Research on Appropriate Image Recommendation for Vocabulary Learning using Educational Big Data and Image Analysis

2019 Fiscal Year Final Research Report

• Project/Area Number: 19K20941
• Project/Area Number (Other): 18H05745 (2018)
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund (2019); Single-year Grants (2018)
• Review Section: :Education and related fields
• Research Institution: Tokyo University of Agriculture and Technology (2019); Kyoto University (2018)
• Principal Investigator: Hasnine Nehal 東京農工大学, 工学(系)研究科(研究院), 特任助教 (30827720)
• Project Period (FY): 2018-08-24 – 2020-03-31
• Keywords: ユビキタス学習 / ライフログ画像分析 / 教育ビッグデータ / 語彙学習

Outline of Final Research Achievements

In this research, lifelog images are analyzed for recommending Feature-based Context-specific Appropriate Images (FCAI) images. FCAI images are determined according to the foreign language learners' current context by using educational big data that are collected from a language learning support system. We proposed a Distributed Semantic Model(DSM) for analyzing lifelog images. We also analyzed words, time, place, vocabulary level and images as the educational big data from SCROLL(System for Capturing and Reminding Learning Log) server. The outcomes of this project are- 1) the creation of a small-scale wordbank using NLP techniques, 2) the development of image analysis tools where AI-based methods are leveraged, and 3) the development of recommendation tools called LFO panels that can recommend authentic and partially-authentic logs to foreign language learners.

Free Research Field

教育工学

Academic Significance and Societal Importance of the Research Achievements

この研究は, CALL,mラーニング,ライフログ画像分析,ユビキタス学習,外国語語彙学習など研究分野に大きく貢献します. この研究は, 言語学習におけるapplied AIの採用を促進します. さらに、この研究は、Society 5.0言語学習のギャップを埋めるのに役立ちます. 関連する科学団体（APSEC,SOLAR,AACE, IIAI AAIなど）からこの研究について肯定的なフィードバックを受けました.