• International Journal of Computer Science & Network Security
• /
• 제21권3호
• /
• pp.112-119
• /
• 2021

Social media has become a global means of communication in people's lives. Most people are using Twitter for communication purposes and its inappropriate use, which has negative effects on people's lives. One of the widely common misuses of Twitter is cyberbullying. As the resources of dialectal Arabic are rare, so for cyberbullying most people are using dialectal Arabic. For this reason, the ultimate goal of this study is to detect and classify cyberbullying on Twitter in the Arabic context in Saudi Arabia. To help in the detection and classification of tweets, Pointwise Mutual Information (PMI) to generate a lexicon, and Support Vector Machine (SVM) algorithms are used. The evaluation is performed on both methods in terms of the F1-score. However, the F1-score after applying the PMI is 50%, while after the SVM application on the resampling data it is 82%. The analysis of the results shows that the SVM algorithm outperforms better.

• Asian Journal for Public Opinion Research
• /
• 제9권3호
• /
• pp.240-265
• /
• 2021

Public services need to be evaluated regularly to identify areas that need further improvement. Data collection via Twitter is affordable and timely, so it has the potential to be utilized to evaluate the quality of public service. This study utilizes tweets mentioning three service units of the provincial government of Jakarta and applies both sentiment analysis and topic classification to predict a rating/score of public service quality. The research goal is to examine if the evaluation of public services based on social media data is possible. The findings indicate that the use of Twitter has an advantage in terms of sample size and variety of opinions. Tweets can be translated into scores as well. Nonetheless, the representativeness issue and the predominance of complaint tweets can affect the reliability of the results.

• 지능정보연구
• /
• 제27권2호
• /
• pp.1-15
• /
• 2021

추천시스템은 사용자의 기호를 파악하여 물품 구매 결정을 도와주는 역할을 할 뿐만 아니라, 비즈니스 전략의 관점에서도 중요한 역할을 하기에 많은 기업과 기관에서 관심을 갖고 있다. 최근에는 다양한 추천시스템 연구 중에서도 NLP와 딥러닝 등을 결합한 하이브리드 추천시스템 연구가 증가하고 있다. NLP를 이용한 감성분석은 사용자 리뷰 데이터가 증가함에 따라 2000년대 중반부터 활용되기 시작하였지만, 기계학습 기반 텍스트 분류를 통해서는 텍스트의 특성을 완전히 고려하기 어렵기 때문에 리뷰의 정보를 식별하기 어려운 단점을 갖고 있다. 본 연구에서는 기계학습의 단점을 보완하기 위하여 BERT 기반 감성분석을 활용한 추천시스템을 제안하고자 한다. 비교 모형은 Naive-CF(collaborative filtering), SVD(singular value decomposition)-CF, MF(matrix factorization)-CF, BPR-MF(Bayesian personalized ranking matrix factorization)-CF, LSTM, CNN-LSTM, GRU(Gated Recurrent Units)를 기반으로 하는 추천 모형이며, 실제 데이터에 대한 분석 결과, BERT를 기반으로 하는 추천시스템의 성과가 가장 우수한 것으로 나타났다.

• Journal of Information Processing Systems
• /
• 제16권2호
• /
• pp.329-342
• /
• 2020

In recent years, emotional text classification is one of the essential research contents in the field of natural language processing. It has been widely used in the sentiment analysis of commodities like hotels, and other commentary corpus. This paper proposes an improved W-LDA (weighted latent Dirichlet allocation) topic model to improve the shortcomings of traditional LDA topic models. In the process of the topic of word sampling and its word distribution expectation calculation of the Gibbs of the W-LDA topic model. An average weighted value is adopted to avoid topic-related words from being submerged by high-frequency words, to improve the distinction of the topic. It further integrates the highest classification of the algorithm of support vector machine based on the extracted high-quality document-topic distribution and topic-word vectors. Finally, an efficient integration method is constructed for the analysis and extraction of emotional words, topic distribution calculations, and sentiment classification. Through tests on real teaching evaluation data and test set of public comment set, the results show that the method proposed in the paper has distinct advantages compared with other two typical algorithms in terms of subject differentiation, classification precision, and F1-measure.

• 한국멀티미디어학회논문지
• /
• 제25권10호
• /
• pp.1499-1511
• /
• 2022

In the context of globalization, circulation between cities has become more frequent. The airport is no longer just a place for boarding, disembarking, and transportation, but a public place that serves as the communication function of the "aviation city". The intervention of public art in the airport space not only gives users a sense of space experience, but also becomes a unique carrier for city and country image shaping. The purpose of this paper is to study the emotional value brought by airport public art to users, and to investigate the correlation analysis of public art design elements and user preferences based on this premise. The research methods are machine learning method and SPSS 21.0. The user's emotional value is introduced in the big data evaluation, and the preference and inclination of airport users to various elements of public art are analyzed by questionnaire. Through the research conclusion, the preference and main contradiction of users in the airport for the four dimensions of public art design elements are obtained. Opinions and optimization methods to provide reference data and theoretical support for public art design.

• 공학교육연구
• /
• 제27권1호
• /
• pp.41-49
• /
• 2024

The purpose of this study is to explore ways to improve major satisfaction that can be applied by universities through the analysis of factors influencing major satisfaction of engineering college students. To this end, Korea-National Survey of Student Engagement(K-NSSE) data involving 814 students from T University were used, and logistic regression analysis and t-test were applied. The main results obtained through this are as follows. First, engineering college students' major satisfaction factors include major-career relevance, college immersion, and positive academic sentiment. Second, depending on the grade, it was confirmed that the factor of major-career relevance in the lower grades, and the factors of meaningful learning experience and college immersion in addition to major-career relevance in the upper grades had a significant influence. Third, the higher the meaningful learning experience, positive academic sentiment, and college immersion, including the major-career relevance, the higher the major satisfaction was found in the middle-class group with a score of BO or higher. This study is meaningful in that it revealed differences in influence by individual characteristics as well as major satisfaction influencing factors that can be practiced in universities such as learning experiences.

