Mitsuru Ishizuka

TIST Journal 2014 Journal Article

Intelligent Interface for Textual Attitude Analysis

• Alena Neviarouskaya
• Masaki Aono
• Helmut Prendinger
• Mitsuru Ishizuka

This article describes a novel intelligent interface for attitude sensing in text driven by a robust computational tool for the analysis of fine-grained attitudes (emotions, judgments, and appreciations) expressed in text. The module responsible for textual attitude analysis was developed using a compositional linguistic approach based on the attitude-conveying lexicon, the analysis of syntactic and dependency relations between words in a sentence, the compositionality principle applied at various grammatical levels, the rules elaborated for semantically distinct verb classes, and a method considering the hierarchy of concepts. The performance of this module was evaluated on sentences from personal stories about life experiences. The developed web-based interface supports recognition of nine emotions, positive and negative judgments, and positive and negative appreciations conveyed in text. It allows users to adjust parameters, to enable or disable various functionality components of the algorithm, and to select the format of text annotation and attitude statistics visualization.

AAAI Conference 2012 Conference Paper

Similarity Is Not Entailment — Jointly Learning Similarity Transformation for Textual Entailment

• Ken-ichi Yokote
• Danushka Bollegala
• Mitsuru Ishizuka

Predicting entailment between two given texts is an important task upon which the performance of numerous NLP tasks depend on such as question answering, text summarization, and information extraction. The degree to which two texts are similar has been used extensively as a key feature in much previous work in predicting entailment. However, using similarity scores directly, without proper transformations, results in suboptimal performance. Given a set of lexical similarity measures, we propose a method that jointly learns both (a) a set of non-linear transformation functions for those similarity measures and, (b) the optimal non-linear combination of those transformation functions to predict textual entailment. Our method consistently outperforms numerous baselines, reporting a micro-averaged F-score of 46. 48 on the RTE- 7 benchmark dataset. The proposed method is ranked 2-nd among 33 entailment systems participated in RTE-7, demonstrating its competitiveness over numerous other entailment approaches. Although our method is statistically comparable to the current state-of-the-art, we require less external knowledge resources.

AAAI Conference 2011 Conference Paper

Cross-Language Latent Relational Search: Mapping Knowledge across Languages

• Nguyen Tuan Duc
• Danushka Bollegala
• Mitsuru Ishizuka

Latent relational search (LRS) is a novel approach for mapping knowledge across two domains. Given a source domain knowledge concerning the Moon, “The Moon is a satellite of the Earth”, one can form a question {(Moon, Earth), (Ganymede, ?)} to query an LRS engine for new knowledge in the target domain concerning the Ganymede. An LRS engine relies on some supporting sentences such as “Ganymede is a natural satellite of Jupiter. ” to retrieve and rank “Jupiter” as the first answer. This paper proposes cross-language latent relational search (CLRS) to extend the knowledge mapping capability of LRS from cross-domain knowledge mapping to cross-domain and cross-language knowledge mapping. In CLRS, the supporting sentences for the source pair might be in a different language with that of the target pair. We represent the relation between two entities in an entity pair by lexical patterns of the context surrounding the two entities. We then propose a novel hybrid lexical pattern clustering algorithm to capture the semantic similarity between paraphrased lexical patterns across languages. Experiments on Japanese-English datasets show that the proposed method achieves an MRR of 0. 579 for CLRS task, which is comparable to the MRR of an existing monolingual LRS engine.

IJCAI Conference 2011 Conference Paper

Interest Prediction on Multinomial, Time-Evolving Social Graph

• Nozomi Nori
• Danushka Bollegala
• Mitsuru Ishizuka

We propose a method to predict users' interests in social media, using time-evolving, multinomial relational data. We exploit various actions performed by users, and their preferences to predict user interests. Actions performed by users in social media such as Twitter, Delicious and Facebook have two fundamental properties. (a) User actions can be represented as high-dimensional or multinomial relations - e. g. referring URLs, bookmarking and tagging, clicking a favorite button on a post etc. (b) User actions are time-varying and user-specific - each user has unique preferences that change over time. Consequently, it is appropriate to represent each user's action at some point in time as a multinomial relational data. We propose ActionGraph, a novel graph representation for modeling users' multinomial, time-varying actions. Each user's action at some time point is represented by an action node. ActionGraph is a bipartite graph whose edges connect an action node to its involving entities, referred to as object nodes. Using real-world social media data, we empirically justify the proposed graph structure. Our experimental results show that the proposed ActionGraph improves the accuracy in a user interest prediction task by outperforming several baselines including standard tensor analysis, a previously proposed state-of-the-art LDA-based method and other graph-based variants. Moreover, the proposed method shows robust performances in the presence of sparse data.

