• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.6 no.3
• /
• pp.249-254
• /
• 2005

In this paper, we analysis the recognition performance of PCA/LDA by distance measures. We are adapt to ORL face database with the fourteen distance measures. In case of PCA, it has high performance for the manhattan distance and the weighted SSE distance to face recognition, In case of PCA/LDA, it has high performance for the angle-based distance and the modified SSE distance. Also, PCA/LDA is better than PCA for reduction of dimension. Therefore, the PCA/LDA method and the angle-based distance have the most performance and a few dimension for face recognition with ORL face database.

• Journal of KIISE:Software and Applications
• /
• v.26 no.9
• /
• pp.1114-1124
• /
• 1999

클러스터링은 동일한 클러스터에 속하는 데이타들 간에는 유사도가 크도록 하고 다른 클러스터에 속하는 데이타들 간에는 유사도가 작도록 주어진 데이타를 몇 개의 클러스터로 묶는 것이다. 어떤 대상을 기술하는 데이타는 수치 속성뿐만 아니라 정성적인 비수치 속성을 갖게 되고, 이들 속성값은 관측 오류, 불확실성, 주관적인 판정 등으로 인해서 정확한 값으로 주어지지 않고 애매한 값으로 주어지는 경우가 많다. 본 논문에서는 애매한 값을 퍼지값으로 표현하는 수치 속성과 비수치 속성을 포함한 데이타에 대한 비유사도 척도를 제안하고, 이 척도를 이용하여 퍼지값을 포함한 데이타에 대하여 퍼지 클러스터링하는 방법을 소개한 다음, 이를 이용한 실험 결과를 보인다. Abstract The objective of clustering is to group a set of data into some number of clusters in a way to minimize the similarity between data belonging to different clusters and to maximize the similarity between data belonging to the same cluster. Many data for real world objects consist of numeric attributes and non-numeric attributes whose values are fuzzily described due to observation error, uncertainty, subjective judgement, and so on. This paper proposes a dissimilarity measure applicable to such data and then introduces a fuzzy clustering method for such data using the proposed dissimilarity measure. It also presents some experiment results to show the applicability of the proposed clustering method and dissimilarity measure.

• Science of Emotion and Sensibility
• /
• v.26 no.2
• /
• pp.25-38
• /
• 2023

Among the discussions on affective representation, the first is to explain the affective representation in the dimensions, and the second is to explain the affective representation according to the modality. In previous studies, to explain affective representation, valence models (signed valence, unsigned valence) and Modality-generality models (modality-general, modality-specific) were presented. In this study, we compared models presented in the previous study using the recently published ASMR to confirm which models explain affective representation well. The data used in this study were behavioral rating values collected by Kim & Kim (2022), and these were obtained for ASMR stimuli that were divided into three affective types (negative, neutral, and positive) and two modalities (auditory and audiovisual). Then, a multidimensional scaling, a representational similarity analysis with a two-way repeated measures ANOVA, and a multiple regression analysis with a two-way repeated measures ANOVA were performed. The results revealed that signed valence and modality-general distinguished between affective types of stimuli better than unsigned valence and modality-specific. Similar to the results of multidimensional scaling, the results of a representational similarity analysis and a multiple regression also showed that the signed valence and modality-general significantly explained affective representation better than the unsigned valence and the modality-specific. These results suggest that the model in which positive and negative are located at the opposite ends of the one dimension explains the affective representation of ASMR well, and that the affective representation was consistent regardless of modality.

• The Journal of Korean Association of Computer Education
• /
• v.18 no.4
• /
• pp.55-62
• /
• 2015

Recommender systems through collaborative filtering has been utilized successfully in various areas by providing with convenience in searching information. Measuring similarity is critical in determining performance of these systems, because it is the criteria for the range of recommenders. This study analyzes distributions of similarity from traditional measures and investigates relations between similarities and the number of co-rated items. With this, this study suggests a method for selecting reliable recommenders by restricting similarities, which compensates for the drawbacks of previous measures. Experimental results showed that restricting similarities of neighbors by upper and lower thresholds yield superior performance than previous methods, especially when consulting fewer nearest neighbors. Maximum improvement of 0.047 for cosine similarity and that of 0.03 for Pearson was achieved. This result tells that a collaborative filtering system using Pearson or cosine similarities should not consult neighbors with very high or low similarities.

• Korean Journal of Cognitive Science
• /
• v.17 no.1
• /
• pp.53-74
• /
• 2006

Similarity has been considered as one of basic concepts of cognitive psychology which is useful for explaining cognitive structure and process. MDS models(Shepard, 1964; Nosofsky, 1991) and Contrast model(Tversky, 1977) were proposed as early models of similarity comparison process. But, there have been a lot of theoretical doubts about the conceptual validity of similarity as a result of empirical findings which could not be explained by early models. Goldstone(1994) assumed that similarity could be defined by alignment processes, and suggested structural alignment as a prospective alternative for solving conceptual controversies so far. In this study, basic assumption and algorithms of MDS models(Shepard, 1944; Nosofsky, 1991) and Contrast model(Tversky, 1977) were described shortly and some theoretical limitations such as arbitrariness of selective attention and correlated structures were discussed as well. The conceptual characteristics and algorithms of SIAM(Goldstone, 1994) were described and how it has been applied to cognitive psychology areas such as categorization, conceptual combination, and analogical reasoning were reviewed. Finally, some theoretical limitations related with data-driven processing and alternative processing and possible directions for structural alignment were discussed.

