• Journal of Intelligence and Information Systems
• /
• v.16 no.4
• /
• pp.99-112
• /
• 2010

Ensemble learning is a method for improving the performance of classification and prediction algorithms. It is a method for finding a highly accurateclassifier on the training set by constructing and combining an ensemble of weak classifiers, each of which needs only to be moderately accurate on the training set. Ensemble learning has received considerable attention from machine learning and artificial intelligence fields because of its remarkable performance improvement and flexible integration with the traditional learning algorithms such as decision tree (DT), neural networks (NN), and SVM, etc. In those researches, all of DT ensemble studies have demonstrated impressive improvements in the generalization behavior of DT, while NN and SVM ensemble studies have not shown remarkable performance as shown in DT ensembles. Recently, several works have reported that the performance of ensemble can be degraded where multiple classifiers of an ensemble are highly correlated with, and thereby result in multicollinearity problem, which leads to performance degradation of the ensemble. They have also proposed the differentiated learning strategies to cope with performance degradation problem. Hansen and Salamon (1990) insisted that it is necessary and sufficient for the performance enhancement of an ensemble that the ensemble should contain diverse classifiers. Breiman (1996) explored that ensemble learning can increase the performance of unstable learning algorithms, but does not show remarkable performance improvement on stable learning algorithms. Unstable learning algorithms such as decision tree learners are sensitive to the change of the training data, and thus small changes in the training data can yield large changes in the generated classifiers. Therefore, ensemble with unstable learning algorithms can guarantee some diversity among the classifiers. To the contrary, stable learning algorithms such as NN and SVM generate similar classifiers in spite of small changes of the training data, and thus the correlation among the resulting classifiers is very high. This high correlation results in multicollinearity problem, which leads to performance degradation of the ensemble. Kim,s work (2009) showedthe performance comparison in bankruptcy prediction on Korea firms using tradition prediction algorithms such as NN, DT, and SVM. It reports that stable learning algorithms such as NN and SVM have higher predictability than the unstable DT. Meanwhile, with respect to their ensemble learning, DT ensemble shows the more improved performance than NN and SVM ensemble. Further analysis with variance inflation factor (VIF) analysis empirically proves that performance degradation of ensemble is due to multicollinearity problem. It also proposes that optimization of ensemble is needed to cope with such a problem. This paper proposes a hybrid system for coverage optimization of NN ensemble (CO-NN) in order to improve the performance of NN ensemble. Coverage optimization is a technique of choosing a sub-ensemble from an original ensemble to guarantee the diversity of classifiers in coverage optimization process. CO-NN uses GA which has been widely used for various optimization problems to deal with the coverage optimization problem. The GA chromosomes for the coverage optimization are encoded into binary strings, each bit of which indicates individual classifier. The fitness function is defined as maximization of error reduction and a constraint of variance inflation factor (VIF), which is one of the generally used methods to measure multicollinearity, is added to insure the diversity of classifiers by removing high correlation among the classifiers. We use Microsoft Excel and the GAs software package called Evolver. Experiments on company failure prediction have shown that CO-NN is effectively applied in the stable performance enhancement of NNensembles through the choice of classifiers by considering the correlations of the ensemble. The classifiers which have the potential multicollinearity problem are removed by the coverage optimization process of CO-NN and thereby CO-NN has shown higher performance than a single NN classifier and NN ensemble at 1% significance level, and DT ensemble at 5% significance level. However, there remain further research issues. First, decision optimization process to find optimal combination function should be considered in further research. Secondly, various learning strategies to deal with data noise should be introduced in more advanced further researches in the future.

• ETRI Journal
• /
• v.28 no.1
• /
• pp.77-83
• /
• 2006

In order to improve the accuracy of cell selection in heterogeneous cellular systems, this paper proposes a fuzzy multiple-objective decision-based cell selection (FMDCS) strategy. Since heterogeneous cellular systems have different access technologies and multiple traffic classes, the strategy adopts cell type, data rate, coverage, transmission delay, and call arrival rate as evaluation indices, and uses different weight vectors according to the traffic classes of the mobile host. Then, a fuzzy multiple-objective decision algorithm is applied to select the optimal cell from all candidates. This paper also gives an instance analysis and simulation. The instance analysis shows FMDCS makes different selections for different traffic classes. Simulation results of the after-handoff quality-of-service (QoS) show the selected cell can provide MH optimal service.

