• KIPS Transactions on Computer and Communication Systems
• /
• v.7 no.1
• /
• pp.1-8
• /
• 2018

It is not easy to learn simple expressions of moving picture data since it contains noise and a lot of information in addition to time-based information. In this study, we propose a similarity representation method and a deep learning method between sequential data which can express such video data abstractly and simpler. This is to learn and obtain a function that allow them to have maximum information when interpreting the degree of similarity between image data vectors constituting a moving picture. Through the actual data, it is confirmed that the proposed method shows better classification performance than the existing moving image classification methods.

• Korean Institute of Interior Design Journal
• /
• v.23 no.5
• /
• pp.139-145
• /
• 2014

As life is changing, modern people are interested in culture life more than the past and as a result Art Exhibition increases. Nowadays many methods of exhibit are introduced, especially Direct experience. Trick Art Experience exhibition introduced Direct experirence to a work of art for interests and used props, objets, images, and sounds for maximizing experience element. Trick Art Experience exhibition added humor as imitating famous painting and gave a deep impression through the optical illusion technique which make flat structure look three-dimensional structure. Recently, demands for Experience exhibition which are like Trick Art increase. This study figure out pros and cons through a comparison analysis of direction and component about Trick Art Experience exhibition and find a check list about making an audience active about experience. From now on, we have an object about showing design data including customer satisfaction measurement to make various expressions of direction and component about Experience Art through a case analysis. We have firsthand visited case spaces to experience the spaces and prepared a check list for the relevant types of expressions. Referring to the exhibit work classification system, components that correspond to Trick Art Experience exhibition were grouped according to constituting items. Among the types of expressions, the levels of satisfaction with Experience exhibition and Participating exhibit were found to be high and the level of satisfaction with Heart Moving exhibit was found to be low through surveys. Among the components, the levels of satisfaction with stereograms and formative effect elements were found to be high and those of spatial changes and media effects were found to be low through surveys. Therefore, the expressions should be reinforced and the components should be improved.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.12
• /
• pp.1595-1603
• /
• 2020

Online defamation incidents such as Internet news comments on portal sites, SNS, and community sites are increasing in recent years. Bias and hate expressions threaten online service users in various forms, such as invasion of privacy and personal attacks, and defamation issues. In the past few years, academia and industry have been approaching in various ways to solve this problem The purpose of this study is to build a dataset and experiment with deep learning classification modeling for detecting various bias expressions as well as hate expressions. The dataset was annotated 7 labels that 10 personnel cross-checked. In this study, each of the 7 classes in a dataset of about 137,111 Korean internet news comments is binary classified and analyzed through deep learning techniques. The Proposed technique used in this study is multi-channel CNN model with attention. As a result of the experiment, the weighted average f1 score was 70.32% of performance.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.3
• /
• pp.747-752
• /
• 2011

Microarray classification typically possesses two striking attributes: (1) classifier design and error estimation are based on remarkably small samples and (2) cross-validation error estimation is employed in the majority of the papers. A Microarray data of ovarian cancer consists of the expressions of thens of thousands of genes, and there is no systematic procedure to analyze this information instantaneously. In this paper, gene markers are selected by ranking genes according to statistics, popular classification rules - linear discriminant analysis, k-nearest-neighbor and decision trees - has been performed comparing classification accuracy of data selecting gene markers and not selecting gene markers. The Result that apply linear classification analysis at Microarray data set including marker gene that are selected using ANOVA method represent the highest classification accuracy of 97.78% and the lowest prediction error estimate.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10a
• /
• pp.598-600
• /
• 2001

DNA chip technology enables us to monitor thousands of gene expressions per sample simultaneously. Typically, DNA microarray data has at least several thousands of variables (genes) wish relatively smal1 number of samples. Thus feature (gene) selection by dimensionality reduction is necessary for efficient data analysis. In this paper we employ the partial least squares (PLS) method for gene selection and the principal component analysis (PCA) method for classification. The useful behavior of the PLS is verified by computer simulations.

• Genomics & Informatics
• /
• v.3 no.2
• /
• pp.55-60
• /
• 2005

Peroxiredoxins (Prx's) are a superfamily of peroxidases that are ubiquitous in all super-kingdoms. Previous biochemical and structural studies have suggested that Prx's could be divided into five subfamilies (1-Cys, Typical 2-Cys, Atypical 2-Cys C-, L- and R- types). In this work, we have developed a set of regular expression patterns describing subfamily-specific spatial constraints of the key catalytic residues. Using these patterns, 1,016 Prx's available in public databases were classified into the five subfamilies. Our method performed well for most of the types except for Atypical 2 Cys R type.

