• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2013, Vol. 35 ›› Issue (9): 141-145.

• 论文 • 上一篇    下一篇

网络新词识别算法研究

刘哲1,黄永峰1,罗芳2,陈跻2,王丙坤1   

  1. (1.清华大学电子工程系,北京 100084;2.华中科技大学计算机学院,湖北 武汉 430074)
  • 收稿日期:2013-04-09 修回日期:2013-07-07 出版日期:2013-09-25 发布日期:2013-09-25
  • 基金资助:

    国家863计划资助项目(2012AA011004);清华大学自主科研项目基金(20111081023)

Research on algorithm for networks new words identification       

LIU Zhe1,HUANG Yongfeng1,LUO Fang2,CHEN Ji2,WANG Bingkun1   

  1. (1.Department of Electronic Engineering,Tsinghua University,Beijing 100084;
    2.School of Computer Science and Technology,Huazhong University of Science and Technology,Wuhan 430074,China)
  • Received:2013-04-09 Revised:2013-07-07 Online:2013-09-25 Published:2013-09-25

摘要:

针对社交网络新词识别过程中“旧词新义”所引起的语义模糊问题,提出了网络新词识别算法。通过检测词语频度变化、共现词语分布一致性、情感倾向性迁移三项指标综合分析判断网络新词产生变化的规律特点,从而设计一种网络新词识别算法。最后以实验验证了该算法对提高现有系统网络新词识别准确率的可行性和有效性。

关键词: 社交网络, 新词识别, 准确率

Abstract:

Aiming at the problem of ambiguity caused by Neologism in new words identification in social networks, an algorithm for networks new words identification is proposed. The algorithm analyzes the changing patterns and features of networks new words by three indicators: frequency gain, cooccurrence distribution consistency and emotional tendencies transference. Finally, real experiments validate the feasibility and effectiveness of the algorithm improving existing systems of new words identification.

Key words: social networks;new words identification;precision