• Journal of the military operations research society of Korea
• /
• v.24 no.2
• /
• pp.146-161
• /
• 1998

This article develops model of the nonstationary state behavior of the multiechelon spare parts provisioning systems. This study is concerned with a problem of determining the near optimal requirements level of the spare parts, especially Concurrent Spare Parts(CSP). CSP is supplied with the procurement of new equipment system, and is used to sustain the equipment without resupply during the initial coverage period. We consider this situation as a multiechelon inventory model with several bases and one depot. And we assume an equipment system which consists of many types of parts would grounded if one of the parts fail. Also this multiechelon CSP problem is considering the nonstationary poisson failure process and nonstationary exponential repair process in a dynamic environment. We develop an efficient computational procedure to find the near optimal number of spare parts minimizing the total expected cost, while achieving the required system availability. Finally we present a simple example of suggested method.

• Korean Journal of Remote Sensing
• /
• v.27 no.6
• /
• pp.637-651
• /
• 2011

This study presents a rule-based urban image classification method for time series analysis of changes in the vicinity of Asan-si and Cheonan-si in Chungcheongnam-do, using Landsat satellite images (1991-2006). The area has been highly developed through the relocation of industrial facilities, land development, construction of a high-speed railroad, and an extension of the subway. To determine the yearly changing pattern of the urban area, eleven classes were made depending on the trend of development. An algorithm was generalized for the rules to be applied as an unsupervised classification, without the need of training area. The analysis results show that the urban zone of the research area has increased by about 1.53 times, and each correlation graph confirmed the distribution of the Built Up Index (BUI) values for each class. To evaluate the rule-based classification, coverage and accuracy were assessed. When Optimal allowable factor=0.36, the coverage of the rule was 98.4%, and for the test using ground data from 1991 to 2006, overall accuracy was 99.49%. It was confirmed that the method suggested to determine the maximum allowable factor correlates to the accuracy test results using ground data. Among the multiple images, available data was used as best as possible and classification accuracy could be improved since optimal classification to suit objectives was possible. The rule-based urban image classification method is expected to be applied to time series image analyses such as thematic mapping for urban development, urban development, and monitoring of environmental changes.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.7
• /
• pp.33-40
• /
• 2015

The time slot assignment problem (TSAP) or Satellite Communications scheduling problem (SCSP) for a satellite performs $n{\times}n$ ground station data traffic switching has been known NP-hard problem. This paper suggests $O(n^2)$ time complexity algorithm for TSAP of a satellite that performs $n^2{\times}n^2$ ground station data traffic switching. This problem is more difficult than $n{\times}n$ TSAP as NP-hard problem. Firstly, we compute the average traffic for n-transponder's basic coverage zone and applies ground station exchange method that swap the ground stations until all of the transponders have a average value as possible. Nextly, we transform the D matrix to $D_{LB}$ traffic matrix that sum of rows and columns all of transponders have LB. Finally, we select the maximum traffic of row and column in $D_{LB}$, then decide the duration of kth switch mode to minimum traffic from selected values. The proposed algorithm can be get the optimal solution for experimental data.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.2055-2058
• /
• 2002

It designs the general description of 8-VSB DTV transmission system will be installed and operated in future on the basis of NTSC TV transmission system in operation now. First of all, it investigates a condition that can maintain optimum received field in case of ground wave broadcasting with analog NTSC and digital 8-VSB DTV. Also it calculates critical value of minimum field strength fur receiving by using field strength calculation method of Longley-Rice. The 8-VSB DTV transmission sets its goal in transmission of data with 19.39Mbps at high speed in order to be available of HDTV in the band of NTSC 6MHz. Accordingly, the optimum of field strength between transmission and receiving points most of all is important problem. So,8-VSB DTV transmission should be enlarged or effective compared with NTSC service coverage. This study is to calculate the transmission power necessary in switching from NTSC VHF to DTV UHF and from NTSC UHF to DTV UHF for maintaining service coverage that NTSC is equal to 8-VSB DTV as a result of calculation.