IJCAI Conference 2011 Conference Paper

Mining Longitudinal Network for Predicting Company Value

• Yingzi Jin
• Ching-Yung Lin
• Yutaka Matsuo
• Mitsuru Ishizuka

Real-world social networks are dynamic in nature. Companies continue to collaborate, align strategically, acquire, and merge over time, and receive positive/negative impact from other companies. Consequently, their performance changes with time. If one can understand what types of network changes affect a company's value, he/she can predict the future value of the company, grasp industry innovations, and make business more successful. However, it often requires continuous records of relational changes, which are often difficult to track for companies, and the models of mining longitudinal network are quite complicated. In this study, we developed algorithms and a system to infer large-scale evolutionary company networks from public news during 1981--2009. Then, based on how networks change over time, as well as the financial information of the companies, we predicted company profit growth. This is the first study of longitudinal network-mining-based company performance analysis in the literature.

IJCAI Conference 2011 Conference Paper

Relation Adaptation: Learning to Extract Novel Relations with Minimum Supervision

• Danushka Bollegala
• Yutaka Matsuo
• Mitsuru Ishizuka

Extracting the relations that exist between two entities is an important step in numerousWeb-related tasks such as information extraction. A supervised relation extraction system that is trained to extract a particular relation type might not accurately extract a new type of a relation for which it has not been trained. However, it is costly to create training data manually for every new relation type that one might want to extract. We propose a method to adapt an existing relation extraction system to extractnew relation types with minimum supervision. Our proposed method comprises two stages: learning a lower-dimensional projection between different relations, and learning a relational classifier for the target relation type with instance sampling. We evaluate the proposed method using a dataset that contains 2000 instances for 20 different relation types. Our experimental results show that the proposed method achieves a statistically significant macro-average F-score of 62. 77. Moreover, the proposed method outperforms numerous baselines and a previously proposed weakly-supervised relation extraction method.

AAMAS Conference 2008 Conference Paper

Dynamic Bayesian Network Based Interest Estimation for Visual Attentive Presentation Agents

• Boris Brandherm
• Helmut Prendinger
• Mitsuru Ishizuka

In this paper, we report on an interactive system and the results ofa formal user study that was carried out with the aim of comparing two approaches to estimating users’ interest in a multimodal presentation based on their eye gaze. The scenario consists of a virtual showroom where two 3D agents present product items in an entertaining way, and adapt their performance according to users’ (in)attentiveness. In order to infer users’ attention and visual interest with regard to interface objects, our system analyzes eye movements in real-time. Interest detection algorithms used in previous research determine an object of interest based on the time that eye gaze dwells on that object. However, this kind of algorithm does not seem to be well suited for dynamic presentations where the goal is to assess the user’s focus of attention with regard to a dynamically changing presentation. Here, the current context of the object of interest has to be considered, i. e. , whether the visual object is part of (or contributes to) the current presentation content or not. Therefore, we propose to estimate the interest (or non-interest) of a user by means of dynamic Bayesian networks that may take into account the current context of the attention receiving object. In this way, the presentation agents can provide timely and appropriate response. The benefits of our approach will be demonstrated both theoretically and empirically.

ECAI Conference 2008 Conference Paper

WWW sits the SAT: Measuring Relational Similarity on the Web

• Danushka Bollegala
• Yutaka Matsuo
• Mitsuru Ishizuka

Measuring relational similarity between words is important in numerous natural language processing tasks such as solving analogy questions and classifying noun-modifier relations. We propose a method to measure the similarity between semantic relations that hold between two pairs of words using a web search engine. First, each pair of words is represented by a vector of automatically extracted lexical patterns. Then a Support Vector Machine is trained to recognize word pairs with similar semantic relations. We evaluate the proposed method on SAT multiple-choice word-analogy questions. The proposed method achieves a score of 40% which is comparable with relational similarity measures which use manually created resources such as WordNet. The proposed method significantly reduces the time taken by previously proposed computationally intensive methods, such as latent relational analysis, to process 374 analogy questions from 8 days to less than 6 hours.

