• Journal of Computing Science and Engineering
• /
• v.9 no.4
• /
• pp.204-213
• /
• 2015

Under the conditional independence assumption among local features, the Naive Bayes Nearest Neighbor (NBNN) classifier has been recently proposed and performs classification without any training or quantization phases. While the original NBNN shows high classification accuracy without adopting an explicit training phase, the conditional independence among local features is against the compositionality of objects indicating that different, but related parts of an object appear together. As a result, the assumption of the conditional independence weakens the accuracy of classification techniques based on NBNN. In this work, we look into this issue, and propose a novel Bayesian network for an NBNN based classification to consider the conditional dependence among features. To achieve our goal, we extract a high-level feature and its corresponding, multiple low-level features for each image patch. We then represent them based on a simple, two-level layered Bayesian network, and design its classification function considering our Bayesian network. To achieve low memory requirement and fast query-time performance, we further optimize our representation and classification function, named relation-based Bayesian network, by considering and representing the relationship between a high-level feature and its low-level features into a compact relation vector, whose dimensionality is the same as the number of low-level features, e.g., four elements in our tests. We have demonstrated the benefits of our method over the original NBNN and its recent improvement, and local NBNN in two different benchmarks. Our method shows improved accuracy, up to 27% against the tested methods. This high accuracy is mainly due to consideration of the conditional dependences between high-level and its corresponding low-level features.

• The Journal of the Korea Contents Association
• /
• v.20 no.7
• /
• pp.525-533
• /
• 2020

In recent years, the VR-interactive action contents industry, which utilizes five senses of human bodies, has continued to grow through areas such as games, tourism, movies, performances and exhibitions, but it has reached to breaking point by unrealistic visual elements. Therefore, the purpose of this study is to analyze the effect of each evaluation factor based on visual immersion of interactive type of VR action contents to overcome the limitations. For this study, firstly, prior research is reviewed. Secondly, the evaluation factors of visual immersion of interactive type of VR action contents and hypothesis are to be derived. Research finding is that there is no difference to recognize proximity, three-dimensionality, visibility and immersion by gender. Also, in order to influence visual immersion, it is important that 3D modeling of characters and objects must be sophisticated to be fit in with their surroundings and lighting. This makes user to be confused where they are actually in.

• Journal of the Korea Computer Industry Society
• /
• v.4 no.12
• /
• pp.887-900
• /
• 2003

Many applications require the clustering of large amounts of high dimensional data. Most automated clustering techniques have been developed but they do not work effectively and/or efficiently on high dimensional (numerical) data, which is due to the so-called “curse of dimensionality”. Moreover, the high dimensional data often contain a significant amount of noise, which causes additional ineffectiveness of algorithms. Therefore, it is necessary to look over the structure and various characteristics of high dimensional data and to develop algorithm that support clustering adapted to applications of the high dimensional database. In this paper, we investigate and classify the existing high dimensional clustering methods by analyzing the strength and weakness of each method for specific applications and comparing them. Especially, in terms of efficiency and effectiveness, we compare the traditional algorithms with CLIP which are developed by us. This study will contribute to develop more advanced algorithms than the current algorithms.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.4
• /
• pp.427-431
• /
• 2010

Multidimensional scaling (MDS) is a widely used method for dimensionality reduction, of which purpose is to represent high-dimensional data in a low-dimensional space while preserving distances among objects as much as possible. MDS has mainly been applied to data visualization and feature selection. Among various MDS methods, the classical MDS is not readily applicable to data which has large numbers of objects, on normal desktop computers due to its computational complexity. More precisely, it needs to solve eigenpair problems on dissimilarity matrices based on Euclidean distance. Thus, running time and required memory of the classical MDS highly increase as n (the number of objects) grows up, restricting its use in large-scale domains. In this paper, we propose an efficient approximation algorithm for the classical MDS based on divide-and-conquer and CUDA. Through a set of experiments, we show that our approach is highly efficient and effective for analysis and visualization of data consisting of several thousands of objects.

• Korean Journal of Cognitive Science
• /
• v.23 no.3
• /
• pp.295-322
• /
• 2012

In this paper, we cluster similar word senses applying vector space model and HAL (Hyperspace Analog to Language). HAL measures corelation among words through a certain size of context (Lund and Burgess 1996). The similarity measurement between a word pair is cosine similarity based on the vector space model, which reduces distortion of space between high frequency words and low frequency words (Salton et al. 1975, Widdows 2004). We use PCA (Principal Component Analysis) and SVD (Singular Value Decomposition) to reduce a large amount of dimensions caused by similarity matrix. For sense similarity clustering, we adopt supervised and non-supervised learning methods. For non-supervised method, we use clustering. For supervised method, we use SVM (Support Vector Machine), Naive Bayes Classifier, and Maximum Entropy Method.

