人工知能学会論文誌
Online ISSN : 1346-8030
Print ISSN : 1346-0714
検索
OR
閲覧
検索
22 巻 , 6 号
選択された号の論文の7件中1~7を表示しています
    • |<
    • <
    • 1
    • >
    • >|
論文
  • 平尾 努, 鈴木 潤, 磯崎 秀樹
    22 巻 (2007) 6 号 p. 574-584
    公開日: 2007/08/24
    ジャーナル フリー
    In the study of automatic summarization, the main research topic was `important sentence extraction' but nowadays `sentence compression' is a hot research topic. Conventional sentence compression methods usually transform a given sentence into a parse tree or a dependency tree, and modify them to get a shorter sentence. However, this method is sometimes too rigid. In this paper, we regard sentence compression as an combinatorial optimization problem that extracts an optimal subsequence of words. Hori et al. also proposed a similar method, but they used only a small number of features and their weights were tuned by hand. We introduce a large number of features such as part-of-speech bigrams and word position in the sentence. Furthermore, we train the system by discriminative learning. According to our experiments, our method obtained better score than other methods with statistical significance.
    抄録全体を表示
  • 中渡瀬 秀一, 相澤 彰子
    22 巻 (2007) 6 号 p. 585-594
    公開日: 2007/09/11
    ジャーナル フリー
    When extracting IS-A relationship between nouns from texts, the majority of conventional methods rely on specific expression patterns such as ``A such as B''. However, those expressions cover only a restricted subset of nouns comparing to the entire set contained in the corpus. Based on the observation, this paper investigates a method for identifying IS-A relationship that does not depend on particular expression patterns. In the paper, we first clarify our notion of ``IS-A relationship'' as a subset relation between the instance sets represented by two distinct nouns. This follows the assumption that hyponyms that appear in texts are most likely substituted by their hypernyms while the substitutionality does not hold in the opposite cases. Based on this, we propose a new method for detecting possible hypernym-hyponym pairs by examining the substitutionality of the two nouns using their co-occurring verbs and their dependency/case structure. Then, the effectiveness of the proposed method is confirmed through some experimental study. In the experiments, 47 target words were first selected from the Word List by Semantic Principles. Next, using 11-years newspaper articles as input texts, a list of candidate hyponyms was generated for each selected word using the proposed method. Also, a conventional pattern-based method was applied to the same newspaper articles to obtain hypernym-hyponym candidate pairs. Then, the candidate lists were examined by a human reviewer. Through the experiments, it is confirmed that 94.3% pairs obtained by our method were not covered by the conventional method. Also, the average accuracy was about 36% for the top 200 ranked candidates of the proposed method, the performance of which was quite comparable to the 31% of the existing method.
    抄録全体を表示
  • Haiyang Yu, Seizaburo Niitsuma
    22 巻 (2007) 6 号 p. 595-603
    公開日: 2007/09/11
    ジャーナル フリー
    Although there were in-depth discussions in the 1970s on the question of whether the human visual system contains 'curvature detectors' or contour detectors, which respond to the tangents of curves [Blakemore 74], they yielded no definite conclusions. Until now, the end-stopped cell model of curve detection has been the predominant one [Dobbins 87]. However, this model detects curvature with a low degree of accuracy, so a better model is required. Long ago, people discovered that the human brain is a network of numerous neurons. The hypothesis of achieving a highly accurate calculation of curvature through a network composed of biological elements (simple cells) is readily accepted. However, neither Blakemore et al. nor Dobbins et al. explain the function of simple cells in the calculation of curvature. This article illustrates the function of simple cells in calculating curvature. Moreover, in this article we attempt to construct a computational model for describing the mechanism for calculating curvatures along suggestions of Blakemore and Over. This model gives a key for answering to a question why the Helmholtz irradiation disappears when two squares are replaced by two circles.
    抄録全体を表示
  • 篠原 修二, 入部 百合絵, 河嶋 宏明, 桂田 浩一, 新田 恒雄
    22 巻 (2007) 6 号 p. 604-611
    公開日: 2007/09/21
    ジャーナル フリー
    Recently, e-learning systems for self-learning with various types of retrieval functions have been developed. This paper describes a method for keyword extraction using retrieval information stored from many students through the retrieval functions. Firstly, we show that (1) teachers tend to consider technical terms as important, while students unfamiliar with the technical terms tend to retrieve the terms, therefore (2) there is a clear correlation between keywords extracted by the teachers and the retrieval words by the students. Secondly, we propse a method utilizing retrieval information from the students for keyword extraction, and show that the mehod can achieve quite better performance than a method extracting keywords using only lecture information.
    抄録全体を表示
  • 松山 科子, 寺野 隆雄
    22 巻 (2007) 6 号 p. 612-620
    公開日: 2007/09/21
    ジャーナル フリー
    This paper discusses dynamic properties of human communications networks, which appears as a result of informationexchanges among people. We propose agent-based simulation (ABS) to examine implicit mechanisms behind the dynamics. The ABS enables us to reveal the characteristics and the differences of the networks regarding the specific communicationgroups. We perform experiments on the ABS with activity data from questionnaires survey and with virtual data which isdifferent from the activity data. We compare the difference between them and show the effectiveness of the ABS through theexperiments.
    抄録全体を表示
  • 大村 英史, 片上 大輔, 新田 克己
    22 巻 (2007) 6 号 p. 621-630
    公開日: 2007/10/10
    ジャーナル フリー
    In this paper, we propose a learning method for an agent to interact with other agents effectively. This method, MULA-C, improves efficiency of the learning, by clustering agents, and influences the learning experience of one agent to other agents which belong to the same cluster. Similarity among agents is evaluated by similarity among Q-values of agents. We give the detail explanation of learning method of MULA-C, and present the result of experiments which shows the effectiveness of MULA-C.
    抄録全体を表示
正誤表.
    • |<
    • <
    • 1
    • >
    • >|
feedback
Top