• Journal of KIISE:Software and Applications
• /
• v.33 no.11
• /
• pp.953-964
• /
• 2006

Due to the usefulness of the collaborative filtering, it has been widely used in both the research and commercial field. However, there are still some challenges for it to be more efficient, especially the scalability problem, the sparsity problem and the cold start problem. In this paper. we address these problems and provide a novel distributed approach based on agents collaboration for the problems. We have tried to solve the scalability problem by making each agent save its users ratings and broadcast them to the users friends so that only friends ratings and his own ratings are kept in an agents local database. To reduce quality degradation of recommendation caused by the lack of rating data, we introduce a method using friends opinions instead of real rating data when they are not available. We also suggest a collaborative filtering algorithm based on user profile to provide new users with recommendation service. Experiments show that our suggested approach is helpful to the new user problem as well as is more scalable than traditional centralized CF filtering systems and alleviate the sparsity problem.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.9
• /
• pp.281-290
• /
• 2020

The Vision-and-Language Navigation(VLN) task is a complex intelligence problem that requires both visual and language comprehension skills. In this paper, we propose a new learning model for visual-language navigation agents. The model adopts a hybrid learning that combines imitation learning based on demo data and reinforcement learning based on action reward. Therefore, this model can meet both problems of imitation learning that can be biased to the demo data and reinforcement learning with relatively low data efficiency. In addition, the proposed model uses a novel path-based reward function designed to solve the problem of existing goal-based reward functions. In this paper, we demonstrate the high performance of the proposed model through various experiments using both Matterport3D simulation environment and R2R benchmark dataset.

• KIISE Transactions on Computing Practices
• /
• v.22 no.8
• /
• pp.393-398
• /
• 2016

Collective Adaptive System is an adaptive multi-agent system which accomplishes its goal by collaborating various agents. Because the collective property of the Collective Adaptive System is accomplished by the goal of the system being based on collaboration, testing the goal accomplishment and their interactions among heterogeneous agents is important. This paper presents a feasibility study of applying a model-based testing approach using task-based test model to a Collective Adaptive System. This paper describes additional information to be applied for Collective Adaptive System for future studies. To analyze our approach, we applied the proposed approach to a smart home system as a case study; our results indicated that we can systematically derive test cases to check whether the Collective Adaptive System successfully achieved its goals by modifying and extending the existing task model.

• The KIPS Transactions:PartC
• /
• v.8C no.6
• /
• pp.855-864
• /
• 2001

With the increasing importance and complexity of EC (Electronic Commerce) across the Internet, the need and expectation for intelligent software agents to support both consumers and suppliers through the whole process of EC are growing rapidly. To realize the intelligent EC. a multiagent based EC system. which includes foundational technologies such as the establishment of standard product ontology the definition of message and negotiation protocol and brockering, is required. In this paper we propose an intelligent EC System named ICOMA(Intelligent electronic CO mmerce system based on Multi-Agent) as an open infrastructure of multiagent-based EC. Concretely we have proposed. designed and implemented an architecture of multiagent-based EC system including 6-types of agents message protocol for inter-agent negotiation, personalized produst retrieval and filtering., We have confirmed the effectiveness of the system through experiments.

• KIISE Transactions on Computing Practices
• /
• v.21 no.4
• /
• pp.290-298
• /
• 2015

This paper suggests a multi-agent message transport architecture based on the message-oriented middleware ZeroMQ. Compared with the other middlewares such as CORBA, Ice, and Thrift, ZeroMQ receives a good score in the evaluation of performance, QoS (Quality of Service), patterns, user friendliness, and resources. The suggested message transfer architecture borrowed many basic concepts like agent platform, AMS (Agent Management System), and MTS (Message Transfer System) from FIPA (Foundation for Intelligent Physical Agents) standard multi-agent specifications, and the architecture inherited the strength of the architecture from the multi-agent framework SMAF (Smart Multi-Agent Framework). The architecture suggested in this paper is a novel peer-to-peer architecture which is not known to the ZeroMQ community. In the suggested architecture, every MTS agent uses only one ZeroMQ router socket to support peer-to-peer communication among MTS agents. The suggested architecture can support closely collaborating software areas such as intelligent robots as well as the traditional application areas of multi-agent architecture. The suggested architecture has interoperability and scalability with the ZeroMQ devices and patterns.

