Preliminary Research on the Knowledge Discovery in Chemistry Based on the Word Embedding

Abstract

Abstract: In the text data, knowledge representation is abstract, unstructured, and latent. It is difficult to find knowledge by co-occurrence, rule or association. Based on Word2Vec, this paper proposes to utilize the WP-Word2Vec model with word attribute, and designs knowledge discovery model based on word vector. Experiments show that the WP-Word2Vec model can integrate word class information into the training word vector, and the similarity in the similar word discovery is improved.

Key words: Morpheme, Word embedding, Word2vec, Knowledge discovery

Wang Xin, Ji Jiuming, Li Nan, Sun Jiqing. Preliminary Research on the Knowledge Discovery in Chemistry Based on the Word Embedding[J]. Libraly Journal, 2018, 37(11): 90-98.

References

[ 1 ] 高劲松, 李迎迎, 刘龙, 等. 基于关联数据的知识发现模型构建研究[J]. 情报科学, 2016(6):10-13,18.
[ 2 ] Amancio D R.Network analysis of named entity cooccurrencesin written texts[J]. 2016, 114(5):58005.
[ 3 ] Farid D M, Al-Mamun M A, Manderick B, etal.An adaptive rule-based classifier for mining bigbiological data[J]. Expert Systems with Applications,2016(64):305-316.
[ 4 ] 王培吉, 赵玉琳, 吕剑峰. 基于Apriori算法的关联规则数据挖掘研究[J]. 统计与决策, 2011(23):19-21.
[ 5 ] Ho Chung Wu, Robert Wing Pong Luk, Kam FaiWong, et al. Interpreting TF-IDF term weights amaking relevance decisions[J]. ACM Transactions onInformation Systems(TOIS), 2008, 26(3):1—37.
[ 6 ] 刘惠. 知识发现理论与方法及其应用的研究[D].大连：大连理工大学, 2004.
[ 7 ] Wu X, Zhu X, Wu G Q, et al. Data Mining with BigData[J]. IEEE Transactions on Knowledge & DataEngineering, 2014, 26(1):97-107.
[ 8 ] 蔡云雷. 基于潜在语义分析的专利文本分类技术研究[D]. 沈阳：沈阳航空航天大学, 2011.
[ 9 ] 孟宪军. 互联网文本聚类与检索技术研究[D]. 哈尔滨：哈尔滨工业大学, 2009.
[10] 黎旭荣. Forestnet: 一种结合深度学习和决策树集成的方法[D]. 广州：中山大学, 2015.
[11] 宋艳雪. 基于关联规则和图排序的句子情感倾向性研究[D]. 大连：大连理工大学, 2011.
[12] 贾修一. 基于语义的粗糙集理论和模型研究[D].南京：南京大学, 2011.
[13] 张志平. 基于遗传算法的汉语基本词汇自动提取研究[D]. 呼和浩特：内蒙古师范大学, 2007.
[14] Santos C N D, Gattit M. Deep Convolutional NeuralNetworks for Sentiment Analysis of Short Texts[C]//International Conference on ComputationalLinguistics. 2014.
[15] Bengio Y,Courville A,Vincent P.RepresentationLearning: A Review and New Perspectives[J].IEEE Transactions on Pattern Analysis & MachineIntelligence, 2013, 35(8):1798-1828.
[16] Turian J, Ratinov L, Bengio Y.Word representations:A simple and general method for semi-supervisedlearning[C]//Proc of ACL.Stroudsburg, PA: ACL,2010:384-394.
[17] Manning C D,Raghavan P, Schutze H.Introductionto Information Retrieval [M]. Cambridge, UK:Cambridge University Press, 2008.
[18] 佚名. word2vec中的数学原理详解[EB/OL]. blog.csdn.net/itpus/article/details/37969519/, 2016-10-01.
[19] 王倩倩, 陈荣, 李楠, 等. 面向化学名称的术语知识量计算模型研究[J]. 图书馆杂志, 2015(10):59-62.
[20] 李楠, 孙济庆,吉久明.汉语词素语义与知识发现研究初探[J]. 图书情报工作, 2013, 57(17):109-113.

[1]	Ruan Guangce, Xie Fan, Tu Shiwen. Application Research Based on Word2vec Diversity in Library Recommender System [J]. Libraly Journal, 2020, 39(3): 124-132.
[2]	Howell Y. Wang, Fred Y. Ye. A Concise Discussion on the Theoretical and Technical Foundations of Data Science from the View of Information Science [J]. Libraly Journal, 2020, 39(12): 20-28.
[3]	Zhou Qianhao, Dai Zefan, Zhu Yifan, Sheng Jiaqi, Xu Xin. Design and Implementation of inBooks Digital Humanities Tool: A WeChat Mini Program Based on the Open Data of Shanghai Library [J]. Libraly Journal, 2019, 38(2): 41-48.
[4]	Hu Yuting, Huang Chen. Research on Keywords Association Recommendation of Knowledge Discovery System [J]. Libraly Journal, 2018, 37(2): 71-77.
[5]	Wang Qianqian, Chen Rong,Li Nan, Sun Jiqing. Research on Knowledge Quantity Calculation Model forTerminology Oriented to Chemical Names [J]. LIBRARY JOURNAL, 2015, 34(10): 59-62.