| 新任助教講演会(Lectures from New Assistant Professors) |
|---|
| 日時： | 平成28年4月27日(水)3限 (13:30 -- 15:00), 2016/04/27, Wednesday |
|---|---|
| 場所(Location)： | L1 |
| 司会(Chair)： | 畑 秀明 (Hideaki HATA) |
| 講演者(Presenter)： | 能地 宏 (Hiroshi NOJI), 自然言語処理学研究室 (Computational Linguistics Lab.) |
|---|---|
| 題目(Title)： | 言語の普遍性に基づく教師なし構文解析 Unsupervised Syntactic Parsing based on Language Universality |
| 概要(Abstract)： |
教師なし構文解析は自然言語処理において古くから存在する問題であると同時に、最も困難なタスクの一つとして認識されている。構文解析とは、文が与えられたときにその構造、例えば文内の単語間の係り関係を推定する問題である。自然言語処理において、教師あり構文解析はある程度の成功を収めているのに対し、明示的な正解データを与えずに学習を行う教師なし構文解析の精度は依然非常に低い。本講演では、自然言語処理、構文解析、そして我々のアプローチである、言語に普遍的に存在する構造上の偏りを利用した教師なし構文解析の手法について紹介を行う。 Unsupervised syntactic parsing has been recognized as one of the most challenging problems in natural language processing (NLP) for a long time. The goal of syntactic parsing is to predict the syntactic structure of a sentence, e.g., word to word dependencies on a sentence. While supervised syntactic parsing has been quite successful, the accuracy of unsupervised parsing is still very bad. In this talk, I will introduce natural language processing, syntactic, parsing, and our approach to unsupervised syntactic parsing exploiting universal syntactic biases, which may exist across all languages in the world. |
| 講演者(Presenter)： | 崔 恩瀞 (Eunjong CHOI), ソフトウェア設計学研究室 (Software Design and Analysis Lab.) |
|---|---|
| 題目(Title)： | Detection and management of code clones in large-scale software systems |
| 概要(Abstract)： | Large-scale software systems have complicated the maintenance process by increasing efforts such as understanding of the existing source code. A code clone (i.e., a code fragment that has other code fragments identical or similar to it in the source code) a well-known factor hindering the software maintenance task. For example, when changes to code clones in a clone set (i.e., a set of code clones that are identical or similar to each other) are inconsistent, the developer needs to identify inconsistently changed code clones and apply consistent changes to them. In this talk, I will introduce approaches for detecting and managing code clone in large-scale software systems. |