• Journal of People, Plants, and Environment
• /
• v.22 no.2
• /
• pp.209-217
• /
• 2019

Generally human sensibility is expressed in a certain language. To discover the sensibility of visitors in relation to the forest environment, it is first necessary to determine their exact meanings. Furthermore, it is necessary to sort these terms according to their meanings based on an appropriate classification system. This study attempted to develop a classification system for forest sensibility vocabulary by extracting Korean words used by forest visitors to express their sensibilities in relation to the forest environment, and established the structure of the system to classify the accumulated vocabulary. For this purpose, we extracted forest sensibility words based on literature review of experiences reported in the past as well as interviews of forest visitors, and categorized the words by meanings using the Standard Korean Language Dictionary maintained by the National Institute of the Korean Language. Next, the classification system for these words was established with reference to the classification system for vocabulary in the Korean language examined in previous studies of Korean language and literature. As a result, 137 forest sensibility words were collected using a documentary survey, and we categorized these words into four types: emotion, sense, evaluation, and existence. Categorizing the collected forest sensibility words based on this Korean language classification system resulted in the extraction of 40 representative sensibility words. This experiment enabled us to determine from where our sensibilities that find expressions in the forest are derived, that is, from sight, hearing, smell, taste, or touch, along with various other aspects of how our human sensibilities are expressed such as whether the subject of a word is person-centered or object-centered. We believe that the results of this study can serve as foundational data about forest sensibility.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.1 no.1
• /
• pp.45-48
• /
• 1978

If two inspectors classify items in a lot into m classes, it is possible that each of them makes wrong classification in some cases, thus causing bias. Expressions have been obtained for the limits of this bias in estimating the proportion of the different classes. From the results of the classification they obtained limit for the estimates of Proportions have been worked out, based on assumption regarding the magnitudes of probabilities of misclassification. Now we suppose that $P_{ti}{\;}(t=1.2)$ is the probability that t the inspector classifies correctly an item in class $A_i$ and $q_{tji}$ is the probability that he misclassifies in $A_j$ an item actually belonging to $A_i$, therefor, $P_{ti}+ \sum\limits_{j{\neq}i}q_{tji}=1$ An estimate for the proportion $P_k$ of the class $A_k$ in the lot would be $\hat{P}_k=r_{kk}+(\frac{1}{2})\sum\limits_{j{\neq}k}r_{kj}+r_{jk}$ The % Bias in proportion $\hat{P}_k$ is $\frac{E(\hat{P}_k)-P_k}{P_k}{\times}100$

• Journal of Information Processing Systems
• /
• v.1 no.1 s.1
• /
• pp.22-26
• /
• 2005

In this paper, we classified the objectionable texts into four rates according to their harmfulness and proposed the hierarchical text rating system for objectionable documents. Since the documents in the same category have similarities in used words, expressions and structure of the document, the text rating system, which uses a single classification model, has low accuracy. To solve this problem, we separate objectionable documents into several subsets by using their properties, and then classify the subsets hierarchically. The proposed system consists of three layers. In each layer, we select features using the chi-square statistics, and then the weight of the features, which is calculated by using the TF-IDF weighting scheme, is used as an input of the non-linear SVM classifier. By means of a hierarchical scheme using the different features and the different number of features in each layer, we can characterize the objectionability of documents more effectively and expect to improve the performance of the rating system. We compared the performance of the proposed system and performance of several text rating systems and experimental results show that the proposed system can archive an excellent classification performance.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.539-554
• /
• 2018

Effective interaction between user and device is considered an important ability of IoT devices. For some applications, it is necessary to recognize human facial expressions in real time and make accurate judgments in order to respond to situations correctly. Therefore, many researches on facial image analysis have been preceded in order to construct a more accurate and faster recognition system. In this study, we constructed an automatic recognition system for facial expressions through two steps - a facial recognition step and a classification step. We compared various models with different sets of data with pixel information, landmark coordinates, Euclidean distances among landmark points, and arctangent angles. We found a fast and efficient prediction model with only 30 principal components of face landmark information. We applied several prediction models, that included linear discriminant analysis (LDA), random forests, support vector machine (SVM), and bagging; consequently, an SVM model gives the best result. The LDA model gives the second best prediction accuracy but it can fit and predict data faster than SVM and other methods. Finally, we compared our method to Microsoft Azure Emotion API and Convolution Neural Network (CNN). Our method gives a very competitive result.