• Science of Emotion and Sensibility
• /
• v.5 no.1
• /
• pp.61-70
• /
• 2002

The purpose of this study was to develop 'Emotion Rating Scale induced by Odors'and to identify the structure of odor emotion induced by odors. At first 37 adjectives that describe odor to develop a rating scale were selected. Subjects were to rate odor emotion on a 7-point bipolar scale. 304 subjects participated and were as a group instructed to rate odor emotion. 53 out of 304 subjects were retested to test for reliability of the scale two weeks after under the same condition and finally 25 adjectives were then selected based on high test-retest reliability and factor loading, high contributing to one factor. 24 subjects each in 10s, 20s, 30s & 40s were to rate odor emotion induced by 5 different odors on the scale developed. The structure of odor emotion consisted of 'Esthetics', 'Intensity', 'Romance', 'Nature'and 'Character'. The structure of odor emotion by age appeared quite similar but that by different odors was little bit different.

• Journal of the Korea Convergence Society
• /
• v.9 no.10
• /
• pp.393-406
• /
• 2018

The depression and anxiety are the most popular mental disorders and not easy to distinguish because of their lots of similarities in the diagnostic criteria, related theories, and clinical symptoms. In this article, we developed the affect scale for distinguishable diagnosis, utilized the relationships between the Positive and Negative affect, and the depression and anxiety. We made up the seed scale of the items which selected by partial correlation, and set the scoring up by multiple regression method. The Multi-dimentional affect scale is reliable and working similarly as the scales used before, but less correlated to each other. We conclude that the affect scale achieved the diagnosis for distinguish between depression and anxiety. Our suggestions for the further study are to redeem the cultural differences, modify by the elaborate methods, and validate by the actual clinical data.

• Science of Emotion and Sensibility
• /
• v.2 no.1
• /
• pp.69-75
• /
• 1999

본 연구는 우리말 형용사를 이용하여 각종 생활공간음에 대해 느끼는 사람들이 감성적 반응의 차원을 밝히기 위한 기초 연구로 수행되었다. 형용사 사전에서 추출된 어휘들에 대한 적합성 및 유사성 평가를 통해 14개의 형용사가 최종적으로 선정되었고, 이 형용사들의 유사성 점수에 대해 요인분석, 다차원 척도 분석 및 군집분석을 시행하였다. 연구 결과 생활공간음에 대한 사람들의 감성적 반응은 크게 3가지 차원, 즉 쾌-불쾌, 민감-둔감 및 지각된 크기가 추출되었고, 소음에 대해 느끼는 감성은 이들 차원 상에서 측정될 수 있는 것으로 나타났다. 이들 차원들은 서로 직선적인 관계가 있는 것으로 보인다. 즉, 소음에 대해 민감한 사람은 그 소음에 대해 불쾌하거나 짜증을 일으키기 쉽지만 소음에 대해 불쾌하거나 짜즈을 일으킨 사람이 소음에 대해 민감한 사람이라고 볼 수는 없을 것이다. 따라서 이들 차원간에 구체적인 통로분석이 이루어져야 할 것으로 보이며 이는 후속연구에서 밝혀질 것이다. 또한 실제 소음을 제시하여 각각의 소음을 평가하는 경험적 연구도 수행될 것이다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.91-93
• /
• 2005

대용량 데이타에 대한 복잡한 질의 처리가 요구되는 응용에서 빠른 시간 안에 응답을 돌려주기 위해, 데이타를 작은 크기로 근사하여 질의를 처리하는 방법이 연구되고 있다. 빠른 응답을 위해 주어진 저장 공간의 제약 하에서 얼마나 원본 데이타와 유사하게 근사할 수 있는지가 데이타 근사의 성능을 결정한다. 본 논문에서는 데이타 근사에서 유사도의 척도 중 하나인 최대 상대 오차를 줄이기 위하여 Haar 웨이블릿을 변형한 조화 웨이블릿 기법을 제안한다. 조화 웨이블릿은 데이타 변환 과정 중 조화평균을 이용하여 상대 오차 정보를 손쉽게 얻어낼 수 있어 근사 데이타의 상대 오차를 최소화하는 데 적합한 기법이다.

• Science of Emotion and Sensibility
• /
• v.12 no.1
• /
• pp.1-10
• /
• 2009

The purpose of this study is to find out the relationship between the design factor and the sensitivity in web site. The classification of sensitivity-types consists of the research of books and the survey, and the language specialist's review and the analysis of factor. The research of the Image Scale accomplished through the analysis of the result of sensitivity-types. The major findings of the analysis are summarized as follows. The webpage sensitivity-types are classified into the 7 types, namely 'refreshment', 'calm', 'refinement', 'strongness', 'youth', 'uniqueness', 'futurity'. As a result of analyzing of similarity between the adjectives by multiple standards, the web site Image Scale space consists of the axis between 'heavy-light' and 'soft-hard'. As a result of the research of relationship between the web site design factor and the emotion, the color and the layout influenced into 'soft-hard' much, and the light and the color influenced into 'heavy-light' much.