Junichiro Mori

POLYPHONET: an advanced social network extraction system from the web

Social networks play important roles in the Semantic Web: knowledge management, information retrieval, ubiquitous computing, and so on. We propose a social network extraction system called POLYPHONET, which employs several advanced techniques to extract relations of persons, detect groups of persons, and obtain keywords for a person. Search engines, especially Google, are used to measure co-occurrence of information and obtain Web documents.Several studies have used search engines to extract social networks from the Web, but our research advances the following points: First, we reduce the related methods into simple pseudocodes using Google so that we can build up integrated systems. Second, we develop several new algorithms for social networking mining such as those to classify relations into categories, to make extraction scalable, and to obtain and utilize person-to-word relations. Third, every module is implemented in POLYPHONET, which has been used at four academic conferences, each with more than 500 participants. We overview that system. Finally, a novel architecture called Super Social Network Mining is proposed; it utilizes simple modules using Google and is characterized by scalability and Relate-Identify processes: Identification of each entity and extraction of relations are repeated to obtain a more precise social network.

POLYPHONET: An advanced social network extraction system from the Web

Social networks play important roles in the Semantic Web: knowledge management, information retrieval, ubiquitous computing, and so on. We propose a social network extraction system called POLYPHONET, which employs several advanced techniques to extract relations of persons, to detect groups of persons, and to obtain keywords for a person. Search engines, especially Google, are used to measure co-occurrence of information and obtain Web documents. Several studies have used search engines to extract social networks from the Web, but our research advances the following points: first, we reduce the related methods into simple pseudocodes using Google so that we can build up integrated systems. Second, we develop several new algorithms for social network mining such as those to classify relations into categories, to make extraction scalable, and to obtain and utilize person-to-word relations. Third, every module is implemented in POLYPHONET, which has been used at four academic conferences, each with more than 500 participants. We overview that system. Finally, a novel architecture called Iterative Social Network Mining is proposed. It utilizes simple modules using Google and is characterized by scalability and relate-identify processes: identification of each entity and extraction of relations are repeated to obtain a more precise social network.

Recognizing, modeling, and responding to users' affective states

We describe a system that recognizes physiological data of users in real-time, interprets this information as affective states, and responds to affect by employing an animated agent. The agent assumes the role of an Empathic Companion in a virtual job interview scenario where it accompanies a human interviewee. While previously obtained results with the companion with were not significant, the analysis reported here demonstrates that empathic feedback of an agent may reduce user arousal while hearing interviewer questions. This outcome may prove useful for educational systems or applications that induce user stress.

Persona Effect Revisited

The so-called ‘persona effect’ describes the phenomenon that a life-like interface agent can have a positive effect on the user’s perception of a computer-based interaction task. Whereas previous empirical studies rely on questionnaires to evaluate the persona effects, we utilize bio-signals of users in order to precisely associate the occurrence of interface events with users’ autonomic nervous system (ANS) activity. In this paper, we first report on the results of an experiment with an agent-guided mathematical game suggesting that an interface character with affective behavior may significantly decrease user stress. Then, we describe a character-based job interview scenario where a user’s affective state derived from physiological data is projected back (or ‘mirrored’) to the user in real-time. Rather than measuring the effect of an interface agent, the focus here is on employing a character as a medium to reflect the user’s emotional state, a concept with some potential for emotional intelligence training and the medical domain, especially e-Healthcare.

Machine learning approach for finding business partners and building reciprocal relationships

Business development is vital for any firms. However, globalization and the rapid development of technologies have made it difficult to find appropriate business partners such as suppliers and customers, and build reciprocal relationships among them, while it simultaneously offers many opportunities. In this contribution, we propose AI-based approach to find plausible candidates of business partners using firm profiles and transactional relationships among them. We employ machine learning techniques to build a prediction model of customer-supplier relationships. We applied our approach to the large amount of actual business data. The results showed that our approach successfully found potential business partners with F-values of about 84% and reciprocity among them with F-values of about 77%. Using our method, we also developed the Web-based system that helps people in actual businesses to find their new business partners. These contribute to developing ones own business in the complicated, specialized and rapidly changing business environments of recent years.

