• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.51-53
• /
• 2015

In this paper, we propose an efficient sparse signal recovery algorithm referred to as the matching pursuit with a tree pruning (TMP). Two key ingredients of TMP are the pre-selection to put a restriction on columns of the sensing matrix to be investigated and the tree pruning to eliminate unpromising paths from the search tree. In our analysis, we show that the sparse signal is accurately reconstructed when the sensing matrix satisfies the restricted isometry property. In our simulations, we confirm that TMP is effective in recovering sparse signals and outperforms conventional sparse recovery algorithms.

• Journal of Internet Computing and Services
• /
• v.16 no.6
• /
• pp.23-30
• /
• 2015

As BigData is issued, many applications that operate based on the results of data analysis have been developed, typically applications are products recommend service of e-commerce application service system, search service on the search engine service and friend list recommend system of social network service. In this paper, we suggests a decision frequent pattern tree that is combined the origin frequent pattern tree that is mining similar pattern to appear in the data set of the existing data mining techniques and decision tree based on the theory of computer science. The decision frequent pattern tree algorithm improves about problem of frequent pattern tree that have to make some a lot's pattern so it is to hard to analyze about data. We also proposes to model for a Mapredue framework that is a programming model to help to operate in distributed environment.

• Journal of Information Technology Services
• /
• v.16 no.4
• /
• pp.137-148
• /
• 2017

Researches on the Artificial Intelligence has been explosively activated in various fields since the advent of AlphaGo. Particularly, researchers on the application of multi-layer neural network such as deep learning, and various machine learning algorithms are being focused actively. In this paper, we described a development of an artificial intelligence Janggi game based on reinforcement learning algorithm and MCTS (Monte Carlo Tree Search) algorithm with accumulated game data. The previous artificial intelligence games are mostly developed based on mini-max algorithm, which depends only on the results of the tree search algorithms. They cannot use of the real data from the games experts, nor cannot enhance the performance by learning. In this paper, we suggest our approach to overcome those limitations as follows. First, we collects Janggi expert's game data, which can reflect abundant real game results. Second, we create a graph structure by using the game data, which can remove redundant movement. And third, we apply the reinforcement learning algorithm and MCTS algorithm to select the best next move. In addition, the learned graph is stored by object serialization method to provide continuity of the game. The experiment of this study is done with two different types as follows. First, our system is confronted with other AI based system that is currently being served on the internet. Second, our system confronted with some Janggi experts who have winning records of more than 50%. Experimental results show that the rate of our system is significantly higher.

• The KIPS Transactions:PartA
• /
• v.16A no.4
• /
• pp.289-298
• /
• 2009

Generally, RFID systems are composed of one reader and several passive tags, and share the single wireless channel. For this reason, collisions occurwhen more than two tags simultaneously respond to the reader's inquiry. To achieve this problem, many papers, such as QT[8], HCT[10], BSCTTA[2], and QT-BCS[9], have been proposed. In this paper, we propose the tree-based anti-collision algorithm using a bit change sensing unit (TABCS) based on BSCTTA algorithm. The proposed algorithm can identify bits returned from tags through bit change sensing unit, even if multi collisions occur. So, it rapidly generates the unique prefix to indentify each tag, and reduce the total of bits. As the result, the cost of identifying all tag IDs is relatively reduced as compared with existing algorithms. It is verified through simulations that the proposed algorithm surpass other existing algorithms.

• Korean Journal of Oriental Medicine
• /
• v.17 no.2
• /
• pp.121-127
• /
• 2011

Objectives : In the clinical field, it is important to understand the factors that have effects on a certain disease or symptom. For this, many researchers apply Data Mining method to the clinical data that they have collected. One of the efficient methods for Data Mining is decision tree induction. Many researchers have studied to find the best split criteria of decision tree; however, various split criteria coexist. Methods : In this paper, we applied several split criteria(Information Gain, Gini Index, Chi-Square) to Sasang constitutional clinical information and compared each decision tree in order to find optimal split criteria. Results & Conclusion : We found BMI and body measurement factors are important factors to Sasang constitution by analyzing produced decision trees with different split measures. And the decision tree using information gain had the highest accuracy. However, the decision tree that produced highest accuracy is changed depending on given data. So, researcher have to try to find proper split criteria for given data by understanding attribute of the given data.

