Tamotsu Shirado

Extraction and visualization of numerical and named entity information from a large number of documents

We have developed a system that can semi automatically extract numerical and named entity sets from a large number of Japanese documents and can create various kinds of tables and graphs. In our experiments, our system has semiautomatically created approximately 300 kinds of graphs and tables at precisions of 0.2-0.8 with only two hours of manual preparation from a two-year stack of newspapers articles. Note that these newspaper articles contained a large quantity of data, and all of them could not be read or checked manually in such a short amount of time. From this perspective, we concluded that our system is useful and convenient for extracting information from a large number of documents.

Machine-Learning-Based Transformation of Passive Japanese Sentences into Active by Separating Training Data into Each Input Particle

We developed a new method of transforming Japanese case particles when transforming Japanese passive sentences into active sentences. It separates training data into each input particle and uses machine learning for each particle. We also used numerous rich features for learning. Our method obtained a high rate of accuracy (94.30%). In contrast, a method that did not separate training data for any input particles obtained a lower rate of accuracy (92.00%). In addition, a method that did not have many rich features for learning used in a previous study (Murata and Isahara, 2003) obtained a much lower accuracy rate (89.77%). We confirmed that these improvements were significant through a statistical test. We also conducted experiments utilizing traditional methods using verb dictionaries and manually prepared heuristic rules and confirmed that our method obtained much higher accuracy rates than traditional methods.

A system to indicate honorific misuse in spoken japanese

We developed a computational system to indicate the misuse of honorifics in word form and in performance of expressions in Japanese speech sentences. The misuse in word form was checked by constructing a list of expressions whose word form is bad in terms of honorifics. The misuse in performance was checked by constructing a consistency table. The consistency table defined the consistency between the honorific features of sentences and the social relationship among the people involved in the sentence. The social relationship was represented by combinations of [the number of people involved in the sentence] × [relative social position among the people] × [in-group/out-group relationship among the people]. The consistency table was obtained by using a machine learning technique. The proposed system was verified using test data prepared by the authors and also by third-party linguistic researchers. The results showed that the system was able to discriminate between the correct and the incorrect honorific sentences in all but a few cases. Furthermore, differences in the educational importance among the norms used in the system were revealed based on experiments using sentences written by people who are not linguistic experts.

Acquiring Concept Hierarchies of Adjectives from Corpora: Towards Construction of Ontology of Adjectives from a real data

本研究は, 実データに基づいた形容詞の観点からみた概念体系の自動構築をめざし, その一環として, 形容詞概念の階層関係構築に焦点を当てたものである.包含関係の尺度によって上位下位関係を求め, その単語間の上位下位関係に基づき概念階層を自動構築した.結果については, カバー率などの表層的な面と, 階層の作られ方についての質的な面での評価を行った.階層の質的な面での評価にあたって, 概念の継承関係と事例 (形容詞) の各概念の成員としての連続性という観点から心理実験を行い, 既存の人手によって作られたEDR辞書の階層と比較した.実験手法はScheffeの一対比較法を用いた.その結果, 自動構築がよい, あるいは既存の辞書と有意差がないと判断された階層は, 全体の43%となった.抽出した概念数の不足や階層構築の際の問題点など課題も抱えているが, 自動生成の階層が既存辞書の階層に対して, その結果の半分弱の階層で問題を提起するという意味で, ベースラインとなる数値と考える.

Decreased-Adding-Based Question-Answering System Using Simple Connection Method for Contextual Questions

We participated in NTCIR QAC-1, QAC-2, and QAC-3, which were evaluation workshops for answering questions held by the National Institute of Informatics of Japan and studied question-answering systems for contextual questions. Contextual questions are defined as a series of questions with contexts. For example, the first question is “What is the capital of Japan?” and the one succeeding is related to the first such as “What was it called in ancient times?”. Contextual question-answering can be considered interactive. This paper describes our system for contextual questions. Our question-answering system used numerous effective methods to answer questions. We used a method of connecting question sentences to handle contextual questions. We also used a method of using multiple documents to obtain more accurate answers. We confirmed that these methods were effective in the experiments. Our question-answering system obtained the second best accuracy in QAC-1 and the best accuracy in both QAC-2 and QAC-3 for contextual question answering. It is thus a high-performance system.

Acquiring concept hierarchies of adjectives from corpora

We describe a method to acquire a distribution of the concepts of adjectives automatically by using a self-organizing map and a directional similarity measure. A means of evaluating concept hierarchies of adjectives extracted automatically from corpora is elucidated. We used Scheffes method of paired comparison to test experimentally the validity of hierarchies thus obtained with human intuition and found that our method was effective for 43% of the hierarchies considered.

A Computational Model of Change in Politeness with the Addition of Word Endings

Many polite expressions can be synthesized using simple expressions in Japanese. It is expected that such syntheses are dominated by psychological mechanisms in politeness. This study reveals one such mechanism by using a computational model for politeness in Japanese to describe changes in politeness through the addition of word endings. In this model, two stochastic features are assumed: (1) For each expression, a situation in which the expression would be used can be represented by a probability distribution of the politeness value in a psychological space, and (2) For each word ending e, a probability distribution exists in a one-dimensional psychometrical space of politeness, where the distribution represents the ideal distribution of the most suitable (or ideal) expression to which the word ending e would be added. The change in politeness that arise from the addition of word endings is calculated by the difference between these probability distributions. The information theory is utilized in this calculation. A linear relationship is expected to exist between the change in politeness that arise from the addition of word endings to expressions and the politeness of the original expressions. Psychological experiments were performed to verify the validity of the model. The degree of politeness of expressions was evaluated using Thurstones scaling. Experimental results show the expected linearity to be true which qualitatively verifies the validity of the model. The results are also discussed in terms of linguistic intuition.

A computational model for politeness of expressions

待遇表現の丁寧さの計算モデルとして, 待遇表現に語尾を付加した際の待遇値 (待遇表現の丁寧さの度合い) の変化に関する定量的なモデルを提案した. このモデルでは (1) それぞれの待遇表現に対し, その表現が用いられるべき状況が待遇値に関する正規分布として表される, (2) それぞれの語尾に対し, その語尾が付加される待遇表現が用いられるべき状況が待遇値に関する正規分布として表される, というふたつの仮定を立て, 待遇表現に語尾を付加した際に得られる情報量を定義した. そして更に, 語尾の付加による待遇値の変化量は, 付加の際得られる情報量に関する一次式で表すことができる, という仮定を立て, 語尾の付加による待遇値の変化量を, 語尾が付加される前の待遇表現に対する待遇値の関数として定義した. このモデルの妥当性を検証するため, ふたつの異なった発話状況において用いられる待遇表現のグループそれぞれに対し, 語尾の付加による待遇値変化を求める心理実験を行った. その結果, いずれのグループにおいても語尾の付加による待遇値変化は, 提案されたモデルによって予測された傾向に従い, モデルの妥当性が支持された.