Detection method of emerging leading papers using time transition

To survive worldwide competitions of research and development in the current rapid increase of information, decision-makers and researchers need to be supported to find promising research fields and papers. But finding those fields from an available data in too much heavy flood of information becomes difficult. We aim to develop a methodology supporting to find emerging leading papers with a bibliometric approach. The analyses in this work are about four academic domains using our time transition analysis. In the time transition analysis, after citation networks are constructed, centralities of each paper are calculated and their changes are tracked. Then, the centralities are plotted, and the features of the leading papers are extracted. Based on the features, we proposed ways to detect the leading papers by focusing on in-degree centrality and its transition. This work will contribute to finding the leading paper, and it is useful for decision-makers and researchers to decide the worthy research topic to invest their resources.

A Framework for Ubiquitous Content Sharing

Examining three content-sharing scenarios reveals requirements for a generic content-sharing framework. The SharedLife framework helps combine different content types and balance user control versus effort when sharing information.

Extraction of business relationships in supply networks using statistical learning theory.

Supply chain management represents one of the most important scientific streams of operations research. The supply of energy, materials, products, and services involves millions of transactions conducted among national and local business enterprises. To deliver efficient and effective support for supply chain design and management, structural analyses and predictive models of customer–supplier relationships are expected to clarify current enterprise business conditions and to help enterprises identify innovative business partners for future success. This article presents the outcomes of a recent structural investigation concerning a supply network in the central area of Japan. We investigated the effectiveness of statistical learning theory to express the individual differences of a supply chain of enterprises within a certain business community using social network analysis. In the experiments, we employ support vector machine to train a customer–supplier relationship model on one of the main communities extracted from a supply network in the central area of Japan. The prediction results reveal an F-value of approximately 70% when the model is built by using network-based features, and an F-value of approximately 77% when the model is built by using attribute-based features. When we build the model based on both, F-values are improved to approximately 82%. The results of this research can help to dispel the implicit design space concerning customer–supplier relationships, which can be explored and refined from detailed topological information provided by network structures rather than from traditional and attribute-related enterprise profiles. We also investigate and discuss differences in the predictive accuracy of the model for different sizes of enterprises and types of business communities.

E-mail networks and leadership performance

Online communication is an indispensable tool for communication and management. The network structure of communication is considered to affect team and individual performances, but it has not been not empirically tested. In this article, we collected a set of 1-month e-mail logs of a company and conducted an e-mail network analysis. We calculated the network centralities of 72 managerial candidates, and investigated the relationship between positions in the network and leadership performance with partial least squares structural equation modeling. Betweenness and in-degree network centralities of those middle managers are correlated with their leadership performance; on the other hand, for this management group, out-degree has no correlation, and PageRank is a negative indicator of leadership. Leaders with high performance are trusted in their communities as a hub of the information channel of the communication network.

Shedding light on a neglected area: a new approach to knowledge creation

Awareness is needed of sustainability issues that are currently neglected but potentially important. A computer-based approach can highlight unconscious and neglected areas because it can structure vast amounts of knowledge. In this article, we propose a methodology to perceive unconscious areas of sustainability with the support of a computer-based approach, which creates a matrix, named the recognized-unrecognized matrix, which highlights both local and globally neglected problems. A case study is conducted to consider the potential contribution of the aviation industry to sustainability issues. We demonstrate that a citation network analysis is an effective methodology to chart the recognized-unrecognized matrix. We highlight issues of water use in the aviation industry by designing an innovative water and air circulation system, which significantly reduces water and fuel consumption on board airplane flights. We also suggest a new approach to integrating both explicit and tacit knowledge to enable knowledge creation.