• 한국의류산업학회지
• /
• 제24권5호
• /
• pp.593-604
• /
• 2022

The purpose of this study is to analyze the flow of change in perception of metaverse before and after COVID-19 through big data analysis. This research method used Textom to collect all data, including metaverse for two years before COVID-19 (2018.1.1~2019.11.30) and after COVID-19 outbreak (2020.1.11~2021.12.31), and the collection channels were selected by Naver and Google. The collected data were text mining, and word frequency, TF-IDF, word cloud, network analysis, and emotional analysis were conducted. As a result of the analysis, first, hotels, weddings, and glades were commonly extracted as social issues related to metaverse before and after COVID-19, and keywords such as robots and launches were derived, so the frequency of keywords related to hotels and weddings was high. Second, the association of the pre-COVID-19 metaverse keywords was platform-oriented, content-oriented, economic-oriented, and online promotion-oriented, and post-COVID-19 clusters were event-oriented, ontact sales-oriented, stock-oriented, and new businesses. Third, positive keywords such as likes, interest, and joy before COVID-19 were high, and positive keywords such as likes, joy, and interest after COVID-19. In conclusion, through this study, it was found that metaverse has firmly established itself as a new platform business model that can be used in various fields such as tourism, travel, festivals, and education using smart technology and metaverse.

• 한국의류산업학회지
• /
• 제22권5호
• /
• pp.639-649
• /
• 2020

This study investigates consumer opinions of clothing care and provides fundamental data to decision-making for oncoming development of clothing care system. Textom, a web-matrix program, was used to analyze big data collected from Naver and Daum with a keyword of "clothing care" from March 2019 to February 2020. A total of 22, 187 texts were shown from the big data collection. Collected big data were analyzed using text-mining, network, and CONCOR analysis. The results of this study were as follows. First, many keywords related to clothing care were shown from the result of frequency analysis such as style, Dryer, LG Electronics, Product, Customer, Clothing, and Styler. Consumers were well recognizing and having an interest in recent information related to the clothing care system. Second, various keywords such as product, function, brand, and performance, were linked to each other which were fundamentally related to the clothing care. The interest in products of the clothing care system were linked to product brands that were also naturally linked to consumer interest. Third, the keywords in the network showed similar attributes from the result of CONCOR analysis that were classified into 4 groups such as the characteristics of purchase, product, performance, and interest. Lastly, positive emotions including goodwill, interest, and joy on the clothing care system were strongly expressed from the result of the sentimental analysis.

• 한국의류학회지
• /
• 제44권5호
• /
• pp.1019-1035
• /
• 2020

This study examines the millennial generation, who express themselves and share information on social media after experiencing constantly changing 'hot places' (places of interest) in contemporary cities, with the goal of analyzing space consumption behaviors. Data were collected via an Instagram crawler application developed with Python 3.4 administered to 19,262 posts using the term 'hot places' from November 1 and December 15, 2019. Issues were derived from a text mining technique using Textom 2.0; in addition, semantic network analysis using Ucinet6 and the NetDraw program were also conducted. The results are as follows. First, a frequency analysis of keywords for hot places indicated words frequently found in nouns were related to food, local names, SNS and timing. Words related to positive emotions felt in experience, and words related to behavior in hot places appeared in predicate. Based on importance, communication is the most important keyword and influenced all issues. Second, the results of visualization of semantic network analysis revealed four categories in the scope of the definition of "hot place": (1) culinary exploration, (2) atmosphere of cafés, (3) happy daily life of 'me' expressed in images, (4) emotional photos.

• 아태비즈니스연구
• /
• 제12권2호
• /
• pp.161-174
• /
• 2021

Purpose - The purpose of this study was to examine the recent popular consumption trend, the hocance phenomenon, using social media big data. The study intended to present practical directions and marketing measures for the recovery and growth of the hotel industry after COVID-19 pandemic. Design/methodology/approach - Big data analysis has been used in various fields, and in this study, it was used to understand the hocance phenomenon. For three years from January 1, 2018 to December 31, 2020, we collected text data including the keyword 'hocance' from the blog and cafe of NAVER and Daum. TEXTOM and UCINET 6 were used to collect and analyze the data. Findings - According to the results of analysis, the words such as 'hocance', 'hotel', 'Seoul', 'travel', 'swimming pool', 'Incheon', 'breakfast', 'child' and 'friend' were identified with high frequency. The results of CONCOR analysis showed similar results in all three years. It has been confirmed that 'swimming pool', 'breakfast', 'child' and 'friend' are important when deciding on the hocance package. Research implications or Originality - The study was differentiated in that it used social media big data instead of traditional research methods. Furthermore, it reflected social phenomena as a consumption trend so there was practical value in establishing marketing strategies for the tourism and hotel industry.