• Journal of Communications and Networks
• /
• v.14 no.5
• /
• pp.578-587
• /
• 2012

One of the key elements in the emerging, packet-based long term evolution (LTE) cellular systems is the deployment of multiple femtocells for the improvement of coverage and data rate. However, arbitrary overlaps in the coverage of these femtocells make the handover operation more complex and challenging. As the existing handover strategy of LTE systems considers only carrier to interference plus noise ratio (CINR), it often suffers from resource constraints in the target femtocell, thereby leading to handover failure. In this paper, we propose a new efficient, multi-objective handover solution for LTE cellular systems. The proposed solution considers multiple parameters like signal strength and available bandwidth in the selection of the optimal target cell. This results in a significant increase in the handover success rate, thereby reducing the blocking of handover and new sessions. The overall handover process is modeled and analyzed by a three-dimensional Markov chain. The analytical results for the major performance metrics closely resemble the simulation results. The simulation results show that the proposed multi-objective handover offers considerable improvement in the session blocking rates, session queuing delay, handover latency, and goodput during handover.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.5
• /
• pp.1951-1972
• /
• 2018

Wi-Fi Access Point (AP) optimization approaches are used in indoor positioning systems for signal coverage enhancement, as well as positioning precision improvement. Although the huge power consumption of the AP optimization forms a serious problem due to the signal coverage requirement for large-scale indoor environment, the conventional approaches treat the problem of power consumption independent from the design of indoor positioning systems. This paper proposes a new Fast Water-filling algorithm Group Power Constraint (FWA-GPC) based Wi-Fi AP optimization approach for indoor positioning in which the power consumed by the AP optimization is significantly considered. This paper has three contributions. First, it is not restricted to conventional concept of one AP for one candidate AP location, but considered spare APs once the active APs break off. Second, it utilizes the concept of water-filling model from adaptive channel power allocation to calculate the number of APs for each candidate AP location by maximizing the location fingerprint discrimination. Third, it uses a fast version, namely Fast Water-filling algorithm, to search for the optimal solution efficiently. The experimental results conducted in two typical indoor Wi-Fi environments prove that the proposed FWA-GPC performs better than the conventional AP optimization approaches.

• Journal of Korean Dental Science
• /
• v.17 no.2
• /
• pp.92-104
• /
• 2024

Managing multiple non-carious cervical lesions (NCCLs) with gingival recession and dentin hypersensitivity can be challenging. Herein, we present two cases of successful treatment procedure for multiple NCCLs with gingival recession and dentin hypersensitivity using an envelope coronally advanced flap with CTG and composite resin restoration. Through the combined approach of restorative and periodontal procedure, both patients showed adequate extent of gingival coverage and esthetic outcome based on the Modified Root Coverage Esthetic Score (MRES) at 6 months postoperatively. Also, dentin hypersensitivity was reduced effectively during the follow up period. Although the pocket depth slightly increased in patient 1, possibly due to the amount of restoration located sub-gingivally, pocket depth remained within 3 mm. This suggest that re-establishing the clinical CEJ and performing partial restoration is advantageous for periodontal tissue and is expected to contribute to maintain gingival height in the long term. These case reports emphasize the efficacy of the combined approach for treating multiple NCCLs with gingival recession and dentin hypersensitivity, highlighting the importance of careful restoration planning for optimal clinical and aesthetic outcomes.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.4
• /
• pp.47-52
• /
• 2009

The transmitted signal from a source is transmitted to a destination through wireless channels. But if the mobile destination is out of the coverage of the source or exists in the shady side of the coverage, the destination can not receiver the signal from the source and they can not maintain communication. In order to overcome these problems, we adopt relays. A system employing relays is a multi-hop relay system. In the multi-hop relay system, coverages of each relay that is used for different systems can overlap each other in some place. When there is a destination in this place, interference occurs at the destination. In this paper, we study on the efficient co-channel interference (CCI) cancellation algorithm. In the proposed strategy, CCI is mitigated by zero forcing (ZF) or minimum mean square error (MMSE) receivers. Moreover, successive interference cancellation (SIC) with optimal ordering algorithm is applied for rejecting CCI efficiently. And we analyzed and simulated the proposed system performance in Rayleigh fading channel. In order to justify the benefit of the proposed strategy, the overall system performance is illustrated in terms of bit error probability.