• Korean Journal of Cognitive Science
• /
• v.13 no.3
• /
• pp.1-10
• /
• 2002

This study was performed to analyze the physical attributes of the faces and affective words on the fares. For analyzing physical attributes inside of a face, 36 facial features were selected and almost of them were the lengths or distance values. For analyzing facial contour 14 points were selected and the lengths from nose-end to them were measured. The values of these features except ratio values normalized by facial vortical length or facial horizontal length because the face size of each person is different. The principal component analysis (PCA) was performed and four major factors were extracted: 'facial contour' component, 'vortical length of eye' component, 'facial width' component, 'eyebrow region' component. We supposed the five-dimensional imaginary space of faces using factor scores of PCA, and selected representative faces evenly in this space. On the other hand, the affective words on faces were collected from magazines and through surveys. The factor analysis and multidimensional scaling method were performed and two orthogonal dimensions for the affections on faces were suggested: babyish-mature and sharp-soft.

• Journal of KIISE:Software and Applications
• /
• v.37 no.7
• /
• pp.548-557
• /
• 2010

In this paper, we propose a semi-automatic modeling approach of ontology to annotate VOD to realize the IPTV's intelligent searching. The ontology is made by combining partial tree that extracts hypernym, hyponym, and synonym of keywords related to a service domain from WordNet. Further, we add to the partial tree new keywords that are undefined in WordNet, such as foreign words and words written in Chinese characters. The ontology consists of two parts: generic hierarchy and specific hierarchy. The former is the semantic model of vocabularies such as keywords and contents of keywords. They are defined as classes including property restrictions in the ontology. The latter is generated using the reasoning technique by inferring contents of keywords based on the generic hierarchy. An annotation generates metadata (i.e., contents and genre) of VOD based on the specific hierarchy. The generic hierarchy can be applied to other domains, and the specific hierarchy helps modeling the ontology to fit the service domain. This approach is proved as good to generate metadata independent of any specific domain. As a result, the proposed method produced around 82% precision with 2,400 VOD annotation test data.

• Science of Emotion and Sensibility
• /
• v.11 no.2
• /
• pp.285-292
• /
• 2008

The purpose of this study is to provide the fundamental data needed in analyzing customer needs and understanding products in developing designs, to help designers to have better understanding of digital camera products, and to support the setting of concepts in developing designs, by understanding the specific properties of products that have specific purposes. In this study, homogeneity analysis was performed to the collected products launched from 2000 until now and representative products were selected to extract the questions on the adjectives and preferences felt form such products. Based on the questions, basic questionnaire survey and subject image analysis was performed in relation to the elements of images preferred by customers through the regression analysis of dependent variables and preferences and the regression analysis of purchasing power. When we design for digital camera, we must consider about the elements of digital cameras and the terms convenient, sensitive, functional, and grace. In terms of whole trend of shape, the shape highlighting grips and the digital cameras having grips, large LCD, dark colors, and manual buttons were preferred.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.785-790
• /
• 2009

Widgets are the instrument for representing user's character and embossing the value of blogs. The compound word of the Windows and Gadget the application, widgets are the functional program to displayed on the screen graphical user interface (GUI) tools as a kind of service that user want to see. On the operating system, the Web, and mobile area, widgets offer the delivery of information, convenience and efficiency. However widgets have been never gave satisfaction to user because it focused transmitting information and representing circumstance than fun. This study is for recognized fun elements that user feel interest and categorized fun elements each type of widgets. Fun elements of widget never been defined, we use fun elements on design and product area and emotional word that is representative of affectivity. And we make up an online questionnaire to blog users. The widget selected by popular degree among the domestic widgets and the Japanese widget. And the results of the questionnaire that 5-scales used based on user preferences to identify the elements that are fun.

• Korean Journal of Cognitive Science
• /
• v.19 no.4
• /
• pp.499-517
• /
• 2008

In this paper, we propose an effective emotion feature extraction method for Korean and evaluate their availability in sentiment classification. Korean emotion features are expanded from several representative emotion words and they play an important role in building in an effective sentiment classification system. Firstly, synonym information of English word thesaurus is used to extract effective emotion features and then the extracted English emotion features are translated into Korean. To evaluate the extracted Korean emotion features, we represent each document using the extracted features and classify it using SVM(Support Vector Machine). In experimental results, the sentiment classification system using the extracted Korean emotion features obtained more improved performance(14.1%) than the system using content-words based features which have generally used in common text classification systems.

• Journal of Digital Contents Society
• /
• v.18 no.3
• /
• pp.585-592
• /
• 2017

As development of personal devices have made everyday use of internet much easier than before, it is getting generalized to find information and share it through the social media. In particular, communities specialized in each field have become so powerful that they can significantly influence our society. Finally, businesses and governments pay attentions to reflecting their opinions in their strategies. The stock market fluctuates with various factors of society. In order to consider social trends, many studies have tried making use of bigdata analysis on stock market researches as well as traditional approaches using buzz amount. In the example at the top, the studies using text data such as newspaper articles are being published. In this paper, we analyzed the post of 'Paxnet', a securities specialists' site, to supplement the limitation of the news. Based on this, we help researchers analyze the sentiment of investors by generating a domain-specific sentiment lexicon for the stock market.