• 인지과학
• /
• 제21권4호
• /
• pp.481-503
• /
• 2010

이 연구는 초등학교 3학년 아동들이 나이에 알맞은 동화와 설명글을 읽을 때 보이는 안구 운동의 전반 특징과 낱말/어절 수준에 따른 특징을 우리글에서 처음으로 알아보았다. 아동들은 동화를 읽을 때 어절을 대략 213ms 동안 보다가 글 방향으로 3.6자 눈을 움직였고, 설명문을 읽을 때는 214ms 동안 보다가 글 방향으로 3.3자 정도 눈을 움직였다. 동화와 설명글 모두에서 앞으로 가서 다시 읽는 눈 움직임은 전체의 약 31%정도를 차지했다. 이런 전반 양상과 더불어, 어절 수준에서 어절 길이, 낱말빈도, 착지점 효과를 살펴보았는데, 아이들은 동화와 설명 글 모두에서 긴 어절을 짧은 어절보다 오래 주시하였고, 성인 독자와 마찬가지로 눈을 어절 가운데 근처에 착지하려는 경향성을 보였다. 또한 반복되지 않은 어절들을 분석했을 때 설명글에서 빈도가 낮은 어절들은 길이에 따른 주시시간이 차이를 크게 보였고 빈도가 높은 어절들은 길이에 따른 차이를 거의 보이지 않았다.

• 지능정보연구
• /
• 제24권2호
• /
• pp.59-83
• /
• 2018

고객과 대중의 니즈를 파악하기 위한 감성분석의 중요성이 커지면서 최근 영어 텍스트를 대상으로 다양한 딥러닝 모델들이 소개되고 있다. 본 연구는 영어와 한국어의 언어적인 차이에 주목하여 딥러닝 모델을 한국어 상품평 텍스트의 감성분석에 적용할 때 부딪히게 되는 기본적인 이슈들에 대하여 실증적으로 살펴본다. 즉, 딥러닝 모델의 입력으로 사용되는 단어 벡터(word vector)를 형태소 수준에서 도출하고, 여러 형태소 벡터(morpheme vector) 도출 대안에 따라 감성분석의 정확도가 어떻게 달라지는지를 비정태적(non-static) CNN(Convolutional Neural Network) 모델을 사용하여 검증한다. 형태소 벡터 도출 대안은 CBOW(Continuous Bag-Of-Words)를 기본적으로 적용하고, 입력 데이터의 종류, 문장 분리와 맞춤법 및 띄어쓰기 교정, 품사 선택, 품사 태그 부착, 고려 형태소의 최소 빈도수 등과 같은 기준에 따라 달라진다. 형태소 벡터 도출 시, 문법 준수도가 낮더라도 감성분석 대상과 같은 도메인의 텍스트를 사용하고, 문장 분리 외에 맞춤법 및 띄어쓰기 전처리를 하며, 분석불능 범주를 포함한 모든 품사를 고려할 때 감성분석의 분류 정확도가 향상되는 결과를 얻었다. 동음이의어 비율이 높은 한국어 특성 때문에 고려한 품사 태그 부착 방안과 포함할 형태소에 대한 최소 빈도수 기준은 뚜렷한 영향이 없는 것으로 나타났다.

• 디지털융복합연구
• /
• 제19권10호
• /
• pp.469-489
• /
• 2021

본 연구는 정서 지원 애완용 로봇에 대한 소비자의 담론을 살펴보고 키워드를 통해 해당 제품에 대한 노년 소비자의 반응을 파악하고자 아마존 사이트 내 하즈브로(Hasbro)의 Companion Pets 제품 사용에 대한 미국 소비자들의 리뷰를 수집하고, R을 이용하여 단어 빈도분석, 토픽모델링 LDA 분석을 실시하였다. 첫째, 키워드 빈도분석 결과 애완용 로봇의 형태가 실제 동물과 유사한지에 대한 관심이 높은 것으로 나타났다. 둘째, 토픽모델링 결과 5개의 토픽으로 인지, 감정, 행동적 반응이 도출되었으며 이는 긍정 및 부정으로 크게 분류되었다. 셋째, 소비자와 애완용 로봇의 상호작용에 영향을 미치는 사용자, 제품 및 환경적 특성이 확인되었다. 애완용 로봇은 반려동물을 키우기 어려운 사람들이 이를 대체하기 위하여 사용하고, 인지적 어려움이 있는 노년 소비자와 신체적 어려움이 있는 소비자가 이를 이용하는 것으로 나타났다. 본 연구는 코로나19와 같은 팬데믹 상황에서 정서 지원 기능을 수행하는 애완용 로봇에 대하여 이해하고, 소비자의 효용을 극대화하는 서비스를 제공하는데 도움을 줄 것으로 기대한다.

