| 出内 将夫 | D, 中間発表 | 自然言語処理学 | 渡辺 太郎, | 中村 哲, | 進藤 裕之 | |
|
title: Applying Natural Language Processing Technologies to Specific Business Area
abstract: In recent years, AI technology have begun to take place in various business areas, and applying natural language processing technology are increasing. However, in order to apply natural language processing to a specific business area, it is necessary to overcome a lack of data and a specific representation in the area. I presentate that utilizing machine translation technology can be applying technologies and methodologies for business application of natural language processing. language of the presentation: Japanese 発表題目: *** 特定ビジネス業務領域への自然言語処理技術適用 *** 発表概要: *** 昨今、AI技術のビジネス適用があらゆるビジネス領域で行われ始めており、自然言語処理技術のビジネス適用が求められることも増えてきている。しかし、特定のビジネス領域に自然言語処理を適用するには、データ不足やその領域特有の表現に対応する必要がある。機械翻訳を応用した自然言語処理技術のビジネス適用を促進できる応用技術を開発する。 *** | ||||||
| 山元 勇輝 | D, 中間発表 | 自然言語処理学 | 渡辺 太郎, | 中村 哲, | 進藤 裕之, | 大内 啓樹 |
|
title: Complex Sentence Parsing with Abstract Meaning Representation
abstract: Abstract Meaning Representation is a sentence-level meaning representation based on predicate argument structure. One of the challenges found in AMR parsing task is to capture the structure of complex sentences that expresses the relation between predicates. In this presentation, we show our pioneering approach to the problem by capturing skeletal structure of an input complex sentence in advance to full parsing. The approach is a two-step process of dependency matching of complex sentence patterns and semantic disambiguation of output skeletons. With the help of weakly supervised data, we establish a novel baseline of 75.65% on Macro Average and 83.94% on Micro Average for semantic disambiguation task of complex sentence constructions in AMR framework. We also discuss our future plans to integrate the skeletons into AMR parsers. language of the presentation: Japanese | ||||||
| CHOI JUNGMIN | D, 中間発表 | 自然言語処理学 | 渡辺 太郎, | 中村 哲, | 進藤 裕之 | |
|
title: Towards Information Retrieval in Legal Domain
abstract: Law is one of the areas where application of NLP will make a significant impact in the society. My research aims to automatically retrieve relevant law articles to issues described in ordinary language. language of the presentation: English | ||||||