership function Membership (s is defined firsti,Classj)
To represent siRelative to class ClassjDegree of membership.Equation is as follows:
Wherein bkIt is ClassjIn k-th of short sentence, count (Classj) mean to calculate ClassjIn how many is short
Sentence.
In order to be clustered, there is provided a threshold value 0.8, is selected according to the result of multiple cluster experiments.If si
Relative to ClassjDegree of membership be more than 0.8, it may be determined that siIt may belong to Classj.Because there may be several classifications,
Calculate siRelative to the degree of membership of each classification, and find maximum membership degree.If maximum is less than threshold value, one is generated
New category, siTo be first element of this new category.Otherwise, will be siIt is attributed to the corresponding classification of maximum membership degree.
Because only considered the frequency information of same word in clustering algorithm, and different terms semantic information is not considered
Influence.So, some classifications according to the manual merger of semantic similarity.
Up to the present, the element in classification is still similar short sentence.In order to obtain by one group of word or short sentence rule
The final soft skill classification represented, while the data collected after convenient analysis, it is desirable to have a rule-based grader.This
The data of secondary collection will carry out initial work to grader, in the research after classifier rules will likely increase with comprising
Device identification but satisfactory classification phrase are not classified, so constructing a grader with this.For each soft skill point
Class, extracts keyword as the rule of grader from the short sentence in each classification that previous step is obtained.Obviously, a classification can
It can produce multiple rules.Ensure that each short sentence at least conforms to a rule.
For example, it was discovered that class n is (expressive) to have the implication similar to class 1 (communication capability is good), they are integrated into friendship
Flow in technical ability classification.Classification after merging, the category is represented by extracting classification rule, such as, in above-mentioned classification
Arrived communication skill require classification, be then extracted " communication capability " and " being good at exchange " the two rule combine be used as this class
Representative.Simultaneously in order to ensure the short sentence in each classification be correctly validated, it is necessary to classification short sentence carry out grader screening,
To ensure each short sentence in classification by regular correct differentiation.
(dirty data refers to not filter out by the screening of hard technical ability dirty data in data prediction, but and soft skill
Unrelated job requirement, such as " resume asks attached works " this printed words) can not be removed in pretreatment, therefore with the addition of one it is dirty
Class, to handle dirty data, and builds classifier rules for it, but such will not be shown in classification.All not phases in cluster
The class of pass will be all added in this classification, for being clustered in next circulation.
Cluster and the process of extracting rule are repeated, until the data without packet are sky, or final result slowly improves,
Categorical measure and regular quantity are slowly increased.Fig. 3 is the flow chart and DFD of cluster and Rule Extraction.Finally, we obtain
The classification of Mobile Development personnel soft skill requirement was obtained, soft skill category result is as shown in table 1.5423 tricks have been crawled in implementation
Engage in advertisement, table 1 list of quantity one to be shown with how many wanted advertisements and include this technical ability.
The soft skill classification of table 1
Sequence number | Content | Quantity | Ratio |
1 | Communication capability | 2668 | 61.33% |
2 | Team unity | 2608 | 59.95% |
3 | Analysis and ability to solve problem | 2039 | 46.87% |
4 | The sense of duty | 1683 | 38.69% |
5 | Fast Learning ability | 1458 | 33.51% |
6 | Carry out challenging job | 1138 | 26.16% |
7 | Coding custom | 930 | 21.38% |
8 | Bear pressure | 704 | 16.18% |
9 | Work independently | 558 | 12.83% |
10 | Positive working attitude | 471 | 10.83% |
11 | Organizational capacity | 339 | 7.79% |
12 | Innovation ability | 326 | 7.49% |
13 | English reading ability | 278 | 6.39% |
Above example is provided just for the sake of the description purpose of the present invention, and is not intended to limit the scope of the present invention.This
The scope of invention is defined by the following claims.The various equivalent substitutions that do not depart from spirit and principles of the present invention and make and repair
Change, all should cover within the scope of the present invention.
