2024-03-28T17:14:52Zhttps://ipsj.ixsq.nii.ac.jp/ej/?action=repository_oaipmhoai:ipsj.ixsq.nii.ac.jp:001782662023-04-27T10:00:04Z01164:03500:09054:09094
A Structure-based Method for Mathematical Document ClassificationA Structure-based Method for Mathematical Document Classificationeng検索関連技術http://id.nii.ac.jp/1001/00178178/Technical Reporthttps://ipsj.ixsq.nii.ac.jp/ej/?action=repository_action_common_download&item_id=178266&item_no=1&attribute_id=1&file_no=1Copyright (c) 2017 by the Information Processing Society of JapanTokyo Institute of TechnologyTokyo Institute of TechnologyTokinori, SuzukiAtsushi, FujiiMathematical document classification (MDC) is a task to classify mathematical documents consisting of text and mathematical expressions (ME) to mathematical categories, e.g. probability theory and set theory. This task is an important task for supporting user search on recent wide-spreaded digital libraries and archiving services. Although ME could bring an important information as being in a central part of communication especially in math fields, how to utilize ME for MDC is not matured. In this paper, we propose the classification method based on texts combined with structures of ME, which are expected to reflect mathematical concepts and rules specific to a category. We demonstrate classification results that our proposed method outperforms existing method with state-of-the-art ME modeling on F-measure.Mathematical document classification (MDC) is a task to classify mathematical documents consisting of text and mathematical expressions (ME) to mathematical categories, e.g. probability theory and set theory. This task is an important task for supporting user search on recent wide-spreaded digital libraries and archiving services. Although ME could bring an important information as being in a central part of communication especially in math fields, how to utilize ME for MDC is not matured. In this paper, we propose the classification method based on texts combined with structures of ME, which are expected to reflect mathematical concepts and rules specific to a category. We demonstrate classification results that our proposed method outperforms existing method with state-of-the-art ME modeling on F-measure.AN10114171研究報告情報基礎とアクセス技術(IFAT)2017-IFAT-12511182017-03-032188-88842017-03-02