• The Journal of the Korea Contents Association
• /
• v.12 no.11
• /
• pp.30-38
• /
• 2012

Sentiment analysis is concerned with extracting and analyzing different kinds of user sentiment expressed in a variety of social media such as blog and twitter. Although sentiment analysis techniques are actively studied for these days, evaluation sets are not developed yet for Korean sentiment analysis. In this paper, we constructed an evaluation set for Korean sentiment analysis. To evaluate sentiment analysis systems more throughly, each sentence in our evaluation set is tagged with the polarity of the sentiment as well as the category and the strength of the sentiment. We divide kinds of sentiment into 7 positive categories and 15 negative categories. Each category is given the strength of the sentiment from 1 to 3. Our evaluation set consists of 3,270 sentences extracted from various social media. For each sentence, 5 human taggers assigned the category and the strength of the sentiment expressed in the sentence. The ratio of inter-taggers agreement was 93% in the polarity, 70% in the category, 58% in the strength of sentiment. The ratio of inter-taggers agreement our evaluation set is a bit higher than other evaluation sets developed for German and Spanish. This result shows our evaluation set can be used as a reliable resource for the evaluation of sentiment analysis systems.

• The Journal of Society for e-Business Studies
• /
• v.21 no.3
• /
• pp.15-28
• /
• 2016

This paper proposes a system recommending spatial information what user wants with collecting and analyzing tweets around the user's location by using the GPS information acquired in mobile. This system has built an emotion dictionary and then derive the recommendation score of morphological analyzed tweets to provide not just simple information but recommendation through the emotion analysis information. The system also calculates distance between the recommended tweets and user's latitude-longitude coordinates and the results showed the close order. This paper evaluates the result of the emotion analysis in a total of 10 areas with two keyword 'Restaurants' and 'Performance.' In the result, the number of tweets containing the words positive or negative are 122 of the total 210. In addition, 65 tweets classified as positive or negative by analyzing emotions after a morphological analysis and only 46 tweets contained the meaning of the positive or negative actually. This result shows the system detected tweets containing the emotional element with recall of 38% and performed emotion analysis with precision of 71%.

• Annual Conference on Human and Language Technology
• /
• 2010.10a
• /
• pp.42-45
• /
• 2010

본 연구에서는 한국의 대중가요의 가사 정보를 형태소 단위로 분석하고 이 정보를 기반으로 노래의 감정을 분류하여 추천하는 시스템을 제안한다. 이 시스템을 구축하기 위해서 수집된 노래의 가사는 형태소를 분석하여 각 형태소를 자질로 결정하고, 사용되는 분류기는 ME 모델을 이용해서 학습된다. 이 학습된 분류기는 자질의 수에 따라 그 성능이 분석되고, 분류기를 사용한 추천 시스템은 랜덤하게 생성된 데이터 집합에 대해서 얼마나 정확하게 노래를 추천하는 지를 분석한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.04a
• /
• pp.515-518
• /
• 2009

이동단말기가 보급이 확산됨에 따라 많은 사용자들이 이동단말기를 사용하고 필연적으로 많은 통신행동을 하고 있다. 특히 SMS 는 시간과 장소의 제한이 적어 사용자들의 통신행동 중 큰 비중을 차지하고 있다. SMS 통신행동에서 이모티콘의 사용이 많이 나타나고 있으며 이는 텍스트 기반의 의사소통의 한계를 극복하기 위한 방안으로 볼 수 있다. SMS 로부터 사용자의 감정을 추론하려는 기존의 연구가 있었지만 SMS 텍스트에 국한된다는 문제점이 있다. 본 논문에서는 최근 휴대폰, PDA, 스마트폰 등 이동단말기의 발전에 따라 통신행동 기록, 위치 정보와 같은 컨텍스트 정보를 수집하고 이용할 수 있음에 착안하여 SMS 텍스트와 함께 이동단말기의 컨텍스트 정보를 추론에 사용하였다. 의사결정트리를 이용하여 가용한 컨텍스트 정보로부터 추론한 정황 정보를 SMS 통신에서 사용하여 기존의 텍스트 기반의 의사소통의 한계를 극복할 수 있는 Visual SMS 를 제안한다. 사전에 정의한 훈련 데이터 집합을 통하여 의사결정트리를 생성하고 이를 기반으로 Visual SMS 를 구현, 시뮬레이션하여 추론 결과를 통해 그 기대효과를 확인한다.

