Hokyin Lai

A commonsense knowledge base supported multi-agent architecture

As the global communication and knowledge exchange between different domains in multi-agent systems (MAS) becomes more important, domain-specific knowledge can no longer solely support reasoning in MAS, whereas commonsense knowledge becomes more critical to the reasoning quality. Incorporating the commonsense knowledge bases (CKB) is at the cutting edge of this field. We are interested in how to reuse CKB in multi-agent systems architectures. Our target is to identify the best way, in terms of efficiency and effectiveness, for MAS to reason with the support of the commonsense knowledge base. In this study, we analyze the reasons for and feasibility of commonsense knowledge reuse in MAS during the knowledge management process, and investigate the commonsense knowledge base-supported MAS architecture design. Four possible approaches to incorporating commonsense knowledge into MAS, with either homogeneous or heterogeneous data, are proposed and tested with the same scenario. The performances have been evaluated using standardized metrics. After comparing the results, one approach is recommended and justifications are given, which constitutes the main contribution of this paper.

A general knowledge mediation infrastructure for multi-agent systems

Intelligent decision making needs to be equipped with broader knowledge in order to enhance the decision quality. Knowledge for decision making can be categorized as domain specific and general. Applying domain knowledge in intelligent systems is not new, but applying general knowledge to support business decision making is a possible way to obtain an edge over competitors. For this reason, the paper focuses primarily on designing a general knowledge mediation infrastructure (GKMI) which supports the use of general knowledge from multiple heterogeneous sources, and provides an unified access point for typical multi-agent systems (MAS) to access that knowledge. The finite state automaton (FSA) is used to model and analyze the commonsense inference ability of GKMI. By carrying out two use cases of GKMI for MAS development and operation the effectiveness of this infrastructure is examined.

An ICA-Based multivariate discretization algorithm

Discretization is an important preprocessing technique in data mining tasks. Univariate Discretization is the most commonly used method. It discretizes only one single attribute of a dataset at a time, without considering the interaction information with other attributes. Since it is multi-attribute rather than one single attribute determines the targeted class attribute, the result of Univariate Discretization is not optimal. In this paper, a new Multivariate Discretization algorithm is proposed. It uses ICA (Independent Component Analysis) to transform the original attributes into an independent attribute space, and then apply Univariate Discretization to each attribute in the new space. Data mining tasks can be conducted in the new discretized dataset with independent attributes. The numerical experiment results show that our method improves the discretization performance, especially for the nongaussian datasets, and it is competent compared to PCA-based multivariate method.

A Cyc-Based Multi-agent System

Multi-agent systems (MAS) are systems in which many intelligent agents interact with each other to accomplish a common goal, e.g. solving a complicated problem in a distributed environment. Domain specific knowledge can no longer solely support reasoning in MAS, whereas common sense knowledge becomes more critical to the reasoning quality, especially when e-commerce has become more popular and more merchants are getting excited to the worldwide market. Incorporating common sense knowledge to the MAS is on edge. However, the use of common sense knowledge induces implementation dilemmas. For example, OpenCyc is not only a common sense knowledge base, but also has its own inference engine. Developers have to determine whether to use the existing inference engine or to use the one that OpenCyc provided. In this paper, four approaches to incorporate common sense knowledge to MAS are proposed and evaluated, and we finally advocate our favorite.

An Agent-Based Approach to Process Management in E-Learning Environments

Learning is a process to acquire new knowledge. Ideally, this process is the result of an active interaction of key cognitive processes, such as perception, imagery, organization, and elaboration. Quality learning has emphasized on designing a course curriculum or learning process, which can elicit the cognitive processing of learners. However, most e-learning systems nowadays are resources-oriented instead of process-oriented. These systems were designed without adequate support of pedagogical principles to guide the learning process. They have not explained the sequence of how the knowledge was acquired, which, in fact, is extremely important to the quality of learning. This study aims to develop an e-learning environment that enables students to get engaged in their learning process by guiding and customizing their learning process in an adaptive way. The expected performance of the Agent-based e-learning Process model is also evaluated by comparing with traditional e-learning models.