• Journal of KIISE:Software and Applications
• /
• v.36 no.7
• /
• pp.560-570
• /
• 2009

RFID, Radio Frequency Identification, is a technology of automated identification replacing bar-code. RFID technology has advantages that it recognizes fast and it is strong to contamination using wireless communication. However, there are difficult problems that should be solved for popularization of RFID. Among of these, tag anti-collision problem is dealed in this paper. It affected the performance of RFID system directly. This paper analyzes conventional algorithms and proposes new algorithms of tag anti-collision. The algorithm proposed was composed of appropriate properties to each phase of distribution and recognition as hybrid between ALOHA-based algorithm and QT-based algorithm. At phase of distribution, the number of tags recognizing at a frame was reduced using ALOHA-based algorithm. It addressed the delay problem because of deep depth of tree. At phase of recognition, it solved ALOHA-based chronic problem that couldn't recognize all the tags sometimes. Moreover, QTR algorithm that recognize by reversed tag IDs was adopted for the performance. The FQTR algorithm proposed in this paper showed brilliant performance as compared with convention algorithms by simulation.

• The Journal of the Acoustical Society of Korea
• /
• v.32 no.2
• /
• pp.174-180
• /
• 2013

This paper proposes a decision tree based context clustering algorithm for HMM-based speech synthesis systems using the cross likelihood ratio with a hierarchical prior (CLRHP). Conventional algorithms tie the context-dependent HMM states that have similar statistical characteristics, but they do not consider the statistical similarity of split child nodes, which does not guarantee the statistical difference between the final leaf nodes. The proposed CLRHP algorithm improves the reliability of model parameters by taking a criterion of minimizing the statistical similarity of split child nodes. Experimental results verify the superiority of the proposed approach to conventional ones.

• Land and Housing Review
• /
• v.1 no.1
• /
• pp.1-7
• /
• 2010

Case-based reasoning (CBR) method can make estimators understand the estimation process more clearly. Thus, CBR is widely used as a methodology for cost estimation. In CBR, the quality of case retrieval affects the relevance of retrieved cases and hence the overall quality of the reminding capability of CBR system. Thus, it is essential to retrieve relevant past cases for establishing a robust CBR system. Case retrieval needs the following tasks to obtain appropriate case(s); indexing, search, and matching (Aamodt and Plaza 1994). However, the previous CBR researches mostly deal with matching process that has limits such as accuracy and efficiency of case retrieval. In order to address this issue, this research presents a CBR cost model for building projects that has two-step retrieval process: decision tree and nearest neighbor methods. Specifically, the proposed cost model has indexing, search and matching modules. Features in the model are divided into shape-based and scale-based attributes. Based on these, decision tree is established for facilitating the search task and nearest neighbor method was utilized for matching task. In regard to applying nearest neighbor method, attribute weights are assigned using GA optimization and similarity is calculated using the principle of distance measuring. Thereafter, the proposed CBR cost model is developed using 174 cases and validated using 12 test cases.

• The Journal of Korea Robotics Society
• /
• v.13 no.2
• /
• pp.142-149
• /
• 2018

$RRT^*$ (Rapidly exploring Random $Tree^*$) based algorithms are widely used for path planning. Informed $RRT^*$ uses $RRT^*$ for generating an initial path and optimizes the path by limiting sampling regions to the area around the initial path. $RRT^*$ algorithms have several limitations such as slow convergence speed, large memory requirements, and difficulties in finding paths when narrow aisles or doors exist. In this paper, we propose an algorithm to deal with these problems. The proposed algorithm applies the image skeletonization to the gridmap image for generating an initial path. Because this initial path is close to the optimal cost path even in the complex environments, the cost can converge to the optimum more quickly in the proposed algorithm than in the conventional Informed $RRT^*$. Also, we can reduce the number of nodes and memory requirement. The performance of the proposed algorithm is verified by comparison with the conventional Informed $RRT^*$ and Informed $RRT^*$ using initial path generated by $A^*$.

• Journal of KIISE:Information Networking
• /
• v.36 no.3
• /
• pp.153-162
• /
• 2009

It is a challenging work for service operators to accurately classify different services, which runs on various wireless networks based upon numerous platforms. This works focuses on design and implementation of a classifier, which accurately classifies applications, which are captured horn WiBro Network. Notion of session is introduced for the classifier, instead of commonly used Flow to develop a classifier. Based on session information of given traffic, two classification algorithms are presented, Classification and Regression Tree and Support Vector Machine. Both algorithms are capable of classifying accurately and effectively with misclassification rate of 0.85%, and 0.94%, respectively. This work shows that classifier using CART provides ease of interpreting the result and implementation.