• Science of Emotion and Sensibility
• /
• v.24 no.3
• /
• pp.91-114
• /
• 2021

The first step of this study is to collect appropriate words from the list of words in the relationship. All vocabularies that are unfamiliar-but capable of guessing the meaning and expressing interpersonal relationships-were collected from three Korean dictionaries. Consequently, a compilation of 2,725 words was created; overlapping words were selected; and 910 words were chosen. Only grammatical forms were found; however, words with similar meanings-or identical meanings-were also found, and a reclassification process was required to reflect this. These procedures were repeated seven times, resulting in a total of 249 words being screened. However, due to the characteristics of this study, the number of words needs to be reduced because the meaning of words is more specific and summarized, and the overall interpersonal aspect is well expressed. Therefore, the process of reclassifying 249 words by their familiarity and appropriateness was subsequently undertaken, and the word with the highest level of familiarity and appropriateness was finally selected.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.13 no.2
• /
• pp.102-110
• /
• 2017

Current word sense disambiguation techniques employ various machine learning-based methods. Various approaches have been proposed to address this problem, including the knowledge base approach. This approach defines the sense of an ambiguous word in accordance with knowledge base information with no training corpus. In unsupervised learning techniques that use a knowledge base approach, graph-based and similarity-based methods have been the main research areas. The graph-based method has the advantage of constructing a semantic graph that delineates all paths between different senses that an ambiguous word may have. However, unnecessary semantic paths may be introduced, thereby increasing the risk of errors. To solve this problem and construct a fine-grained graph, in this paper, we propose a model that iteratively constructs the graph while eliminating unnecessary nodes and edges, i.e., senses and semantic paths. The hybrid similarity estimation model was applied to estimate a more accurate sense in the constructed semantic graph. Because the proposed model uses BabelNet, a multilingual lexical knowledge base, the model is not limited to a specific language.

• 한국정보교육학회:학술대회논문집
• /
• 2021.08a
• /
• pp.141-148
• /
• 2021

The mainstream tool for software education for elementary students is Educational Programming Language. It is essential for upper graders to advance from EPL to text based programming language. However, many students experience difficulty in adopting to this change since Python is run in English. Python is an actively used TPL. This study focuses on developing an education program to facilitate learning Python for Korean speaking students. We have extracted the necessary reserved words needed for data analysis in Python. Then we replaced the extracted words into Korean terms that could be understood in elementary level. The replaced terms were matched on one-to-one correspondence with reserved words used in Python. This devised program would assist students in experiencing data analysis with Python. We expect that this education program will be applied effectively as a basic resource to learn TPL.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.197-202
• /
• 2019

토큰화는 입력 텍스트를 더 작은 단위의 텍스트로 분절하는 과정으로 주로 기계 학습 과정의 효율화를 위해 수행되는 전처리 작업이다. 현재까지 자연어 처리 분야 과업에 적용하기 위해 다양한 토큰화 방법이 제안되어 왔으나, 주로 텍스트를 효율적으로 분절하는데 초점을 맞춘 연구만이 이루어져 왔을 뿐, 한국어 데이터를 대상으로 최신 기계 학습 기법을 적용하고자 할 때 적합한 토큰화 방법이 무엇일지 탐구 해보기 위한 연구는 거의 이루어지지 않았다. 본 논문에서는 한국어 데이터를 대상으로 최신 기계 학습 기법인 전이 학습 기반의 자연어 처리 방법론을 적용하는데 있어 가장 적합한 토큰화 방법이 무엇인지 알아보기 위한 탐구 연구를 진행했다. 실험을 위해서는 대표적인 전이 학습 모형이면서 가장 좋은 성능을 보이고 있는 모형인 BERT를 이용했으며, 최종 성능 비교를 위해 토큰화 방법에 따라 성능이 크게 좌우되는 과업 중 하나인 기계 독해 과업을 채택했다. 비교 실험을 위한 토큰화 방법으로는 통상적으로 사용되는 음절, 어절, 형태소 단위뿐만 아니라 최근 각광을 받고 있는 토큰화 방식인 Byte Pair Encoding (BPE)를 채택했으며, 이와 더불어 새로운 토큰화 방법인 형태소 분절 단위 위에 BPE를 적용하는 혼합 토큰화 방법을 제안 한 뒤 성능 비교를 실시했다. 실험 결과, 어휘집 축소 효과 및 언어 모델의 퍼플렉시티 관점에서는 음절 단위 토큰화가 우수한 성능을 보였으나, 토큰 자체의 의미 내포 능력이 중요한 기계 독해 과업의 경우 형태소 단위의 토큰화가 우수한 성능을 보임을 확인할 수 있었다. 또한, BPE 토큰화가 종합적으로 우수한 성능을 보이는 가운데, 본 연구에서 새로이 제안한 형태소 분절과 BPE를 동시에 이용하는 혼합 토큰화 방법이 가장 우수한 성능을 보임을 확인할 수 있었다.