• Korean journal of communication and information
• /
• v.84
• /
• pp.28-98
• /
• 2017

This work critically deals with the much unstable and fragmented states of young adults in contemporary Korean society. Nowadays, most of young adults in Korea have encountered much severe, uncertain, and 'hellish' socio-economic situations while their dreams of getting regular jobs are hard to achieve. This study aims to present a nuanced and detailed analysis of their sedimented feelings, variegated emotions, as well as embodied positions by utilizing in-depth interviews and critical explorations on their particular modes of survival.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.8
• /
• pp.47-54
• /
• 2020

Developmental disorders are impairments of brain and/or central nervous system and refer to a disorder of brain function that affects languages, communication skills, perception, sociality and so on. In diagnosis of developmental disorders, behavioral response such as expressing emotions in proper situation is one of observable indicators that tells whether or not individual has the disorders. However, diagnosis by observation can allow subjective evaluation that leads erroneous conclusion. This research presents the technological environment and data acquisition system for AI based screening of autism disorder. The environment was built considering activities for two screening protocols, namely Autism Diagnostic Observation Schedule (ADOS) and Behavior Development Screening for Toddler (BeDevel). The activities between therapist and baby during the screening are fully recorded. The proposed software in this research was designed to support recording, monitoring and data tagging for learning AI algorithms.

• Journal of Software Assessment and Valuation
• /
• v.16 no.1
• /
• pp.81-88
• /
• 2020

Various methods of differentiating and servicing digital content for individual users have been studied. Searching for association rules is a very useful way to discover individual preferences in digital content services. The Apriori algorithm is useful as an association rule extractor using frequent itemsets. However, the Apriori algorithm is not suitable for application to an actual content service because it considers only the reference count of each content. In this paper, we propose a new algorithm based on the Apriori that searches association rules by using purchase history and usage-time for each item. The proposed algorithm utilizes the usage time with the weight value according to purchase items. Thus, it is possible to extract the exact preference of the actual user. We implement the proposed algorithm and verify the performance through the actual data presented in the actual content service system.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.11
• /
• pp.206-216
• /
• 2013

The emotion plays a critical role in human's daily life including learning, action, decision and communication. In this paper, emotion discrimination classifier is designed to reduce system complexity through reduced selection of dominant features from biosignals. The photoplethysmography(PPG), skin temperature, skin conductance, fontal and parietal electroencephalography(EEG) signals were measured during 4 types of movie watching associated with the induction of neutral, sad, fear joy emotions. The genetic algorithm with support vector machine(SVM) based fitness function was designed to determine dominant features among 24 parameters extracted from measured biosignals. It shows maximum classification accuracy of 96.4%, which is 17% higher than that of SVM alone. The minimum error features selected are the mean and NN50 of heart rate variability from PPG signal, the mean of PPG induced pulse transit time, the mean of skin resistance, and ${\delta}$ and ${\beta}$ frequency band powers of parietal EEG. The combination of parietal EEG, PPG, and skin resistance is recommendable in high accuracy instrumentation, while the combinational use of PPG and skin conductance(79% accuracy) is affordable in simplified instrumentation.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.23 no.6
• /
• pp.542-553
• /
• 2020

Virtual world and simulation need large scale map rendering. However, rendering too many vertices is a computationally complex and time-consuming process. Some game development companies have developed 3D LOD objects for high-speed rendering based on distance between camera and 3D object. Terrain physics simulation researchers need a way to recognize the original object shape from 3D LOD objects. In this paper, we proposed simply automatic LOD framework using point cloud data (PCD). This PCD was created using a 6-direct orthographic ray. Various experiments are performed to validate the effectiveness of the proposed method. We hope the proposed automatic LOD generation framework can play an important role in game development and terrain physic simulation.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2017.01a
• /
• pp.73-74
• /
• 2017

SVM(Support Vector Machine) 알고리즘은 대표적인 기계 학습 분류 알고리즘으로 감정 분석, 제스처 인식 등 다양한 분야의 문제를 해결하기 위해 사용되고 있다. SVM 알고리즘은 분리경계면(Hyper-Plane) 또는 분리경계면 집합 중 지지벡터(Support Vector)라 불리는 특정한 점들로 이루어진 두 그룹 간의 거리 차이(Margin)를 최대로 하는 분리경계면을 이용하여 데이터를 분류하는 알고리즘이다. 높은 정확도를 제공하지만 처리 속도가 느리며 학습을 위해 대량의 데이터 및 메모리가 필요하기 때문에 자원이 제한적인 IoT 환경에서 사용이 어렵다. 본 논문에서는 자원이 제한된 IoT 노드를 기반으로 효율적으로 데이터를 학습하기 위해 K-means 알고리즘을 이용하여 SVM 알고리즘의 저전력화 방안을 연구한다.