Catálogo de publicaciones - libros

This paper analyzes complex collective behaviors of a multiagent system, which consists of interacting agents with evolutionary learning capabilities. The interaction and learning of the agents are modeled by the concept of Connected Replicator Dynamics expanded from evolutionary Game Theory.The dynamic learning system we analyze shows various behavioral and decision changes including bifurcation of chaos in the sense of physical sciences.The main contributions of the paper are summarized as follows: (1) In amultiagent system, the emergence of chaotic behaviors is general and essential, even if each agent does not have chaotic properties; and (2) However,asimple controlling agent with the Keep-It-Simple-Stupid (KISS) principle, or a sheep-dog agent, is able to domesticate or tame the complex behaviors.

Semantic Computing is to design and operate information systems based on meaning and vocabulary shared by people and computers. It aims at closing the semantic gap, thus enabling closer cooperation between people and information systems, and thereby semantically enriching our life-world. Ontologies and constraints are major technologies to let people and information systems share common meaning. Semantic authoring is to compose information content together with explicit semantic structure based on ontologies. This not only reduces the cost of content composition but also improves the quality of the resulting content, by both freeing the author from worries about the order of presentation and providing her a perspicuous view of the logical content structure. Social interactions are much more generally modelled in terms of constraints than in terms of workflows or procedures. CBTO (compositional business-task organization) is a constraint-based framework to concisely describe uniformities of social interactions and thus provides a semantic-level scheme for coordinating various, possibly interactive, services with each other.

We propose the concept of social summarization as a new technology for semantic computing. Social summarization focuses on human evaluative acts toward information, and provides an alternative to the content-based methods employed in the conventional information summarization technologies. We describe the idea of social summarization and its implementation in the community system TelMeA2003, which is being developed to investigate its effectiveness in supporting collaborative activities in online communities. We also report on the preliminary analysis of TelMeA2003 based on our experience obtained in a distance learning community e-Kyositu.

We present as a preliminary study of knowledge discovery from discussion content of offline meetings. Our system generates minutes for such meetings semi-automatically and links them with audio-visual data of discussion scenes. Then, not only retrieval of the discussion content, but also we are pursuing the method of searching for a similar discussion to an ongoing discussion from the past ones, and the method of generation of an answer to a certain question based on the accumulated discussion content. In terms of mailing lists and online discussion systems such as bulletin board systems, various studies have been done. However, what we think is greatly different from the previous works is that ours includes face-to-face offline meetings. We analyze meetings from diversified perspectives using audio and visual information. We also developed a tool for semantic annotation on discussion content. We consider this research not just data mining but a kind of real-world human activity mining.

Semantics is grounded on intellectual activities in various communities in the human society. Conversations and stories are primary media for substantiating the semantic processes in a community. In this paper, I model a semantic process as a coevolutionary spiral of conversations and stories in a community, and present a simple but practical computational model featuring knowledge cards that can serve as a semantic component, lifecycle support of knowledge cards, and strategic control of information streams. Then, I generalize the approach as a technique called conversation quantization, which is based on the idea of approximating a continuous flow of conversation by a series of conversation quanta that represent points of the discourse. I show some implemented systems to show how these ideas are implemented.

This chapter features seven out of nine awarded papers, selected from JSAI 2004 — the 18th annual conference of Japan Society for Artificial Intelligence. These awarded papers are truly excellent, as they were chosen out of over 290 papers, with the selection rate just about three per cent, and the selection involved approximately seventy reviewers. In artificial intelligence and other related fields, empirical methods have recently been rather prevalent in comparison with theoretical approaches. However, all these awarded papers are both theoretically sound and practically feasibile. Synopses of the seven papers follow, with short comments for the award.

We consider the issue of online anomaly detection from a time sequence of directional data (normalized vectors) in high dimensional systems. In spite of the practical importance, little is known about anomaly detection methods for directional data. Using a novel concept of the effective dimension of the system, we successfully formulated an anomaly detection method which is free from the “curse of dimensionality.” In our method, we derive a probability distribution function (pdf) for an anomaly metric, and use a novel update algorithm for the parameters in the pdf, where the effective dimension is included as a fitting parameter. For directional data from a computer system, we demonstrate the utility of our algorithm in anomaly detection.

This paper reports a new document retrieval method using non-relevant documents. Suppose, we need to find documents interesting to the user in as few iterations of human intervention as possible. In each iteration, a relatively small set of documents is evaluated in terms of the relevance to the user’s interest. Ordinary relevance feedback needs both relevant and non-relevant documents, but the initial set of documents checked by the user may often not include relevant documents. Accordingly we propose a new feedback method using non-relevant documents only. This selects documents classified as “not non-relevant“ and close to the boundary defined by the discriminant function obtained from one-class SVM. Experiments show that this method can efficiently retrieve a relevant documents.

This paper describes how to add tags to design documents in order to extract knowledge from information for intelligent design support. Our project called Universal Abduction Studio (UAS) aims to build a new design support system that supports conceptual design by dynamically integrating knowledge in different design domains. This paper focuses on knowledge description form which can be used to capture knowledge from text-based information and then be used for inference for creative design. We propose so-called containing both human-readable texts and machine-readable knowledge such as propositions and rules.

In this paper, we discuss importance and usefulness of interpersonal network in a community support system. We built a scheduling support system for an academic conference. Our system supports information exchange among participants and information discovery with generating participants’ interpersonal network. This system was used in an academic conference called JSAI2003 involving 276 active users. The analysis of the networks reveals that interpersonal networks can promote information exchange among people by indicating existence of people to the others, and that it can also support information discovery by recommendation.