• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.10C
• /
• pp.965-974
• /
• 2007

In this paper, we proposed the feature selection algorithm for multi-class genre classification. In our proposed algorithm, we developed GMM separation score based on Gaussian mixture model for measuring separability between two genres. Additionally, we improved feature subset selection algorithm based on sequential forward selection for multi-class genre classification. Instead of setting criterion as entire genre separability measures, we set criterion as worst genre separability measure for each sequential selection step. In order to assess the performance proposed algorithm, we extracted various features which represent characteristics such as timbre, rhythm, pitch and so on. Then, we investigate classification performance by GMM classifier and k-NN classifier for selected features using conventional algorithm and proposed algorithm. Proposed algorithm showed improved performance in classification accuracy up to 10 percent for classification experiments of low dimension feature vector especially.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.6C
• /
• pp.560-569
• /
• 2005

This paper proposes an algorithm for transmit antenna selection in a multi-input multi-output(MIMO) spatial multiplexing system with per antenna rate control(PARC) and an ordered successive interference cancellation (OSIC) receiver. The active antenna subset is determined at the receiver and conveyed to the transmitter using feedback information on transmission rate per antenna. We propose a serial decision procedure consisting of a successive process that tests whether antenna selection gain exists when the antenna with the lowest pre-processing signal to interference and noise ratio(SINR) is discarded at each stage. Furthermore, we show that 'reverse detection ordering', whereby the signal with the lowest SINR is decoded at each stage of successive decoding, widens the disparities among fractions of the whole capacity allocated to each individual antenna and thus maximizes a gain of antenna selection. Numerical results show that the proposed reverse detection ordering based serial antenna selection approaches the closed-loop MIMO capacity and that it induces a negligible capacity loss compared with the heuristic selection strategy even with considerably reduced complexity.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.27 no.1
• /
• pp.103-108
• /
• 2023

We study the problem of selecting a subset of sensor nodes in sensor networks in order to maximize the performance of parameter estimation. To achieve a low-complexity sensor selection algorithm, we propose a greedy iterative algorithm that allows us to select one sensor node at a time so as to maximize the log-determinant of the inverse of the estimation error covariance matrix without resort to direct minimization of the estimation error. We apply QR factorization to the observation matrix in the log-determinant to derive an analytic selection rule which enables a fast selection of the next node at each iteration. We conduct the extensive experiments to show that the proposed algorithm offers a competitive performance in terms of estimation performance and complexity as compared with previous sensor selection techniques and provides a practical solution to the selection problem for various network applications.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.7C
• /
• pp.481-487
• /
• 2011

Cyclic delay diversity (CDD) is a simple transmit diversity technique for an OFDM system using multiple transmit antennas. However, the performance of post-FFT estimation, i.e., integer frequency offset (lFO) is deteriorated by high frequency selectivity introduced by CDD. In this paper, the IFO estimation scheme is proposed for OFDM-based DRM+ system with CDD. Based on the pilot subset partitioning, the proposed IFO estimation scheme reduces the effect of performance degradation caused by frequency selectivity in OFDM systems with CDD . The simulation results show that the performance of the proposed IFO estimator is significantly improved when compared to that of the conventional IFO estimator.

• Proceedings of the KIEE Conference
• /
• 1998.07g
• /
• pp.2330-2332
• /
• 1998

This paper suggests a method to recognize the various defect patterns of cold mill strip using binary decision tree automatically constructed by genetic algorithm. The genetic algorithm and K-means algorithm were used to select a subset of the suitable features at each node in binary decision tree. The feature subset with maximum fitness is chosen and the patterns are classified into two classes by a linear decision boundary. This process was repeated at each node until all the patterns are classified into individual classes. The final recognizer is accomplished by neural network learning of a set of standard patterns at each node. Binary decision tree classifier was applied to the recognition of the defect patterns of cold mill strip and the experimental results were given to demonstrate the usefulness of the proposed scheme.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2005.09a
• /
• pp.57-62
• /
• 2005

Microarray gene expression profiling technology is one of the most important research topics in clinical diagnosis of disease. Given thousands of genes, only a small number of them show strong correlation with a certain phenotype. To identify such an optimal subset from thousands of genes is intractable, which plays a crucial role when classify multiple-class genes express models from tumor samples. This paper proposes an efficient classifier design method to simultaneously select the most relevant genes using an intelligent genetic algorithm (IGA) and design an accurate classifier using Support Vector Machine (SVM). IGA with an intelligent crossover operation based on orthogonal experimental design can efficiently solve large-scale parameter optimization problems. Therefore, the parameters of SVM as well as the binary parameters for gene selection are all encoded in a chromosome to achieve simultaneous optimization of gene selection and the associated SVM for accurate tumor classification. The effectiveness of the proposed method IGA/SVM is evaluated using four benchmark datasets. It is shown by computer simulation that IGA/SVM performs better than the existing method in terms of classification accuracy.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1241-1241
• /
• 2001

