• Journal of KIISE:Software and Applications
• /
• 제29권8호
• /
• pp.595-603
• /
• 2002

Only a few studies have been conducted on how to select multiple classifiers from the pool of available classifiers for showing the good classification performance. Thus, the selection problem if classifiers on how to select or how many to select still remains an important research issue. In this paper, provided that the number of selected classifiers is constrained in advance, a variety of selection criteria are proposed and applied to tile construction of multiple classifier systems, and then these selection criteria will be evaluated by the performance of the constructed multiple classifier systems. All the possible sets of classifiers are trammed by the selection criteria, and some of these sets are selected as the candidates of multiple classifier systems. The multiple classifier system candidates were evaluated by the experiments recognizing unconstrained handwritten numerals obtained both from Concordia university and UCI machine learning repository. Among the selection criteria, particularly the multiple classifier system candidates by the information-theoretic selection criteria based on conditional entropy showed more promising results than those by the other selection criteria.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• 제31권4D호
• /
• pp.547-553
• /
• 2011

This study has redefined the concepts of mobility indexes and potential demand, standards to evaluate areas with the worst public transportation system and applied the relevant indexes to select the areas with the worst public transportation mobility and present a method to set direct public transportation lines between these regions. The mobility indexes and indexes to evaluate potential demand were applied to select the regions with the worst public transportation systems in four metropolitan cities and case studies were carried out on direct lines provided between these regions. The analysis results showed that in public transportation mobility blind spots, public transportation takes much longer than driving an automobile or public transportation services are not provided. In addition, the analysis showed that a direct lines system to solve such worst off regions should be built to have public transportation take as much time as driving an automobile by establishing lines for automobiles only, minimize time lost from hopping up and down a bus and maximize connections.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 한국정보통신학회 2012년도 춘계학술대회
• /
• pp.758-760
• /
• 2012

The wireless sensor network is used to get information that location tracing or data of surrounding areas. Unnecessary retransmission request or many energy consumption because the transmission over the wireless links. In order to select the link of reliable and energy efficient to estimate the quality of radio link technique is required using RSSI, LQI, and so on. In this paper, each path between the sensor nodes, a small in the path within standard deviation of shall be determined the priority. Each path a high priority of the node values, respectively LQI is accumulated. Node can be selected the high LQI value path. Among them the less hop count to select the path is proposed. The proposed algorithm is removed the paths of shorten life using high the LQI value of the entire and high hop count even less variation. So its advantage that the sensor nodes can be selected more reliable path.

• Journal of Korea Multimedia Society
• /
• 제11권2호
• /
• pp.163-174
• /
• 2008

The H.264/AVC provides gains in compression efficiency of up to 50% over a wide range of bit rates and video resolutions compared to previous standards. However, to achieve such high coding efficiency, the complexity of H.264/AVC encoder is also increased drastically than previous ones, mainly because of mode decision. In this paper, we propose a three-step mode decision algorithm for fast encoding in H.264/AVC. In the first step, we select skip mode or inter mode by considering the temporal correlation and spatial correlation. In the second step, if the result of the first step is INTER mode, we select one group between two groups for final mode. In the third step, we select final mode by exploiting the pixel values of error macroblock or the modes of adjacent macroblocks. Simulations show that the proposed method reduces the encoding time by 42% on average without any significant PSNR losses.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제16권12호
• /
• pp.8195-8209
• /
• 2015

The purpose of this paper is to propose a method for recommendation and personalization of important news articles based on evaluating news value. Evaluation of news is the approach by which editors select news articles for cover-story in traditional offline news papers area. For this, my study proposes a suite of methods to select and personalize a set of news based on evaluating news articles, not just on the personal preference for them. The aforementioned the value of news articles including social impact, novelty, relevance to each audience, and human interest, all of which have been factorized in many previous studies, is a main concept for a procedural and structural application methodology deduced in this study. After a comparative case study with other online news services, it was shown that my research provides more effective way to select important news articles in terms of user satisfaction than others.