Claims (2)
1. a kind of app software engineers soft skill categorizing system, it is characterised in that:Including Data Collection and cleaning modul, full text
Pretreatment module and set up class Modules;Wherein:
Data Collection and cleaning modul, realize the function that data and data filtering screening are collected from data source recruitment website, finally
Obtain including the text data of mobile software development personnel's skill requirement;The wanted advertisement issued on recruitment website generally comprises four
Aspect core content:Position title, position temptation, job description and job requirement, are followed by work address;One group is built first
Search key:Mobile Development, Android/Android engineer, IOS engineer;Search for and use webpage capture instrument, obtain
The numerous wanted advertisements containing definition of keywords in position title;The content formation data set included in wanted advertisement;
Search operation is carried out in data set, the advertisement not comprising job requirement is found out, and by the advertisement not comprising job requirement from data
Concentrate and delete, finally give the data set comprising job requirement, continue to analyze in pre-processing for full text;
Full text pretreatment module, about screening soft skill word in the sentence of job requirement from the data set comprising job requirement
Language;First, the text between job requirement subtitle and work address is extracted, the text is that to include mobile software development soft
The job requirement of technical ability and hard technical ability;Then cutting operation is carried out, by cut-point of comma, fullstop and branch by job requirement portion
Sentence is divided to be divided into short sentence;Then, the foundation keyword related to programming language carries out filter operation, deletes and is closed comprising such
The hard technical ability sentence of key word, so as to obtain pure soft skill short sentence;Participle is carried out finally by pure soft skill short sentence, soft skill is obtained
Energy set of words, soft skill set of words includes soft skill short sentence and soft skill term data collection, by soft skill short sentence and soft skill
Energy term data collection is called as direct data set for setting up class Modules;
Class Modules are set up, the soft skill short sentence and soft skill term data collection obtained according to full text pretreatment module realizes evidence
This set up soft skill classification and from category result extracting rule function;First, soft skill word is calculated according to TF-IDF methods
Weight of the language in data set, the more high then weight of the word frequency of occurrences is lower;Then, it is semantic similar between calculating soft skill short sentence
Degree, this stage is calculated according to term weighing;Hierarchical clustering algorithm is reused, cluster operation is carried out to soft skill short sentence, obtained
To soft skill cluster result;Last structural classification device, from cluster result extracting rule as grader class representative.
2. a kind of app software engineers soft skill sorting technique, it is characterised in that step is as follows:
(1) Data Collection and cleanup step:The initial data used is obtained from online recruitment website pull hook net, is led to first
Cross and construct last set keyword:Mobile Development, Android/Android engineer, IOS engineer, are searched for and are grabbed using webpage
Instrument is taken, the numerous Mobile Development class wanted advertisements containing definition of keywords are obtained;Next searched in data set, from
The data text not comprising job requirement is filtered out in wanted advertisement, is abandoned, remaining data collection is handed over to full text and pre-processed
Step is used;
(2) full text pre-treatment step:The text extracted in the job requirement of each advertisement is stored into database, with comma, sentence
Number and branch be cut-point these long sentences are shorten to short sentence;Afterwards, duty is fallen using the keyword filtration related to programming language
Position requires the hard technical ability in text, obtains soft skill short sentence data set;Finally, acquired short sentence is carried out using participle instrument
Participle, finally gives the term data collection of each short sentence, soft skill short sentence and soft skill term data collection are for setting up class Modules
Call;
(3) classification step is set up:After the soft skill short sentence and soft skill term data collection that obtain full text pre-treatment step, according to not