IJCAI Conference 2007 Conference Paper

• Junichiro Mori
• Yutaka Matsuo
• Mitsuru Ishizuka

Social networks have recently garnered considerable interest. With the intention of utilizing social networks for the Semantic Web, several studies have examined automatic extraction of social networks. However, most methods have addressed extraction of the strength of relations. Our goal is extracting the underlying relations between entities that are embedded in social networks. To this end, we propose a method that automatically extracts labels that describe relations among entities. Fundamentally, the method clusters similar entity pairs according to their collective contexts in Web documents. The descriptive labels for relations are obtained from results of clustering. The proposed method is entirely unsupervised and is easily incorporated with existing social network extraction methods. Our experiments conducted on entities in researcher social networks and political social networks achieved clustering with high precision and recall. The results showed that our method is able to extract appropriate relation labels to represent relations among entities in the social networks.

IROS Conference 2006 Conference Paper

Comparison of a Humanoid Robot and an On-Screen Agent as Presenters to Audiences

• Johane Takeuchi
• Kushida Kazutaka
• Yoshitaka Nishimura
• Hiroshi Dohi
• Mitsuru Ishizuka
• Mikio Nakano
• Hiroshi Tsujino

Both on-screen agents and humanoid robots are increasingly used as human-computer interfaces. This study evaluates an on-screen agent and a humanoid robot in the task of one-sided presentations. We compared the participant's subjective impressions of nearly identical presentation contents performed by each presenter. The results derived by the semantic differential (SD) method and the direct questioning show that each presenter has different functional advantages. We infer that on-screen agent and robot can complement each other in presentations

ECAI Conference 2006 Conference Paper

Disambiguating Personal Names on the Web Using Automatically Extracted Key Phrases

• Danushka Bollegala
• Yutaka Matsuo
• Mitsuru Ishizuka

When you search for information regarding a particular person on the web, a search engine returns many pages. Some of these pages may be for people with the same name. How can we disambiguate these different people with the same name? This paper presents an unsupervised algorithm which produces unique phrases to disambiguate different people with the same name (i. e. namesakes). Our algorithm takes in a personal name and outputs multiple sets of phrases which uniquely identify the different namesakes on the web. These phrases could then be added to the query to narrow down the search to a specific namesake. We evaluated the algorithm on a collection of documents retreived from the Web. Experimental results show a significant improvement over the existing methods proposed for this task.

ECAI Conference 2004 Conference Paper

Finding Social Network for Trust Calculation

• Yutaka Matsuo
• Hironori Tomobe
• Kôiti Hasida
• Mitsuru Ishizuka

AAAI Conference 1999 Conference Paper

Qualifying the Expressivity/Efficiency Tradeoff: Reformation-Based Diagnosis

• Helmut Prendinger
• Mitsuru Ishizuka
• University of Tokyo

This paper presents an approach to model-based diagnosis that first compilesa first-order system description to a propositional representation, and then solves the diagnostic problem as a linear programminginstance. Relevance reasoning is employed to isolate parts of the systemthat are related to certain observation types andto economically instantiate the theory, while methodsfrom operations research offer promisingresults to generate near-optimaldiagnosesefficiently.

AIJ Journal 1997 Journal Article

Networked bubble propagation: a polynomial-time hypothetical reasoning method for computing near-optimal solutions

• Yukio Ohsawa
• Mitsuru Ishizuka

Hypothetical reasoning (abduction) is an important knowledge processing framework because of its theoretical basis and its usefulness for solving practical problems including diagnosis, design, etc. In many cases, the most probable hypotheses set for diagnosis or the least expensive one for design is desirable. Cost-based abduction, where a numerical weight is assigned to each hypothesis and an optimal solution hypotheses set with minimal sum of element hypotheses' weights is searched, deals with such problems. However, slow inference speed is its crucial problem: cost-based abduction is NP-complete. In order to achieve a tractable inference of cost-based abduction, we aim at obtaining a nearly, rather than exactly, optimal solution. For this approach, an approximate solution method exploited in mathematical programming is quite beneficial. On the other hand, from the standpoint of knowledge processing, it is also important to realize inference on a network which reflects knowledge structure. Knowledge structure is a fruitful information for an efficient inference. In this paper, we propose an inference method which works on a knowledge network, based on a mechanism similar to the pivot and complement method, an efficient approximate 0–1 integer programming method to find a near-optimal solution within a polynomial time of O(N 4), where N is the number of variables or hypotheses. We reformalize this method by a new type of network on which inference is executed by propagating bubbles. This method achieves an inference time of O(N 2) by executing each bubble propagation within a small sub-network, i. e. , by taking advantage of the knowledge structure.

IJCAI Conference 1985 Conference Paper

Prolog-ELF Incorporating Fuzzy Logic

• Mitsuru Ishizuka
• Naoki Kanai