• 한국IT서비스학회지
• /
• 제18권2호
• /
• pp.37-54
• /
• 2019

The purpose of this study is to present an evaluation model to measure the clarification level of stakeholder requirements of public sector software projects in the Republic of Korea. We tried to grasp the quality of proposal request through evaluation model. It also examines the impact of the level of stakeholder requirements on the level of system requirements. To do this, we analyzed existing research models and related standards related to business requirements and stakeholder requirements, and constructed evaluation models for the system operation concept documents in the ISO/IEC/IEEE 29148. The system operation concept document is a document prepared by organizing the requirements of stakeholders in the organization and sharing the intention of the organization. The evaluation model proposed in this study focuses on evaluating whether the contents related to the system operation concept are faithfully written in the request for proposal. The evaluation items consisted of three items: 'organization status', 'desired changes', and 'operational constraints'. The sample extracted 217 RFPs in the national procurement system. As a result of the analysis, the evaluation model proved to be valid and the internal consistency was maintained. The level of system operation concept was very low, and it was also found to affect the quality of system requirements. It is more important to clearly write stakeholders' requirements than the functional requirements. we propose a news classification methods for sentiment analysis that is effective for bankruptcy prediction model.

• 한국의류산업학회지
• /
• 제6권5호
• /
• pp.605-612
• /
• 2004

This is a study which evaluates the brand personality on casual brands according to the sensibilities of consumers. Focus is placed on classifying the sensibilities of consumers through G sensibilities developed by the Fuji Research Institute in Japan, and then on investigating brand personality of casual wear, Levi's in particular. The subjects are 187 male and female college students living in the Busan area. Data were, using SPSS 10. 0 for Window, statistically analyzed by frequency and factor analysis for VARIMAX, Cronbach's coefficient, and ANOVA. The results of data analysis are as follow; First, in the distribution of G sensibility type, the majority of the respondents have G3(whimsical) type, and then followed by G1(my pace) type, G4(active) type, G2 type(active) type and then G5(sensualist) type. Second, as a result of brand personality evaluation on the casual brands, it has been represented by the following 5 factors; passion, competence, honesty, sophistication, and sentiment. Third, the brand personalities according to G sensibilities showed significant differences among respondents; G5 type(sensualist) especially showed the highest brand personality in the passion and sentiment factors, which are distinctive in the brand personality of Levi's. Consumers of G1 type, G2 type, G3 type, and G4 type showed high brand personality in the competence and honesty factors. The evaluation of brand personality, case study for products development and application, and application of the results need to be continued for follow-up study.

• 한국인터넷방송통신학회논문지
• /
• 제20권5호
• /
• pp.175-185
• /
• 2020

모바일 쇼핑은 오늘날 뉴노멀이라고 할 정도로 일반적인 쇼핑 방법의 하나로 자리잡았다. 온라인 환경에서의 쇼핑을 연구하던 초기와 달리 현재에 보다 중요한 요인들은 상당히 달라졌을 것으로 보인다. 이에 본 연구는 모바일 쇼핑에서 중요하게 여겨지는 서비스품질 요인들을 선별하고 이것이 소비자의 이성-감성적 평가에 어떠한 영향을 주는지 살피는 한편, 향후 구매의도와 구매의도 등으로 이어지는 일련의 영향 관계를 파악해 보고자 하였고 유의미한 결과를 얻을 수 있었다. 연구 결과는 서비스품질 중 개인화와 반응성만이 소비자 심리에 정(+)의 영향을 미치는 것으로 나타났으며, 만족과 기대감정은 소비자 행동에 정(+)의 영향을 미치는 것으로 나타났다. 이는 소비자들에게 개인화와 반응성은 중요한 요인임을 증명하는 결과를 얻을 수 있었으며, 소비자의 만족이 높으면 소비자 행동도 긍정적일 것으로 본 연구결과 나타났다.

• 한국전자거래학회지
• /
• 제26권4호
• /
• pp.15-36
• /
• 2021

국제 항공 시장이 꾸준히 성장함에 따라 항공업계의 경쟁이 더욱 심화되고 있다. 경쟁 우위의 원천을 얻기 위해 서비스의 품질 평가는 필수적이며, 이에 다양한 연구에서 고객 리뷰를 바탕으로 서비스 품질을 측정하는 시도를 지속해 왔다. 그러나 고객 리뷰 데이터를 기반으로 기대와 지각 수준의 차이를 파악하고 전략적 방향을 제시하는 연구는 미흡한 실정이다. 본 연구에서는 항공사 서비스를 대상으로 차원별 중요도를 머신러닝을 통해 측정하고, 차원별 지각 수준을 감성분석을 통해 분석한다. 차원별 중요도와 지각 수준의 결과를 활용하여 항공사별 서비스의 성과를 측정하기 위한 전략 매트릭스를 제시하고, 이를 통해 각 항공사의 품질 분석을 수행한다. 본 연구는 항공사의 고객만족을 결정하는 중요한 요인을 파악하는 동시에, 각 항공사의 현재 서비스 수준을 파악하는 틀을 제시함으로써 서비스 품질 평가의 중요한 도구로 활용될 수 있다.