PURPOSE Recently, the need to improve service quality to provide an ideal viewing experience for sport OTT service users has increased. This study aimed to analyze user reviews of the SPOTV Now app, an OTT service specializing in sports in South Korea, to identify major perceptions of viewers and suggest measures for improvement. METHODS Review data (8,512) recorded on the SPOTV Now app were collected from November 21, 2017 to November 30, 2023. LDA-based topic modeling analysis was conducted along with a series of data preprocessing processes. Sentiment analysis was carried out through machine learning logistic regression analysis and odds ratio scores were calculated. RESULTS A total of five topics were derived, including ‘member management service', 'service stability', 'sport streaming service', 'sport video service', and 'subscription service'. Sentiment analysis indicated that 'exclusive', 'server', 'payment', 'log-in', 'authentication', 'pay', 'subscription', 'advertisement', 'play', and 'loading' were the top ten keywords. CONCLUSIONS Sport OTT operators should be more responsive to user requests and feedback, and video playback and server error problems should be improved for service stability. It is necessary to secure diversity as well as exclusivity of content, in addition to minimizing advertising or preparing various rate types, so that app users feel that they receive high value for their money.
PURPOSE This study was conducted to estimate the tendency of psychological factors influencing cycling performance by analyzing the characteristic factors of athlete reputation in the news big-data. METHODS To explore the psychological factors influencing cycling performance, an open questionnaire was conducted on 82 cyclists, and Inductive Content Analyses was performed. Overall, 89,520 news articles were collected through BIGKinds, and forming factors of athlete reputation were derived through LDA topic modeling analysis and inductive categorization. Through regression analysis, time series tendency of the factors of athlete reputation was calculated. Finally, the tendency of psychological factors to influence cycling performance was estimated based on the previously derived results in this study. RESULTS The psychological factors influencing cycling performance were found to be; emotion control, trust capital, cognitive control, motivation and communications with the coach. The forming factors of athlete reputation was found to be; reporting of the sports event, infrastructure creation, analysis to performance, moral issue, social environmental changes and sports gossip. The time series tendency of the forming factors of athlete reputation was found to include the categories of Hot, Warm, Cool and Cold. The psychological factors influencing cycling performance are estimated to expand to exercise performance and moral intelligence. CONCLUSIONS The results of this study suggest that the discussion of psychological factors influencing cycling performance extends not only to exercise performance, but also to moral intelligence, reflecting the socio-cultural context in the discussion of performance.
PURPOSE The purpose of this study is to analyze the type of and interest in home training video contents using the YouTube platform. METHODS Web crawling was performed using Python and a total of 3,937 sets ofvideo information (title, content, number of views, upload date) were obtained, 3,155 of which were finally selected for the study material. Overlapping and unrelated content were excluded. The data of text underwent 3 stages of preprocessing, the TF and TF-IDF of the keywords were calculated to identify the main keywords, and the LDA algorithm was applied in the topic modeling to successfully identify the types. In order to understand the level of interest by type, the number of views was subdivided into the percentage of the assigned type. RESULTS First, the types of home training videos were classified into bare whole body exercise for aerobic and muscular power strengthening, Pilates exercise for core and upper body strengthening, upper body exercise using tools, lower body line exercise, posture correction and upper body stretching exercise for pain relief, hip-up exercise, dance and tabata exercise for diet, diet and lower body correction stretching exercise for diet, and bare body exercise for core and lower body strengthening. Second, it was found that the proportion and interest were high in the contents of bare whole body exercise for aerobic and muscular power strengthening, dance and tabata exercise for diet, diet and lower body correction stretching exercise for diet. CONCLUSIONS The findings of this study may provide baseline data about the development of the active online home training videos in the market.
PURPOSE By analyzing trends in Taekwondo demonstrations, specifically in breaking and performances, to date, this study aims to offer timely insights and set the groundwork for future research. METHODS We used Korean abstracts from a total of 425 papers containing the keyword “Taekwondo demonstrations” spanning 20 years from April 2004 to April 2023. We employed Python 3.5.2 to conduct dynamic topic modeling (Latent Dirichlet Analysis, LDA) and to examine the correlation between the topic distribution by section and the publication year. RESULTS The main findings from the LDA are as follows. Topic 1 (10%): “The development of demonstrations: performance in culture and art, ” Topic 2 (11%): “The development of formalized rules and judgments in a demonstration event,” Topic 3 (08%): “A study on the educational courses and professionalism of Taekwondo coaches,” Topic 4 (11%): “Technical movements and kinematic characteristics,” Topic 5 (09%): “A study on marketing perspectives of demonstration performances,” and Topic 7 (33%): “Global exchange: the development and rise of internationalization.” In the correlation analysis between the topic share by section and the publication year, Topics 1 to 5 exhibited no statistically significant correlation. However, Topic 6, “A study on the attainment of events, training, and the psychological factors influencing athletes” and Topic 7, “Global exchange: the development and rise of internationalization,” also displayed a very statistically significant but negative correlation. CONCLUSIONS Future research should focus on studies on the psychological management of athletes during the performance of specific techniques and training methods. Further research considering the global characteristics of Taekwondo may be required.
Purpose The purpose of this study is to introduce the basic concepts and procedures for topic modeling and to explain topic modeling to news articles about dementia-related physical activities. And it is also to discuss the possibility of using topic Modeling in the field of physical education. Methods In this study, the LDA algorithm of topic modeling is explained and the analysis procedure is summarized step by step by text preprocessing, text formatting, and topic number determination. The application cases were selected from 274 news articles about dementia-related physical activities reported in 13 major daily newspapers from 2000 to 2018. Results When the number of topics is 3, the Coherence Score figure is the highest. Topic 1 is about welfare services for dementia patients, Topic 2 is about prevention of dementia, and Topic 3 is about dementia research. The ratio by each subject is Topic 2 (46.0%), Topic 3 (33.2%) and Topic 1 (20.8%) in order of high ratio. Conclusion Topic modeling is an effective methodology to extract potential information excluding subjectivity of researchers. It is expected to be used when searching for information in massive texts in the field of physical education.
Purpose The purpose of this study was to analyze the research topics of the articles which were published through European Sport Management Quarterly(ESMQ) from 2009 to 2018. The prior topic analysis studies of the ESMQ classified topics based on the key words using NASPE-NASSM SMPS categories. Therefore they couldn't fully reflect the content of the articles. Methods The topic modeling of the current study was conducted with the Latent Dirichlet Allocation(LDA) which generates topics based on the word usage in the article. A total of 265 articles were converted from 'pdf' format to 'txt' ANSI format for topic modeling analysis. The whole topic modeling process was done using R program and the model was set to generate 10 topics from the article. Results The 3 sport management experts were hired to label the name of the topics and the name of the topics are as follow : (1) Impact of mega sport event, (2) Cause-related marketing, (3) Factors affecting the results of the competition, (4) Managing sport organization, (5) European sport leagues, (6) Strategic management, (7) Sport economics, (8) Sport in communities, (9) Sport consumers, (10) Elite sports. It is not quite possible to compare the results of the current topic modeling results with the previous ones because of the methodological differences. However, even though the standards are different, Sport marketing topic showed the largest growth among the 10 topics extracted. Conclusions This study used the LDA probabilistic algorithm to analyze research topics, which made the analyses more objective and wholistic. However, the insights of the researchers were still needed to interpret and labeling the topics.