We propose a new calibration method, which uses the linearization method for spectral responses and the repetitive adoptions of the linearization weight matrices to construct a frature. Weight matrices are estimated through multiple linear regression (or principal component regression or partial least squares) with forward variable selection. The proposed method is applied to three data sets. The first is FTIR spectral data set for FeO content from sinter process and the second is NIR spectra from trans-alkylation process having two constituent variables. The third is NIR spectra of crude oil with three physical property variables. To see the calibration performance, we compare the new method with the PLS. It is found that the new method gives a little better performance than the PLS and the calibration result is stable in spite of the collinearity among each selected spectral responses. Furthermore, doing the repetitive adoptions of linearization matrices in the proposed methods, uninformative variables are disregarded. That is, the new methods include the effect of variables subset selection, simultaneously.

• Journal of Information Technology Services
• /
• v.12 no.3
• /
• pp.23-37
• /
• 2013

Social commerce is a subset of e-commerce that involves using social media, and user contributions to assist in the online buying and selling of products and services. Given the rapid growth of social commerce sites such as Groupon, Ticketmonster, and Coupang, it has become critical to understand customer purchasing decision-making processes in the social commerce environment. This study developed a theoretical model to examine the role of social commerce site's stickiness in customers' repurchasing decision processes. This study identifies price attribute, variety of selection, shopping enjoyment, and anger as the key factors of social commerce site's stickiness. Data collected from 164 users who had more purchasing experiences with social commerce for more than 7 months were empirically tested against the research model. The analysis results indicate that social commerce site's stickiness plays an important role in enhancing customer's purchasing behavior. Moreover, price attribute and shopping enjoyment significantly influence social commerce site's stickiness, whereas anger does not significantly affect consumer purchasing decision-making processes. However, contrary to our expectation, variety of selection negatively influences social commerce site's stickiness. The theoretical and practical implications of the findings are described.

• Communications for Statistical Applications and Methods
• /
• v.24 no.4
• /
• pp.383-396
• /
• 2017

The model in our approach assumes that computer responses are a realization of a Gaussian processes superimposed on a regression model called a Gaussian process regression model (GPRM). Selecting a subset of variables or building a good reduced model in classical regression is an important process to identify variables influential to responses and for further analysis such as prediction or classification. One reason to select some variables in the prediction aspect is to prevent the over-fitting or under-fitting to data. The same reasoning and approach can be applicable to GPRM. However, only a few works on the variable selection in GPRM were done. In this paper, we propose a new algorithm to build a good prediction model among some GPRMs. It is a post-work of the algorithm that includes the Welch method suggested by previous researchers. The proposed algorithms select some non-zero regression coefficients (${\beta}^{\prime}s$) using forward and backward methods along with the Lasso guided approach. During this process, the fixed were covariance parameters (${\theta}^{\prime}s$) that were pre-selected by the Welch algorithm. We illustrated the superiority of our proposed models over the Welch method and non-selection models using four test functions and one real data example. Future extensions are also discussed.

• IMMUNE NETWORK
• /
• v.15 no.3
• /
• pp.111-120
• /
• 2015

Dendritic cells (DCs) play a significant role in establishing self-tolerance through their ability to present self-antigens to developing T cells in the thymus. DCs are predominantly localized in the medullary region of thymus and present a broad range of self-antigens, which include tissue-restricted antigens expressed and transferred from medullary thymic epithelial cells, circulating antigens directly captured by thymic DCs through coticomedullary junction blood vessels, and peripheral tissue antigens captured and transported by peripheral tissue DCs homing to the thymus. When antigen-presenting DCs make a high affinity interaction with antigen-specific thymocytes, this interaction drives the interacting thymocytes to death, a process often referred to as negative selection, which fundamentally blocks the self-reactive thymocytes from differentiating into mature T cells. Alternatively, the interacting thymocytes differentiate into the regulatory T (Treg) cells, a distinct T cell subset with potent immune suppressive activities. The specific mechanisms by which thymic DCs differentiate Treg cells have been proposed by several laboratories. Here, we review the literatures that elucidate the contribution of thymic DCs to negative selection and Treg cell differentiation, and discusses its potential mechanisms and future directions.