Similarity with word in short sentence carries out clustering operation;First, it is proposed that it is a kind of calculate term weighing method, according to
TF-IDF methods, the number of times that word occurs is more, then the weight of word is lower;Afterwards, the phase of short sentence is calculated according to term weighing
Like property;There is provided threshold value 0.8, if short sentence degree of membership is more than threshold value, determines that short sentence belongs to such;If degree of membership is less than threshold
Value, then generate a new category, this short sentence by be new category first element;Then, manually returned according to semantic similarity
And some classifications;Finally, rule-based grader is constructed with researching and analysing after facilitating by cluster result, from upper one
Walk the rule representative that keyword is extracted in the short sentence in each category set obtained as grader.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710544531.0A CN107194617B (en) | 2017-07-06 | 2017-07-06 | App software engineer soft skill classification system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710544531.0A CN107194617B (en) | 2017-07-06 | 2017-07-06 | App software engineer soft skill classification system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107194617A true CN107194617A (en) | 2017-09-22 |
CN107194617B CN107194617B (en) | 2020-11-24 |
Family
ID=59881385
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710544531.0A Active CN107194617B (en) | 2017-07-06 | 2017-07-06 | App software engineer soft skill classification system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107194617B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107506389A (en) * | 2017-07-27 | 2017-12-22 | 北京德塔精要信息技术有限公司 | A kind of method and apparatus for extracting position skill requirement |
CN108460699A (en) * | 2017-12-20 | 2018-08-28 | 卓智网络科技有限公司 | Teaching programme optimization method and device |
CN108549640A (en) * | 2018-04-24 | 2018-09-18 | 易联众信息技术股份有限公司 | One kind being based on statistical enterprise name similarity calculating method |
CN109934302A (en) * | 2019-03-23 | 2019-06-25 | 大国创新智能科技(东莞)有限公司 | New category recognition methods and robot system based on fuzzy theory and deep learning |
CN110033022A (en) * | 2019-03-08 | 2019-07-19 | 腾讯科技(深圳)有限公司 | Processing method, device and the storage medium of text |
CN110377907A (en) * | 2019-07-18 | 2019-10-25 | 中科鼎富(北京)科技发展有限公司 | A kind of recruitment information standardized method and device |
CN112434211A (en) * | 2020-11-04 | 2021-03-02 | 广州视源电子科技股份有限公司 | Data processing method, device, storage medium and equipment |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102622373A (en) * | 2011-01-31 | 2012-08-01 | 中国科学院声学研究所 | Statistic text classification system and statistic text classification method based on term frequency-inverse document frequency (TF*IDF) algorithm |
CN102708096A (en) * | 2012-05-29 | 2012-10-03 | 代松 | Network intelligence public sentiment monitoring system based on semantics and work method thereof |
WO2013085102A1 (en) * | 2011-12-09 | 2013-06-13 | (주)네오위즈게임즈 | Method and apparatus for managing characters in online game |
CN104866572A (en) * | 2015-05-22 | 2015-08-26 | 齐鲁工业大学 | Method for clustering network-based short texts |
CN105468713A (en) * | 2015-11-19 | 2016-04-06 | 西安交通大学 | Multi-model fused short text classification method |
CN105512864A (en) * | 2016-01-28 | 2016-04-20 | 丁沂 | Method for automatically acquiring post professional ability requirements based on internet |
US20160109257A1 (en) * | 2004-11-16 | 2016-04-21 | Microsoft Technology Licensing, Llc | Methods for automated and semiautomated composition of visual sequences, flows, and flyovers based on content and context |
WO2017014124A1 (en) * | 2015-07-21 | 2017-01-26 | 東亞合成株式会社 | Curable composition |
CN106570076A (en) * | 2016-10-11 | 2017-04-19 | 深圳大学 | Computer text classification system |
-
2017
- 2017-07-06 CN CN201710544531.0A patent/CN107194617B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160109257A1 (en) * | 2004-11-16 | 2016-04-21 | Microsoft Technology Licensing, Llc | Methods for automated and semiautomated composition of visual sequences, flows, and flyovers based on content and context |
CN102622373A (en) * | 2011-01-31 | 2012-08-01 | 中国科学院声学研究所 | Statistic text classification system and statistic text classification method based on term frequency-inverse document frequency (TF*IDF) algorithm |
WO2013085102A1 (en) * | 2011-12-09 | 2013-06-13 | (주)네오위즈게임즈 | Method and apparatus for managing characters in online game |
CN102708096A (en) * | 2012-05-29 | 2012-10-03 | 代松 | Network intelligence public sentiment monitoring system based on semantics and work method thereof |
CN104866572A (en) * | 2015-05-22 | 2015-08-26 | 齐鲁工业大学 | Method for clustering network-based short texts |