• Journal of Broadcast Engineering
• /
• 제7권2호
• /
• pp.88-95
• /
• 2002

According to an evolution of digital broadcasting, it is possible that terrestrial and satellite broadcasting media provide multi-channel services. CATV and satellite media have been also extended to hundreds of channels. As the result of channel expanding, viewers came to select lots of channels. But it is difficult that they select the favorite channel among hundreds of channels. In this paper, we propose an efficient automatic method to recommend channels and programs on a viewer's preference in a multi-channel broadcasting receiver like a Set ToP Box(STB). The proposed algorithm selects channels based on the following method. It makes and saves user history data by using MPEG-7 MDS based on the program information a viewer had watched. It recommends programs similar to a viewer's preference based on user history data. It selects the channel in the recommended genre based on the viewer's channel preference. The experimental result shows that the proposed scheme is efficient to select the user preference channel.

• Journal of Korean Society of Transportation
• /
• 제24권3호
• /
• pp.177-187
• /
• 2006

It is known that over one million car navigation devices are being currently used in Korea. Most. if not all, route guidance systems, however, Provide only one "best" route to users, not providing any options for various types of users to select. The current practice dose not consider each individual's different preferences. These days, a vast amount of information became available due to the rapid development in information processing technology. Thus, users Prefer choices to be given and like to select the one that suits him/her the "best" among available information. To provide such options in this Paper, we developed an algorithm that provides alternative routes that may not the "least cost" ones, but ones that are close to the "least cost" routes for users to select. The algorithm developed and introduced in the paper utilizes a link-based search method, rather than the traditional node-based search method. The link-based algorithm can still utilize the existing transportation network without any modifications, and yet enables to provide flexible route guidance to meet the various needs of users by allowing transfer to other modes and/or restricting left turns. The algorithm developed has been applied to a toy network and demonstrated successful implementation of the multi-path search algorithm for multi-purpose activities.

• Journal of Intelligence and Information Systems
• /
• 제11권1호
• /
• pp.169-189
• /
• 2005

In order to derive a better classifier with a limited number of training examples, active teaming alternately repeats the querying stage fur category labeling and the subsequent learning stage fur rebuilding the calssifier with the newly expanded training set. To relieve the user from the burden of labeling, especially in an on-line environment, it is important to minimize the number of querying steps as well as the total number of query examples. We can derive a good classifier in a small number of querying steps by using only a small number of examples if we can select multiple of diverse, representative, and ambiguous examples to present to the user at each querying step. In this paper, we propose a cluster-based batch query selection method which can select diverse, representative, and highly ambiguous examples for efficient active learning. Experiments with various text data sets have shown that our method can derive a better classifier than other methods which only take into account the ambiguity as the criterion to select multiple query examples.

• Journal of Korea Multimedia Society
• /
• 제11권3호
• /
• pp.353-363
• /
• 2008

A multi-agent system has a longer learning period and larger state-spaces than a sin91e agent system. In this paper, we suggest a new method to reduce the learning time of Nash Q-learning in a multi-agent environment. We apply Macro-actions to Nash Q-learning to improve the teaming speed. In the Nash Q-teaming scheme, when agents select actions, rewards are accumulated like Macro-actions. In the experiments, we compare Nash Q-learning using Macro-actions with general Nash Q-learning. First, we observed how many times the agents achieve their goals. The results of this experiment show that agents using Nash Q-learning and 4 Macro-actions have 9.46% better performance than Nash Q-learning using only 4 primitive actions. Second, when agents use Macro-actions, Q-values are accumulated 2.6 times more. Finally, agents using Macro-actions select less actions about 44%. As a result, agents select fewer actions and Macro-actions improve the Q-value's update. It the agents' learning speeds improve.

• The Journal of the Korea Contents Association
• /
• 제9권10호
• /
• pp.51-58
• /
• 2009

Reinforcement learning is a sub area of machine learning concerned with how an agent ought to take actions in an environment so as to maximize some notion of long-term reward. In the case of multi-agent, especially, which state space and action space gets very enormous in compared to single agent, so it needs to take most effective measure available select the action strategy for effective reinforcement learning. This paper proposes a multi-agent reinforcement learning model based on fuzzy inference system in order to improve learning collect speed and select an effective action in multi-agent. This paper verifies an effective action select strategy through evaluation tests based on Robocup Keepaway which is one of useful test-beds for multi-agent. Our proposed model can apply to evaluate efficiency of the various intelligent multi-agents and also can apply to strategy and tactics of robot soccer system.