arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1906.02053cs.CL
收藏

KAS-term:通过监督机器学习从博士论文中提取斯洛文尼亚语术语

KAS-term: Extracting Slovene Terms from Doctoral Theses via Supervised Machine Learning

Jožef Stefan Institute(约热夫·斯特凡研究所);Faculty of Arts, University of Ljubljana(卢布尔雅那大学艺术学院)

Nikola Ljubešić, Darja Fišer, Tomaž Erjavec

摘要:

本文提出了一个数据集和监督学习实验斯洛文尼亚学术文本中的术语提取。通过形态句法模式提取数据集中的候选术语,并由四个注释器对其术语进行注释。数据集上的实验表明,大多数共现统计,应用后的形态句法模式和频率阈值,执行接近随机的,结果可以显着改善相结合,与监督机器学习,所有七个统计措施,包括在数据集中。在多词项上,使用所有统计量的模型获得的AUC为0.736,而最好的单个统计量仅产生AUC 0.590。在众多的候选特征中,仅添加多词形态句法模式信息和候选词的长度就能进一步改善结果。

英文摘要:

This paper presents a dataset and supervised learning experiments for term extraction from Slovene academic texts. Term candidates in the dataset were extracted via morphosyntactic patterns and annotated for their termness by four annotators. Experiments on the dataset show that most co-occurrence statistics, applied after morphosyntactic patterns and a frequency threshold, perform close to random and that the results can be significantly improved by combining, with supervised machine learning, all the seven statistic measures included in the dataset. On multi-word terms the model using all statistics obtains an AUC of 0.736 while the best single statistic produces only AUC 0.590. Among many additional candidate features, only adding multi-word morphosyntactic pattern information and length of the single-word term candidates achieves further improvements of the results.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