WO2017014124A1 (en) * | 2015-07-21 | 2017-01-26 | 東亞合成株式会社 | Curable composition |
CN105468713A (en) * | 2015-11-19 | 2016-04-06 | 西安交通大学 | Multi-model fused short text classification method |
CN105512864A (en) * | 2016-01-28 | 2016-04-20 | 丁沂 | Method for automatically acquiring post professional ability requirements based on internet |
CN106570076A (en) * | 2016-10-11 | 2017-04-19 | 深圳大学 | Computer text classification system |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107506389A (en) * | 2017-07-27 | 2017-12-22 | 北京德塔精要信息技术有限公司 | A kind of method and apparatus for extracting position skill requirement |
CN107506389B (en) * | 2017-07-27 | 2020-05-19 | 北京德塔精要信息技术有限公司 | Method and device for extracting job skill requirements |
CN108460699A (en) * | 2017-12-20 | 2018-08-28 | 卓智网络科技有限公司 | Teaching programme optimization method and device |
CN108549640A (en) * | 2018-04-24 | 2018-09-18 | 易联众信息技术股份有限公司 | One kind being based on statistical enterprise name similarity calculating method |
CN110033022A (en) * | 2019-03-08 | 2019-07-19 | 腾讯科技(深圳)有限公司 | Processing method, device and the storage medium of text |
CN109934302A (en) * | 2019-03-23 | 2019-06-25 | 大国创新智能科技(东莞)有限公司 | New category recognition methods and robot system based on fuzzy theory and deep learning |
CN110377907A (en) * | 2019-07-18 | 2019-10-25 | 中科鼎富(北京)科技发展有限公司 | A kind of recruitment information standardized method and device |
CN110377907B (en) * | 2019-07-18 | 2023-09-08 | 鼎富智能科技有限公司 | Recruitment information standardization method and device |
CN112434211A (en) * | 2020-11-04 | 2021-03-02 | 广州视源电子科技股份有限公司 | Data processing method, device, storage medium and equipment |
Also Published As
Publication number | Publication date |
---|---|
CN107194617B (en) | 2020-11-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110597735B (en) | Software defect prediction method for open-source software defect feature deep learning | |
CN107194617A (en) | A kind of app software engineers soft skill categorizing system and method | |
CN109189901B (en) | Method for automatically discovering new classification and corresponding corpus in intelligent customer service system | |
CN109783639B (en) | Mediated case intelligent dispatching method and system based on feature extraction | |
CN112699246B (en) | Domain knowledge pushing method based on knowledge graph | |
CN107122340B (en) | A kind of similarity detection method of the science and technology item return based on synonym analysis | |
CN107315738B (en) | A kind of innovation degree appraisal procedure of text information | |
CN109189942A (en) | A kind of construction method and device of patent data knowledge mapping | |
CN106777043A (en) | A kind of academic resources acquisition methods based on LDA | |
CN109271477A (en) | A kind of method and system by internet building taxonomy library | |
CN107992633A (en) | Electronic document automatic classification method and system based on keyword feature | |
CN102194013A (en) | Domain-knowledge-based short text classification method and text classification system | |
CN107220295A (en) | A kind of people's contradiction reconciles case retrieval and mediation strategy recommends method | |
CN105550269A (en) | Product comment analyzing method and system with learning supervising function | |
CN105160038A (en) | Data analysis method and system based on audit database | |
CN102495892A (en) | Webpage information extraction method | |
US10387805B2 (en) | System and method for ranking news feeds | |
CN107679110A (en) | The method and device of knowledge mapping is improved with reference to text classification and picture attribute extraction | |
CN111737485A (en) | Human-sentry matching method and human-sentry matching system based on knowledge graph and deep learning | |
CN108595525A (en) | A kind of lawyer's information processing method and system | |
CN109918648B (en) | Rumor depth detection method based on dynamic sliding window feature score | |
CN115952292B (en) | Multi-label classification method, apparatus and computer readable medium | |
CN108681548A (en) | A kind of lawyer's information processing method and system | |
CN109783619A (en) | A kind of data filtering method for digging | |
Scrivens et al. | Sentiment-based classification of radical text on the web |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20220126 Address after: 215488 No. 301, building 11, phase II, Taicang University Science Park, No. 27, Zigang Road, science and education new town, Taicang City, Suzhou City, Jiangsu Province Patentee after: Tianhang Changying (Jiangsu) Technology Co.,Ltd. Address before: 100191 No. 37, Haidian District, Beijing, Xueyuan Road Patentee before: BEIHANG University |
|
TR01 | Transfer of patent right |