A Skill Vocabulary Normalization Method for Recruitment Webpage Combing Literal and Context Similarity
Received date: 2018-09-17
Online published: 2025-04-28
[Purpose/significance] This paper proposes a skill vocabulary normalization method for recruitment webpages, it aims to solve the problem that many English skill word spelling errors exist in the recruitment webpages. [Method/process] The method combines literal similarity and context similarity to measure the similarity of skill word and form a similar skill word network to normalize the skill words in the recruitment webpages. [Result/conclusion] One week’s computer recruitment information was obtained from domestic mainstream recruitment website 51job to evaluate the proposed method. The experiment results show that the proposed method can automatically, accurately and quickly normalize the skill vocabulary in the recruitment webpages.
Key words: recruitment webpage; skill; lexical normalization
Sun Yu , Jiang Jinde . A Skill Vocabulary Normalization Method for Recruitment Webpage Combing Literal and Context Similarity[J]. Knowledge Management Forum, 2018 , 3(6) : 325 -334 . DOI: 10.13266/j.issn.2095-5472.2018.031

。
。笔者定义两个上下文相似性conSetSim和conFreSim,其中conSetSim的定义如下:
;同样地,S j 表示w j 的上下文技能词语集。由公式(3)可知,当技能词语w i 与技能词语w j 的上下文技能词语集中相同的技能词语越多的时候,则w i 与w j 越相似,越可能是表示相同概念的技能词语。
表示词语w在词语w i 的岗位上下文词语文本集D i 中出现的次数;类似地,
表示词语w在词语w j 的岗位上下文词语文本D j 中出现的次数。由公式(4)可知,conFreSim与conSetSim的相同之处都是通过上下文词语集来度量词语相似度,但是conFreSim还考虑了上下文技能词语集中每个技能词语出现的次数。


表1 相似方法方法 |
| 序号 | 比较方法 | 计算方法 |
| 第1组 | strSim | 公式(2) |
| conSetSim | 公式(3) | |
| conFreSim | 公式(4) | |
| 第2组 | strSim | 公式(2) |
| conSetSim | 公式(3) | |
| strSim-conSetSim | 公式(5) | |
| 第3组 | strSim | 公式(2) |
| conFreSim | 公式(4) | |
| strSim-conFreSim | 公式(6) | |
| 第4组 | strSim | 公式(2) |
| strSim-conSetSim | 公式(5) | |
| strSim-conFreSim | 公式(6) |
表2 实例分析 |
| 序号 | 词 | 词 | strSim | conFreSim | strSim-conFreSim |
| 1 | oracle | oralce | 6.00 | 3.15 | 18.90 |
| 2 | fireworks | framework | 6.33 | 0.59 | 3.73 |
| 3 | mapreduce | hadoop | 2.14 | 2.67 | 5.71 |
表3 词语规范化实例 |
| 序号 | 技能词语集 | 规范化词 | 序号 | 技能词语集 | 规范化词 |
| 1 | android andriod anroid andorid andoird androin andiod androd androrid androi andoid | android | 4 | jquery jquey jqueyr jqurey jqeury juqery juqery jqery jquer | jquery |
| 2 | eclipse eclips eclise eclipe elipse | eclipse | 5 | hibernate hibenate herbinate hibernat hiberate hibernet | hibernate |
| 3 | mybatis mybaits mybaties myibatis mybaitis mybiatis mybtis mybastis mbatis mybits | mybatis | 6 | struts structs stucts struct strut strus strust | struts |
| 4 | javascript javascrip javascrpt jscript javasript javasscript javascipt javacript javascripts javascrpit javasprict javascrtip javscript javescript javascritp | javascript | 8 | oracle oralce oracel orcale orcle orcal oraccle oracl orale oracal orace | oracle |
孙瑜: 提出研究思路,实施实验,撰写论文;
姜金德: 分析实验数据,修改论文,进行理论指导。
| [1] |
WOWCKO I. Skills and vacancy analysis with data mining techniques[J]. Informatics, 2015, 2(4):31-49.
|
| [2] |
KIM J Y, LEE C K. An empirical analysis of requirements for data scientists using online job postings[J]. International journal of software engineering and its application, 2016, 10(4): 161-172.
|
| [3] |
夏火松, 潘筱听. 基于Python挖掘的大数据学术研究与人才需求的关系研究[J]. 信息资源管理学报, 2017, 7(1): 4-12.
|
| [4] |
詹川. 基于文本挖掘的专业人才技能需求分析——以电子商务专业为例[J]. 图书馆论坛, 2017, 5(1): 116-123.
|
| [5] |
夏立新, 楚林, 王忠义,等. 基于网络文本挖掘的就业知识需求关系构建[J]. 图书情报知识, 2016, 169(1):94-100.
|
| [6] |
刘睿伦, 叶文豪, 高瑞卿, 等. 基于大数据岗位需求的文本聚类研究[J]. 数据分析与知识发现, 2017, 12(12): 32-40.
|
| [7] |
LUO Q, ZHAO M, JAVED F, et al. Macau: large-scale skill sense disambiguation in the online recruitment domain[C]// IEEE international conference on big data. Piscataway: IEEE, 2015:1324-1329.
|
| [8] |
BRILL E, MOORE R C. An improved error model for noisy channel spelling correction[C]// Meeting of the Association for Computational Linguistics. Piscataway: IEEE, 2000:286-293.
|
| [9] |
TOUTANOVA K, MOORE R C. Pronunciation modeling for improved spelling correction[C]// Proceedings of annual meeting of the Association for Computational Linguistics. Stroudsburg :Association for Computational Linguistics, 2002:144-151.
|
| [10] |
CHOUDHURY M, SARAF R, JAIN V, et al. Investigation and modeling of the structure of texting language[J]. International journal of document analysis & recognition, 2007, 10(3):157-174.
|
| [11] |
LIU F, WENG F, WANG B, et al. Insertion, deletion, or substitution? normalizing text messages without pre-categorization nor supervision[J]. 2012, 15(2):71-76.
|
| [12] |
AW A T, ZHANG M, XIAO J, et al. A phrase-based statistical model for SMS text normalization.[C]//International conference on computational linguistics and meeting of the Association for Computational Linguistics. New York: ACM, 2006: 17-21.
|
| [13] |
PENNELL D L, LIU Y. A character-level machine translation approach for normalization of sms abbreviations[J]. Natural language processing, 2011,20(2):974-982.
|
| [14] |
COOK P, STEVENSON S. An unsupervised model for text message normalization[M]. Stroudsburg:Association for computational linguistics, 2009.
|
| [15] |
SRIDHAR V K R. Unsupervised text normalization using distributed representations of words and phrases[C]// The workshop on vector space modeling for natural language processing. Piscataway: IEEE, 2015:8-16.
|
| [16] |
施振辉, 沙灜, 梁棋,等. 基于字词联合的变体词规范化研究[J]. 计算机系统应用, 2017, 26(10):29-35.
|
| [17] |
罗延根, 李晓, 蒋同海,等. 基于词向量的维吾尔语词项归一化方法[J]. 计算机工程, 2018(2):220-225.
|
| [18] |
DAMERAU F J. A technique for computer detection and correction of spelling errors[J]. Communications of the ACM, 1964, 7(3):171-176.
|
/
| 〈 |
|
〉 |