• 한국컴퓨터정보학회논문지
• /
• 제21권5호
• /
• pp.19-29
• /
• 2016

This paper suggests a method to refine a massive collective intelligence data, and visualize with multilevel sentiment network, in order to understand information in an intuitive and semantic way. For this study, we first calculated a frequency of sentiment words from each movie review. Second, we designed a Heatmap visualization to effectively discover the main emotions on each online movie review. Third, we formed a Sentiment-Movie Network combining the MDS Map and Social Network in order to fix the movie network topology, while creating a network graph to enable the clustering of similar nodes. Finally, we evaluated our progress to verify if it is actually helpful to improve user cognition for multilevel analysis experience compared to the existing network system, thus concluded that our method provides improved user experience in terms of cognition, being appropriate as an alternative method for semantic understanding.

• 디지털산업정보학회논문지
• /
• 제15권1호
• /
• pp.87-97
• /
• 2019

Reaction of people is importantly considered about specific case as a social network service grows. In the previous research on analysis of social network service, they predicted tendency of interesting topic by giving scores to sentences written by user. Based on previous study we proceeded research of sentiment analysis for social network service's sentences, which predict the result as positive or negative for movie reviews. In this study, we used movie review to get high accuracy. We classify the movie review into positive or negative based on the score for learning. Also, we performed embedding and morpheme analysis on movie review. We could predict learning result as positive or negative with a number 0 and 1 by applying the model based on learning result to social network service. Experimental result show accuracy of about 80% in predicting sentence as positive or negative.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권8호
• /
• pp.3169-3181
• /
• 2015

Nowadays, online word-of-mouth has become a powerful influencer to marketing and sales in business. Opinion mining and sentiment analysis is frequently adopted at market research and business analytics field for analyzing word-of-mouth content. However, there still remain several challengeable areas for 1) sentiment analysis aiming for Korean word-of-mouth content in film market, 2) availability of machine learning models only using linguistic features, 3) effect of the size of the feature set. This study took a sample of 10,000 movie reviews which had posted extremely negative/positive rating in a movie portal site, and conducted sentiment analysis with four machine learning algorithms: naïve Bayesian, decision tree, neural network, and support vector machines. We found neural network and support vector machine produced better accuracy than naïve Bayesian and decision tree on every size of the feature set. Besides, the performance of them was boosting with increasing of the feature set size.

• 지능정보연구
• /
• 제22권3호
• /
• pp.71-89
• /
• 2016

누구나 본인이 사용한 제품이나, 이용한 서비스에 대한 후기를 자유롭게 인터넷에 작성할 수 있고, 이러한 데이터의 양은 점점 더 많아지고 있다. 감성분석은 사용자가 생성한 온라인 텍스트 속에 내포된 감성 및 감정을 식별하기 위해 사용된다. 본 연구는 다양한 데이터 도메인 중 영화 리뷰를 분석 대상으로 한다. 영화 리뷰를 이용한 기존 연구에서는 종종 리뷰 평점을 관객의 감성으로 동일시하여 감성분석에 이용한다. 그러나 리뷰 내용과 평점의 실제적 극성 정도가 항상 일치하는 것은 아니기 때문에 연구의 정확성에 한계가 발생할 수 있다. 이에 본 연구에서는 기계학습 기반의 감성 분류기를 구축하고, 이를 통해 리뷰의 감성점수를 산출하여 리뷰에서 나타나는 감성의 수치화를 목표로 한다. 나아가 산출된 감성점수를 이용하여 리뷰와 영화 흥행 간의 연관성을 살펴보았다. 감성분석 모델은 지지벡터 분류기와 신경망을 이용해 구축되었고, 총 1만 건의 영화 리뷰를 학습용 데이터로 하였다. 감성분석은 총 175편의 영화에 대한 1,258,538개의 리뷰에 적용하였다. 리뷰의 평점과 흥행, 그리고 감성점수와 흥행과의 연관성은 상관분석을 통해 살펴보았고, t-검정으로 두 지표의 평균차를 비교하여 감성점수의 활용성을 검증하였다. 연구 결과, 본 연구에서 제시하는 모델 구축 방법은 나이브 베이즈 분류기로 구축한 모델보다 높은 정확성을 보였다. 상관분석 결과로는, 영화의 주간 평균 평점과 관객 수 간의 유의미한 양의 상관관계가 나타났고, 감성점수와 관객 수 간의 상관분석에서도 유사한 결과가 도출되었다. 이에 두 지표간의 평균을 이용한 t-검정을 수행하고, 이를 바탕으로 산출한 감성점수를 리뷰 평점의 역할을 할 수 있는 지표로써 활용 가능함을 검증하였다. 나아가 검증된 결론을 근거로, 트위터에서 영화를 언급한 트윗을 수집하여 감성분석을 적용한 결과를 살펴봄으로써 감성분석 모델의 활용 방안을 모색하였다. 전체적 실험 및 검증의 과정을 통해 본 연구는 감성분석 연구에 있어 개선된 감성 분류 방법을 제시할 수 있음을 보였고, 이러한 점에서 연구의 의의가 있다.