• 한국문헌정보학회지
• /
• 제56권4호
• /
• pp.431-453
• /
• 2022

혐오는 타인에 대한 배타성이 집단적으로 표출된 것으로, 잘못된 대중적 인식을 통하여 양산되고 재생산된다. 이 연구는 우리사회에서 언급되고 있는 '혐오' 양상을 거시적으로 탐색하고자 1990년부터 2020년까지 발행된 뉴스데이터 17,867건을 대상으로 텍스트마이닝 기법을 활용하여 키워드 네트워크와 군집 분석을 수행하였다. 그리고 단어를 추출하기 전에 먼저 기사를 문장으로 분리하는 전처리 과정을 거쳐 '혐오', '편견', '차별'이라는 단어를 포함하고 있는 문장 총 52,520개를 추출하여 분석에 활용함으로써 '혐오'라는 단어와 인접한 단어들로 구성된 키워드 네트워크를 구축하였다. 수집한 뉴스데이터의 단어 동시출현빈도 분석 결과, 우리 사회에서 혐오와 관련되어 가장 빈번하게 등장하는 대상은 여성, 인종, 성소수자 등이며, 관련된 이슈는 이들 집단과 관련된 법과 범죄 등이었다. 키워드 네트워크 군집 분석 결과, 성별(41.4%), 소수자(28.7%), 인종·민족(15.1%), 선택적·이해관계적(8.5%), 정치·이념(5.7%), 환경·생존적(0.3%) 혐오 등 총 6개의 혐오 군집들이 발견되었다. 논의에서는 군집 분석 결과 구체적으로 드러나지 않은 혐오의 표적(대상)을 모두 추출하여 분석하였다.

• 음성과학
• /
• 제2권
• /
• pp.25-44
• /
• 1997

When listening the various speech synthesis systems developed and being used in our country, we find that though the quality of these systems has improved, they lack naturalness. Moreover, since the voice color of these systems are limited to only one recorded speech DB, it is necessary to record another speech DB to create different voice colors. 'Voice Color' is an abstract concept that characterizes voice personality. So speech synthesis systems need a voice color control function to create various voices. The aim of this study is to examine several factors of voice color control rules for the text-to-speech system which makes natural and various voice types for the sounding of synthetic speech. In order to find such rules from natural speech, glottal source parameters and frequency characteristics of the vocal tract for several voice colors have been studied. In this paper voice colors were catalogued as: deep, sonorous, thick, soft, harsh, high tone, shrill, and weak. For the voice source model, the LF-model was used and for the frequency characteristics of vocal tract, the formant frequencies, bandwidths, and amplitudes were used. These acoustic parameters were tested through multiple regression analysis to achieve the general relation between these parameters and voice colors.

• The Journal of Asian Finance, Economics and Business
• /
• 제9권10호
• /
• pp.215-225
• /
• 2022

This study investigates employees' online reviews, conducts in-depth text topic mining, effectively summarizes the dimensions of employer brand value, and seeks effective ways to build employer brands from a multi-dimensional perspective. This study employs samples of employer reviews, filter keywords according to word frequency-inverse document frequency, builds a review network containing the same keywords, explore the community and summarize the theme dimensions. Simultaneously, it makes a dynamic comparison and analysis of the employer brand value dimension of different industries and enterprises. The study shows that the community exploration theme can be summarized into 11 dimensions of employer brand value, and the dimensions of employer brand value are significantly different across industries and among different enterprises within the industry. The attention to the employer brand value dimension has a significant time change. Various industries pay increasing attention to the dimension of work intensity and career development, while employers pay steady attention to the dimension of welfare benefits. The findings of this study suggest that seeking the heterogeneity of employer brand resources from the multi-dimensional differences and changes is an effective way to improve the competitiveness of enterprises in the human capital market.