• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.10B
• /
• pp.845-854
• /
• 2008

In this paper, we propose a clustering algorithm for efficient energy management of sensor network consisted of sensor nodes that have restricted energy to solve these problem. Proposed algorithm improves energy efficiency by controlling sensing power. And it has distinctive feature that is applied in various network environment. The performance evaluation result shows that the energy efficiency is improved by 5% in the case of all sensor node fixed and by $10{\sim}15%$ in the case of all sensor node moving. It is confirmed through experiment process that the proposed algorithm brings energy efficiency ratio improvement of $5{\sim}15%$ more than the existing algorithm. Proposed algorithm derived an upper bound on the energy efficiency for Ubiquitous Computing environment that have various network environment that is with ZigBee technology of IEEE 802.15.4 bases. Also, we can blow bring elevation for lifetime of sensor network greatly for lifetime of sensor node as is small. And we think that may expand practical use extent of a sensor network technology more in fast changed network environment.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.6
• /
• pp.842-848
• /
• 2008

This paper deal with uncertainty problem by using Type-2 fuzzy logic set for nonlinear system modeling. We design Type-2 fuzzy logic system in which the antecedent and the consequent part of rules are given as Type-2 fuzzy set and also analyze the performance of the ensuing nonlinear model with uncertainty. Here, the apexes of the antecedent membership functions of rules are decided by C-means clustering algorithm and the apexes of the consequent membership functions of rules are learned by using back-propagation based on gradient decent method. Also, the parameters related to the fuzzy model are optimized by means of particle swarm optimization. The proposed model is demonstrated with the aid of two representative numerical examples, such as mathematical synthetic data set and Mackey-Glass time series data set and also we discuss the approximation as well as generalization abilities for the model.

• 전자공학회논문지 IE
• /
• v.47 no.1
• /
• pp.41-47
• /
• 2010

It is very important to prolong the lifetime of wireless sensor networks by using their limited energy efficiently, since it is not possible to change or recharge the battery of sensor nodes after deployment. LEACH protocol is a typical routing protocol based on the clustering scheme for the efficient use of limited energy. It is composed of a few clusters, which consist of head nodes and member nodes. Though LEACH starts from the supposition that all nodes have data transferred to a head, there must be some nodes having useless data in actual state. In this paper we propose a power saving scheme by making a member node dormant if previous sensed data and current data is same. We evaluate the performance of the proposed scheme in comparison with original clustering algorithms. Simulation results validate our scheme has better performance in terms of the number of alive nodes as time evolves.

• Journal of the Korea Convergence Society
• /
• v.8 no.9
• /
• pp.319-330
• /
• 2017

Recently, various universities in Korea have started to work on strengthening their liberal arts and character education through the residential college (RC) system, carrying out various community programs for this purpose. However, because most programs are based on student-to-student relationships, problems can often arise within the community living environments. This paper proposes the roommate assignment algorithm in the context of a residential college, as to effectively achieve character education goals. The clustering algorithm we propose is based on the similarity hypothesis. As a result of the assignment, the degree of similarity (euclidean distance) between roommates was significantly higher than that assigned randomly. The algorithm developed in this study was applied to the data of the students living in the international campus of H University.

• The KIPS Transactions:PartC
• /
• v.15C no.4
• /
• pp.311-320
• /
• 2008

In general, routing algorithms that were applied to ad-hoc networks are not suitable for the environment with many nodes over several thousands. To solve this problem, hierarchical management to these nodes and clustering-based protocols for the stable maintenance of topology are used. In this paper, we propose the clustering-based modified LEACH protocol that can applied to an environment which moves around metal containers within communication nodes. In proposed protocol, we implemented a module for detecting the movement of nodes on the clustering-based LEACH protocol and improved the defect of LEACH in an environment with movable nodes. And we showed the possibility of the effective communication by adjusting the configuration method of multi-hop. We also compared the proposed protocol with LEACH in four points of view, which are a gradual network composition time, a reconfiguration time of a topology, a success ratio of communication on an containers environment, and routing overheads. And to conclude, we verified that the proposed protocol is better than original LEACH protocol in the metal containers environment within communication of nodes.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06c
• /
• pp.33-35
• /
• 2012