• Advances in robotics research
• /
• v.2 no.3
• /
• pp.201-217
• /
• 2018

Agriculture production is a critical human intensive task, which takes place in all regions of the world. The process to grow and harvest crops is labor intensive in many countries due to the lack of automation and advanced technology. Much of the difficult, dangerous and dirty labor of crop production can be automated with intelligent and robotic platforms. We propose an intelligent, agent-oriented robotic team, which can enable the process of harvesting, gathering and collecting crops and fruits, of many types, from agricultural fields. This paper describes a novel robotic organization enabling humans, robots and agents to work together for automation of gathering and collection functions. The focus of the research is a model, called HARMS, which can enable Humans, software Agents, Robots, Machines and Sensors to work together indistinguishably. With this model, any capability-based human-like organization can be conceived and modeled, such as in manufacturing or agriculture. In this research, we model, design and implement a technology application of knowledge-based robot-to-robot and human-to-robot collaboration for an agricultural gathering and collection function. The gathering and collection functions were chosen as they are some of the most labor intensive and least automated processes in the process acquisition of agricultural products. The use of robotic organizations can reduce human labor and increase efficiency allowing people to focus on higher level tasks and minimizing the backbreaking tasks of agricultural production in the future. In this work, the HARMS model was applied to three different robotic instances and an integrated test was completed with satisfactory results that show the basic promise of this research.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.9
• /
• pp.399-406
• /
• 2023

Cloud computing has been evolved to support edge computing architecture that combines fog management layer with edge servers. The main reason why it is received much attention is low communication latency for real-time IoT applications. At the same time, various cloud task scheduling techniques based on artificial intelligence have been proposed. Artificial intelligence-based cloud task scheduling techniques show better performance in comparison to existing methods, but it has relatively high scheduling time. In this paper, we propose a deep learning-based dynamic scheduling with multi-agents supporting scalability in edge computing environments. The proposed method shows low scheduling time than previous artificial intelligence-based scheduling techniques. To show the effectiveness of the proposed method, we compare the performance between previous and proposed methods in a scalable experimental environment. The results show that our method supports real-time IoT applications with low scheduling time, and shows better performance in terms of the number of completed cloud tasks in a scalable experimental environment.

• Journal of KIISE:Software and Applications
• /
• v.30 no.7_8
• /
• pp.672-680
• /
• 2003

A shopbot is a software agent whose goal is to maximize buyer´s satisfaction through automatically gathering the price and quality information of goods as well as the services from on-line sellers. In the response to shopbots´ activities, sellers on the Internet need the agents called pricebots that can help them maximize their own profits. In this paper we adopts Q-learning, one of the model-free reinforcement learning methods as a price-setting algorithm of pricebots. A Q-learned agent increases profitability and eliminates the cyclic price wars when compared with the agents using the myoptimal (myopically optimal) pricing strategy Q-teaming needs to select a sequence of state-action fairs for the convergence of Q-teaming. When the uniform random method in selecting state-action pairs is used, the number of accesses to the Q-tables to obtain the optimal Q-values is quite large. Therefore, it is not appropriate for universal on-line learning in a real world environment. This phenomenon occurs because the uniform random selection reflects the uncertainty of exploitation for the optimal policy. In this paper, we propose a Mixed Nonstationary Policy (MNP), which consists of both the auxiliary Markov process and the original Markov process. MNP tries to keep balance of exploration and exploitation in reinforcement learning. Our experiment results show that the Q-learning agent using MNP converges to the optimal Q-values about 2.6 time faster than the uniform random selection on the average.

• Journal of KIISE:Software and Applications
• /
• v.32 no.3
• /
• pp.202-214
• /
• 2005

The semantic web is the web paradigm that represents not general link of documents but semantics and relation of document. In addition it enables software agents to understand semantics of documents. We propose a semantic search based on inference with ontologies, which has the following characteristics. First, our search engine enables retrieval using explicit ontologies to reason though a search keyword is different from that of documents. Second, although the concept of two ontologies does not match exactly, can be found out similar results from a rule based translator and ontological reasoning. Third, our approach enables search engine to increase accuracy and precision by using explicit ontologies to reason about meanings of documents rather than guessing meanings of documents just by keyword. Fourth, domain ontology enables users to use more detailed queries based on ontology-based automated query generator that has search area and accuracy similar to NLP. Fifth, it enables agents to do automated search not only documents with keyword but also user-preferable information and knowledge from ontologies. It can perform search more accurately than current retrieval systems which use query to databases or keyword matching. We demonstrate our system, which use ontologies and inference based on explicit ontologies, can perform better than keyword matching approach .

• Journal of Intelligence and Information Systems
• /
• v.12 no.2
• /
• pp.67-81
• /
• 2006

While the existing Web focuses on the interface with human users based on HTML, the next generation Web will focus on the interaction among software agents by using XML and XML-based standards and technologies. The inference engine, which will serve as brains of software agents in the next generation Web, should thoroughly understand the Semantic Web, the standard language of the next generation Web. As abasis for the service, the W3C (World Wide Web Consortium) has recommended SWRL (Semantic Web Rule Language) which had been made by compounding OWL (Web Ontology Language) and RuleML (Rule Markup Language). In this research, we develop a backward chaining inference engine SMART-B (SeMantic web Agent Reasoning Tools -Backward chaining inference engine), which uses SWRL and OWL to represent rules and facts respectively. We analyze the requirements for the SWRL-based backward chaining inference and design analgorithm for the backward chaining inference which reflects the traditional backward chaining inference algorithm and the requirements of the next generation Semantic Web. We also implement the backward chaining inference engine and the administrative tools for fact and rule bases into Java components to insure the independence and portability among different platforms under the environment of Ubiquitous Computing.