Sungsoo Lim

Intelligent OS process scheduling using fuzzy inference with user models

The process scheduling aims to arrange CPU time to multiple processes for providing users with more efficient throughput. Except the class of process set by user, conventional operating systems have applied the equivalent scheduling policy to every process. Moreover, if the scheduling policy is once determined, it is unable to change without resetting the operating system which takes much time. In this paper, we propose an intelligent CPU process scheduling algorithm using fuzzy inference with user models. It classifies processes into three classes, batch, interactive and real-time processes, and models users preferences to each process class. Finally, it assigns the priority of each process according to the class of the process and users preference through the fuzzy inference. The experimental result shows the proposed method can adapt to user and allow different scheduling policies to multiple users.

Language generation for conversational agent by evolution of plan trees with genetic programming

As dialogue systems are widely demanded, the research on natural language generation in dialogue has raised interest. Contrary to conventional dialogue systems that reply to the user with a set of predefined answers, a newly developed dialogue system generates them dynamically and trains answers to support more flexible and customized dialogues with humans. The paper proposes an evolutionary method for generating sentences using interactive genetic programming. Sentence plan trees, which stand for the sentence structure, are adopted as the representation of genetic programming. With interactive evolution process with the user, a set of customized sentence structures is obtained. The proposed method applies to a dialogue-based travel planning system and the usability test demonstrates the usefulness of the proposed method.

Gesture based dialogue management using behavior network for flexibility of human robot interaction

The usage of robots becomes more sophisticated, direct communication by means of human language is required to increase the efficiency of their performance. However, the dialogue systems that reply to the user with a set of predefined answers tend to be static. In this paper, we propose a gesture based dialogue system using behavior network for flexibility of human robot interaction. Gestures take an important part of interactions. By using gestures in dialogues, it could support a flexible and realistic interaction with humans. We confirm the usability of gestures through several scenarios and SUS subject test.

Online learning of Bayesian network parameters with incomplete data

Learning Bayesian network is a problem to obtain a network that is the most appropriate to training dataset based on the evaluation measures given. It is studied to decrease time and effort for designing Bayesian networks. In this paper, we propose a novel online learning method of Bayesian network parameters. It provides high flexibility through learning from incomplete data and provides high adaptability on environments through online learning. We have confirmed the performance of the proposed method through the comparison with Voting EM algorithm, which is an online parameter learning method proposed by Cohen, et al.

High-Precision Printing Force Control System for Roll-to-Roll Manufacturing

High-precision printing machines are essential for the fabrication of more sophisticated printed electronics. In particular, roll-to-roll printing with flexible substrates and nonideal rolls, which exhibit cylindricity, misalignment, and run-outs, produces nonlinear effects that are hard to manage with conventional control systems. For instance, printing rolls with cylindricity and run-outs cause large perturbations in the printing forces. In this paper, we propose a hybrid printing force control system that provides uniform printing conditions in the regions where printing occurs. The unique parallel arrangement of two force components provides excellent regulation of the printing force; this system also consumes less power. Our experimental results verify that the force control performance of the hybrid force control system is better than that of a conventional system. We also report the results of in situ printing quality tests with a pressure-sensitive film that confirm the effectiveness of the proposed system.

A Spontaneous Topic Change of Dialogue for Conversational Agent Based on Human Cognition and Memory

Mixed-initiative interaction (MII) plays an important role for the flexible dialogues in conversational agent. Since conventional research on MII process dialogues based on the predefined methodologies, they only provide simple and static dialogues rather than complicated and dynamic dialogues through context-aware themselves. In this paper, we proposed a spontaneous conversational agent that provides MII and can change topics of dialogue dynamically based on human cognitive architecture and memory structure. Based on the global workspace theory, one of the simple cognitive architecture models, the proposed agent is aware of the context of dialogue in conscious level and chooses the topic in unconscious level which is the most relevant to the current context as the next topic of dialogues. We represent the unconscious part of memory using semantic network which is a popular representation for storing knowledge in the field of cognitive science, and retrieve the semantic network according to the spreading activation theory which is proven to be efficient for inferring in semantic networks. It is verified that the proposed method spontaneously changes the topics of dialogues through some dialogue examples on the domain of schedule management.

Language learning for the autonomous mental development of conversational agents

Since the manual construction of our knowledge-base has several crucial limitations when applied to intelligent systems, mental development has been investigated in recent years. Autonomous mental development is a new paradigm for developing autonomous machines, which are adaptive and flexible to the environment. Language development, a kind of mental development, is an important aspect of intelligent conversational agents. In this paper, we propose an intelligent conversational agent and its language development mechanism by putting together five promising techniques; Bayesian networks, pattern matching, finite state machines, templates, and genetic programming. Knowledge acquisition implemented by finite state machines and templates, and language learning by genetic programming are developed for language development. Several illustrations and usability tests show the usefulness of the proposed developmental conversational agent.

Automatic construction of bayesian networks for conversational agent

As the information in the internet proliferates, the methods for effectively providing the information have been exploited, especially in conversational agents. Bayesian network is applied to infer the intention of users query. Since the construction of Bayesian network requires large efforts and much time, an automatic method for it might be useful for applying conversational agents to several applications. In order to improve the scalability of the agent, in this paper, we propose a method of automatically generating Bayesian networks from scripts composing knowledge base of the conversational agent. It constructs the structure of hierarchically composing nodes and learns the conditional probability distribution table using Noisy-OR gate. The experimental results with subjects confirm the usefulness of the proposed method.