위치 기반 서비스에 대한 활용이 증가되면서 도로 네트워크 환경에서 차량의 이동 궤적을 통해 밀집구역을 발견하기 위한 연구들이 진행되고 있다. 기존 연구들은 특정 도로 세그먼트에 포함된 차량 수를 고려하여 밀집 구역을 발견하였다. 하지만 실제적인 도로 환경에서는 도로마다 다른 길이나 도로의 폭이 다르기 때문에 차량 수만으로 도로가 밀리는 구간을 발견하기에는 문제가 있다. 또한 기존 밀집 구역 발견 연구들은 도로 내 방향성을 고려하지 않는 밀집 구간을 발견한다. 따라서 본 논문에서는 기존 밀집도 기반 클러스터링 연구와는 달리 도로 내 차량 및 도로 환경을 고려하여 도로 혼잡 구간을 판별하는 기법을 제안한다. 제안하는 혼잡 구간 판별 기법은 도로 네트워크를 거리와 폭이 다른 세그먼트로 분할하여 방향성이 존재하는 각 도로 내에 차량의 속도 및 객체 포화도에 따른 혼잡 세그먼트를 추출하고, 이를 통해 혼잡 구간을 판별하는 클러스터를 수행한다. 성능 평가 결과를 통해 제안하는 기법은 혼잡 구역을 클러스터링하여 방향 별 혼잡 구간을 파악할 수 있음을 확인하였다.

• The Journal of Bigdata
• /
• v.6 no.1
• /
• pp.197-206
• /
• 2021

GPR data is used for underground exploration. The data gathered are interpreted by experts based on experience as the underground facilities often reflect GPR. In addition, GPR data are different in the noise and characteristics of the data depending on the equipment, environment, etc. This often results in insufficient data with accurate labels. Generally, a large amount of training data have to be obtained to apply CNN models that exhibit high performance in image classification problems. However, due to the characteristics of GPR data, it makes difficult to obtain sufficient data. Finally, this makes neural networks unable to learn based on general supervised learning methods. This paper proposes an image classification method considering data characteristics to ensure that the accuracy of each label is similar. The proposed method is based on semi-supervised learning, and the image is classified using clustering techniques after extracting the feature values of the image from the neural network. This method can be utilized not only when the amount of the labeled data is insufficient, but also when labels that depend on the data are not highly reliable.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.11
• /
• pp.535-540
• /
• 2021

Multi-document refers to a document consisting of various topics, not a single topic, and a typical example is online reviews. There have been several attempts to summarize online reviews because of their vast amounts of information. However, collective summarization of reviews through existing summary models creates a problem of losing the various topics that make up the reviews. Therefore, in this paper, we present method to summarize the review with minimal loss of the topic. The proposed method classify reviews through processes such as preprocessing, importance evaluation, embedding substitution using BERT, and embedding clustering. Furthermore, the classified sentences generate the final summary using the trained Transformer summary model. The performance evaluation of the proposed model was compared by evaluating the existing summary model, seq2seq model, and the cosine similarity with the ROUGE score, and performed a high performance summary compared to the existing summary model.

• Journal of Internet Computing and Services
• /
• v.24 no.2
• /
• pp.67-74
• /
• 2023

Stock price prediction research in the financial sector aims to ensure trading stability and achieve profit realization. Conventional statistical prediction techniques are not reliable for actual trading decisions due to low prediction accuracy compared to randomly predicted results. Artificial intelligence models improve accuracy by learning data characteristics and fluctuation patterns to make predictions. However, predicting stock prices using long-term time series data remains a challenging problem. This paper proposes a stable and reliable stock price prediction method using K-means clustering-based data augmentation and normalization techniques and LSTM models specialized in time series learning. This enables obtaining more accurate and reliable prediction results and pursuing high profits, as well as contributing to market stability.

• Knowledge Management Research
• /
• v.25 no.1
• /
• pp.217-233
• /
• 2024

The demand for two-wheelers has increased in recent years, driven by the growing delivery culture, which has also led to a rise in the number of two-wheelers. Although two-wheelers are economically efficient in congested traffic conditions, reckless driving and ambiguous traffic laws for two-wheelers have turned two-wheeler accidents into a significant social issue. Given the high fatality rate associated with two-wheelers, the severity and risk of two-wheeler accidents are considerable. It is, therefore, crucial to thoroughly understand the characteristics of two-wheeler accidents by analyzing their attributes. In this study, the characteristics of two-wheeled vehicle accidents were categorized using the K-prototypes algorithm, based on data from two-wheeled vehicle accidents. As a result, the accidents were divided into four clusters according to their characteristics. Each cluster showed distinct traits in terms of the roads where accidents occurred, the major laws violated, the types of accidents, and the times of accident occurrences. By tailoring enforcement methods and regulations to the specific characteristics of each type of accident, we can reduce the incidence of accidents involving two-wheelers in metropolitan areas, thereby enhancing road safety. Furthermore, by applying machine learning techniques to urban transportation and safety, this study adds to the body of related literature.