Generating Multi-word Expressions for Learners of English as Second Language Using Latent Semantic Indexing

2015 Fiscal Year Final Research Report

• Project/Area Number: 25330417
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Learning support system
• Research Institution: Tsuda College
• Principal Investigator: Kishi Nobuko 津田塾大学, 学芸学部, 教授 (50245990)
• Co-Investigator(Kenkyū-buntansha): KISHI Yasuhito 神奈川大学, 総合理学研究所, 研究員 (50552999) ; KUSHIMA Chizuko 津田塾大学, 言語文化研究所, 研究員 (80623876) ; TAJIKA Hiroko 津田塾大学, 学芸学部, 教授 (80188268)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 潜在意味解析 / 英語学習 / 学習コンテンツ開発支援 / 語彙学習 / 用例生成

Outline of Final Research Achievements

This project is aimed at creating a system for generating collocations for the learners of English as a foreign language, from a large English text data set. With this system, the user will be able to enter a list of words, and then get a list of sentences or phrases which are used often with the words that they entered. We completed the system by using an algorithm called a latent semantic indexing(LSI), which is well known for finding synonymous or relevant multi-word expressions. However, the system is very slow in processing a large data set such as Wikipedia data and its output often contains several expressions which are hard to understand for beginning learners of English. We plan to work on improving the system and using other algorithms as well as LSI in the next project.

Free Research Field

情報学基礎