• 한국지능시스템학회논문지
• /
• 제24권3호
• /
• pp.292-297
• /
• 2014

최근 소셜 네트워크 서비스가 보편화되면서, 이를 활용하여 사람들의 의견이나 감성 등을 파악하기 위한 감성분석 연구가 다양한 분야 진행되고 있다. 기존의 영화 관련 연구의 경우, 대부분이 영화평에 대해 단순 긍/부정으로 감성분석을 하여, 영화에 대한 선호도를 파악하는 데 그쳤다. 사람의 감성은 단순 긍/부정이 아닌 다양한 감성으로 분류될 수 있는데 반해, 이분법적 감성분석은 영화의 평점 정보에서 손쉽게 얻을 수 있는 선호도와 유사한 분석을 하는데 그친다. 따라서 영화의 평점보다 다양하고 유용한 정보를 얻기 위해서는, 영화 리뷰를 세분화된 감성으로 분석하여 영화에 대해 느낀 감성을 다양한 기준으로 분류할 필요가 있다. 본 논문에서는 Thayer 모델을 기반으로 감성 분류 기준을 세우고, 수집한 영화 관련 트윗을 이용하여 각 영화에 대해 대중이 느끼는 감성을 분석한다. 분석된 영화에 대한 감성 비율을 유클리드거리, 코사인유사도, 피어슨 상관계수를 이용하여 영화간의 유사도를 측정하였다. IMDB에서 제공하는 유사 영화 정보를 바탕으로 본 논문에서 제안하는 방식의 유용성을 검증하였다.

• 한국IT서비스학회지
• /
• 제17권1호
• /
• pp.79-89
• /
• 2018

Sentiment analysis is a technique of text mining that extracts feelings of the person who wrote the sentence like movie review. The preliminary researches of sentiment analysis identify sentiments by using the dictionary which contains negative and positive words collected in advance. As researches on deep learning are actively carried out, sentiment analysis using deep learning model with morpheme or word unit has been done. However, this model has disadvantages in that the word dictionary varies according to the domain and the number of morphemes or words gets relatively larger than that of phonemes. Therefore, the size of the dictionary becomes large and the complexity of the model increases accordingly. We construct a sentiment analysis model using recurrent neural network by dividing input data into phoneme-level which is smaller than morpheme-level. To verify the performance, we use 30,000 movie reviews from the Korean biggest portal, Naver. Morpheme-level sentiment analysis model is also implemented and compared. As a result, the phoneme-level sentiment analysis model is superior to that of the morpheme-level, and in particular, the phoneme-level model using LSTM performs better than that of using GRU model. It is expected that Korean text processing based on a phoneme-level model can be applied to various text mining and language models.

• 대한안전경영과학회지
• /
• 제19권4호
• /
• pp.169-182
• /
• 2017

Predicting a box office gross in the film industry is an important goal. Many works have analyzed the elements of a film making. Previous studies have suggested several methods for predicting box office such as a model for distinguishing people's reactions by using a sentiment analysis, a study on the period of influence of word-of-mouth effect through SNS. These works discover that a word of mouth (WOM) effect through SNS influences customers' choice of movies. Therefore, this study analyzes correlations between a box office gross and a ratio of people reaction to a certain movie by extracting their feedback on the film from before and after of the film opening. In this work, people's reactions to the movie are categorized into positive, neutral, and negative opinions by employing sentiment analysis. In order to proceed the research analyses in this work, North American tweets are collected between March 2011 and August 2012. There is no correlation for each analysis that has been conducted in this work, hereby rate of tweets before and after opening of movies does not have relationship between a box office gross.

