研究者詳細

研究発表
分割表示   全件表示 >>

5 件中 1 - 5 件目

年度
Year
題目又はセッション名
Title or Name of Session
細目
Authorship
発表年月(日)
Date
発表学会等名称 Name, etc. of the conference at which the presentation is to be given, 主催者名称 Organizer, 掲載雑誌名等 Publishing Magazine,発行所 Publisher,巻/号 Vol./no.,頁数 Page nos.
2017  コーパスを利用したレポート自動採点品質の向上  共同  2017/09/12 
FIT2017 第16回情報科学技術フォーラム  , 情報処理学会   

概要(Abstract) 筆者らは,レポート作成指導,教員の採点の厳正化,負担軽減を目的として,ルーブリックに基づくLMS上の自動採点システムを構築している.「表計算ソフトによるデータ分析」をテーマとするレポートを対象に実験したところ,SVMによる総合成績レベルの分類精度は53.6%と低かった.これは分類器作成のための学習データの件数が少なく,データ間に大きなレベル格差がないことが一つの要因であると考えられる.そこで,複数ジャンルの言語コーパスから「成績レベルを網羅する学習データ」を抽出し,分類精度を高めるために取り組んだ結果を報告する.
(山本 恵,梅村 信夫,河野 浩之) 

備考(Remarks)  

2017  Automated Essay Scoring System Based on Rubric  共同  2017/07/09 
ACIS ACIT/BCD/CSII Conferences  , International Association for Computer and Information Science (ACIS)    

概要(Abstract) In this paper, we propose an architecture of automated essay scoring system based on rubric, which combines automated scoring with human scoring. Rubrics are valid criteria for grading students’ essays. Our proposed rubric has five evaluation viewpoints “Contents, Structure, Evidence, Style, and Skill” and 25 evaluation items which are subdivided viewpoints. The system is cloud-based application and consists of several tools such as Moodle, R, MeCab, and RedPen. At first, the system automatically scores 11 items included in the Style and Skill such as sentence style, syntax, usage, readability, lexical richness, and so on. Then it predicts scores of Style and Skill from these items’ scores by multiple regression models. It also predicts Contents’ score by the cosine similarity between topics and descriptions. Moreover, our system classifies into five grades “A+, A, B, C, D” as useful information for teachers, by using machine learning techniques such as support vector machine. We try to improve automated scoring algorithms and a variety of input essays in order to improve the accuracy of classification over 90%.
(Megumi Yamamoto, Nobuo Umemura, Hiroyuki Kawano) 

備考(Remarks)  

2015  Cloud Architecture of Traffic Mining Systems under Privacy Preservation  単独  2015/07/15 
27th European Conference on Operational Research (EURO2015)   

概要(Abstract) n Japan, there exist several independent traffic management systems, which have been developed by different expressway companies and organizations. These management systems partially exchange traffic data each other. During several decades, in order to increase the accuracy of traffic monitoring data, various devices and sensors have been developed, such as supersonic wave detectors, image sensors, ETC and others. Currently, it is becoming important to directly monitor traffic data by probe vehicles, including speed, acceleration, location, and other attributes. In order to analyze traffic data dynamically, it is important to collect probe vehicle data widely and totally. But there are privacy issues to probe vehicle data, it is so hard to directly store data into traffic management systems on commercial public clouds. In this presentation, we propose a cloud architecture of traffic mining system under privacy preserving conditions. Secondly, we try to implement a
prototype system of privacy preserving clouds by using CMS software and GPL software "HElib’, which is a fully homomorphic encryption library in C++ and the NTL mathematical library. Finally, we evaluate performance of traffic data mining in our proposed cloud architecture. As a result, in the near future, different organizations will be able to exchange traffic data by using data cloud with privacy preservation. 

備考(Remarks) http://www.theorsociety.com/ygu575kjg/EURO2015_Programme_Handbook_FINAL_for_USB.pdf 

2014  Advanced traffic monitoring system by probe vehicles under privacy preservation  単独  2014/07/17 
20th Conference of the International Federation of Operational Research Societies.  , IFORS   

概要(Abstract) It is increasingly important to correctly measure traffic data by probe vehicles, including speed, acceleration, location, and other data. In our previous researches, we propose that it is possible to be accurately monitored vehicle traffic by integration of supersonic wave detection devices and highway patrol vehicles equipped GPS transceivers. In this paper, we discuss accuracy of traffic volume and flows depending on density of probe vehicles. Furthermore, under privacy preserving conditions, we propose an architecture of traffic management system using our proposed methods. 

備考(Remarks)  

2013  学生主導型総括的授業評価の提案:アンケート構成フレームワークの再構築  未設定  2013/09/05 
私立大学情報教育協会 平成25年度教育改革ICT戦略大会  , 私立大学情報教育協会   

概要(Abstract)  

備考(Remarks) http://www.juce.jp/LINK/taikai/13happyo/t_happyo_schedule.pdf
 

Page: [<<PREV] [1] [NEXT>>]