• Spatial Information Research
• /
• v.8 no.1
• /
• pp.171-182
• /
• 2000

Random sampling or systematic sampling method is commonly used to assess the accuracy of classification results. In remote sensing, with these sampling method, much time and tedious works are required to acquire sufficient ground truth data. So , a more effective sampling method that can retain the characteristics of the population is required. In this study, fractal analysis is adopted as an index for reference sampling . The fractal dimensions of the whole study area and the sub-regions are calculated to choose sub-regions that have the most similar dimensionality to that of whole-area. Then the whole -area s classification accuracy is compared to those of sub-regions, respectively, and it is verified that the accuracies of selected sub regions are similar to that of full-area . Using the above procedure, a new kind of reference sampling method is proposed. The result shows that it is possible to reduced sampling area and sample size keeping up the same results as existing methods in accuracy tests. Thus, the proposed method is proved cost-effective for reference data sampling.

• Journal of Global Scholars of Marketing Science
• /
• v.17 no.4
• /
• pp.223-245
• /
• 2007

Purpose: While empirical importance of dimensionality of perceived value is widely accepted, our understanding of experiential value dimensions in other retail channels and other cultures has not been explicitly tested. This study attempted to determine if the dimensions of experiential value scale (EVS) by Mathwick, Malhotra, and Rigdon (2001) identified in US catalog and Internet contexts could be applied in other international markets (South Korea) and in other retail channels (department store versus Internet shopping mall). Methodology/Approach: Two data sets, one from 220 department store shoppers and the other from 359 Internet shopping mall shoppers, were analyzed. Findings: Confirmatory factor analysis confirmed four different EVS dimensions by retail channels. Overall, entertainment and intrinsic enjoyment values were found to be more important in department store while economic and efficiency value dimensions were interpreted critical in Internet shopping mall context. Visual appeal aspect constitutes distinct value dimension in two channels. Practical Implications: One separate dimension of time efficiency in Internet shopping mall suggests that more efficient web design and functions that can save time and promote convenience are needed to better accommodate their customers. Internet has heavily relied on traditional attributes, such as factual information, price comparability, and brand name reliance. However, this study suggests that Internet shopping mall retailer should offer visual diversion and stimulation just as brick and mortar shopping malls do. Originality /Value of Paper: Although the research findings must be viewed as tentative because the results are from one country, they provide a rich basis for further understanding the dimensions of experiential value in other international markets and other retail channels. Category: Research Paper

• Journal of Digital Convergence
• /
• v.15 no.10
• /
• pp.183-191
• /
• 2017

This study tries to explain the relationships among omni-channel service characteristics, utilitarian/hedonic shopping value, and reuse intention. We derive instant connectivity, localization, consistency, integration, privacy risk as omni-channel service characteristics from previous studies and collect data from 190 omni-channel service users. The major findings are as follows. First, localization, consistency, and privacy risk have a significant effect on utilitarian shopping value but no significant effect on hedonic shopping value. Second, instant connectivity and integration have a positive effect on both utilitarian and hedonic value. Third, utilitarian and hedonic shopping value have a positive effect on reuse intention. This study extends the scope of omni-channel consumer behaviors by focusing on multi-dimensionality of shopping value. The results of this research can provide useful implications for practitioners to build successful strategies on omni-channel service.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.4
• /
• pp.2124-2148
• /
• 2017

Accurate facial expression recognition (FER) requires reliable signal filtering and the effective feature extraction. Considering these requirements, this paper presents a novel approach for FER which is robust to noise. The main contributions of this work are: First, to preserve texture details in facial expression images and remove image noise, we improved the anisotropic diffusion filter by adjusting the diffusion coefficient according to two factors, namely, the gray value difference between the object and the background and the gradient magnitude of object. The improved filter can effectively distinguish facial muscle deformation and facial noise in face images. Second, to further improve robustness, we propose a new feature descriptor based on a combination of the Histogram of Oriented Gradients with the Canny operator (Canny-HOG) which can represent the precise deformation of eyes, eyebrows and lips for FER. Third, Canny-HOG's block and cell sizes are adjusted to reduce feature dimensionality and make the classifier less prone to overfitting. Our method was tested on images from the JAFFE and CK databases. Experimental results in L-O-Sam-O and L-O-Sub-O modes demonstrated the effectiveness of the proposed method. Meanwhile, the recognition rate of this method is not significantly affected in the presence of Gaussian noise and salt-and-pepper noise conditions.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.2
• /
• pp.142-149
• /
• 2004

It is important to retrieve exact information coinciding with user's need from lots of Usenet news and filter desired information quickly. Differently from email system, we must previously register our interesting news group if we want to get the news information. However, it is not easy for a novice to decide which news group is relevant to his or her interests. In this work, we present a service classifying user preferred news groups among various news groups by the use of Kohonen network. We first extract candidate terms from example documents and then choose a number of representative keywords to be used in Kohonen network from them through fuzzy inference. From the observation of training patterns, we could find the sparsity problem that lots of keywords in training patterns are empty. Thus, a new method to train neural network through reduction of unnecessary dimensions by the statistical coefficient of determination is proposed in this paper. Experimental results show that the proposed method is superior to the method using every dimension in terms of cluster overlap defined by using within cluster distance and between cluster distance.