• International Journal of Computer Science & Network Security
• /
• 제23권8호
• /
• pp.101-106
• /
• 2023

In this busy world actually stress is continuously grow up in research and monitoring social websites. The social interaction is a process by which people act and react in relation with each other like play, fight, dance we can find social interactions. In this we find social structure means maintain the relationships among peoples and group of peoples. Its a limit and depends on its behavior. Because relationships established on expectations of every one involve depending on social network. There is lot of difference between emotional pain and physical pain. When you feel stress on physical body we all feel with tensions, stress on physical consequences, physical effects on our health. When we work on social network websites, developments or any research related information retrieving etc. our brain is going into stress. Actually by social network interactions like watching movies, online shopping, online marketing, online business here we observe sentiment analysis of movie reviews and feedback of customers either positive/negative. In movies there we can observe peoples reaction with each other it depends on actions in film like fights, dances, dialogues, content. Here we can analysis of stress on brain different actions of movie reviews. All these movie review analysis and stress on brain can calculated by machine learning techniques. Actually in target oriented business, the persons who are working in marketing always their brain in stress condition their emotional conditions are different at different times. In this paper how does brain deal with stress management. In software industries when developers are work at home, connected with clients in online work they gone under stress. And their emotional levels and stress levels always changes regarding work communication. In this paper we represent emotional intelligence with stress based analysis using machine learning techniques in social networks. It is ability of the person to be aware on your own emotions or feeling as well as feelings or emotions of the others use this awareness to manage self and your relationships. social interactions is not only about you its about every one can interacting and their expectations too. It about maintaining performance. Performance is sociological understanding how people can interact and a key to know analysis of social interactions. It is always to maintain successful interactions and inline expectations. That is to satisfy the audience. So people careful to control all of these and maintain impression management.

• Journal of Information Processing Systems
• /
• 제15권5호
• /
• pp.1029-1035
• /
• 2019

In recent years, artificial intelligence (AI) services have become one of the most essential parts to extend human capabilities in various fields such as face recognition for security, weather prediction, and so on. Various learning algorithms for existing AI services are utilized, such as classification, regression, and deep learning, to increase accuracy and efficiency for humans. Nonetheless, these services face many challenges such as fake news spread on social media, stock selection, and volatility delay in stock prediction systems and inaccurate movie-based recommendation systems. In this paper, various algorithms are presented to mitigate these issues in different systems and services. Convolutional neural network algorithms are used for detecting fake news in Korean language with a Word-Embedded model. It is based on k-clique and data mining and increased accuracy in personalized recommendation-based services stock selection and volatility delay in stock prediction. Other algorithms like multi-level fusion processing address problems of lack of real-time database.

• 한국산업정보학회논문지
• /
• 제28권5호
• /
• pp.15-30
• /
• 2023

근래에 트랜스포머(Transformer) 구조를 기초로 하는 ChatGPT와 같은 생성모델이 크게 주목받고 있다. 트랜스포머는 다양한 신경망 모델에 응용되는데, 구글의 BERT(bidirectional encoder representations from Transformers) 문장생성 모델에도 사용된다. 본 논문에서는, 한글로 작성된 영화 리뷰에 대한 댓글이 긍정적인지 부정적인지를 판단하는 텍스트 이진 분류모델을 생성하기 위해서, 사전 학습되어 공개된 BERT 다국어 문장생성 모델을 미세조정(fine tuning)한 후, 새로운 한국어 학습 데이터셋을 사용하여 전이학습(transfer learning) 시키는 방법을 제안한다. 이를 위해서 104 개 언어, 12개 레이어, 768개 hidden과 12개의 집중(attention) 헤드 수, 110M 개의 파라미터를 사용하여 사전 학습된 BERT-Base 다국어 문장생성 모델을 사용했다. 영화 댓글을 긍정 또는 부정 분류하는 모델로 변경하기 위해, 사전 학습된 BERT-Base 모델의 입력 레이어와 출력 레이어를 미세 조정한 결과, 178M개의 파라미터를 가지는 새로운 모델이 생성되었다. 미세 조정된 모델에 입력되는 단어의 최대 개수 128, batch_size 16, 학습 횟수 5회로 설정하고, 10,000건의 학습 데이터셋과 5,000건의 테스트 데이터셋을 사용하여 전이 학습시킨 결과, 정확도 0.9582, 손실 0.1177, F1 점수 0.81인 문장 감정 이진 분류모델이 생성되었다. 데이터셋을 5배 늘려서 전이 학습시킨 결과, 정확도 0.9562, 손실 0.1202, F1 점수 0.86인 모델을 얻었다.