TY - GEN
T1 - YouTube Trending Video Analysis in Sri Lanka
AU - Diddeniya, S. I.A.P.
AU - Gunasinghe, H. N.
AU - Premachandra, C.
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - YouTube, owned by Google LLC, is the most widespread video platform. It has gained much popularity in Sri Lanka within the last couple of years. Youth has an interest of becoming full-Time content creators, but lacks technical and marketing knowledge. In this paper, we present a computer aided analysis to provide insights of trending YouTube videos in Sri Lanka. The system gets YouTube trending video data of one and a half months in year 2021 as input and analyze them to report information on video title, description and thumbnail which are the main parameters of YouTube algorithm when deciding on a trending video. Particularly, the system uses YOLO and RetinaNet for detecting objects in thumbnails. Moreover, Sri Lankan video content mainly include Sinhala language which is the mostly used language in the country. We used Sinhala language processing techniques for reporting various information. In summary, using the provided information, a YouTuber can decide on the: (1) colors and objects to include in the thumbnail, (2) title and the description by considering word count and symbols, (3) number of tags, words and phrases to include in a tag and (4) date and time to publish the video. The results indicate that the proposed approach can significantly support an accurate decision of YouTube video upload with a little computational effort.
AB - YouTube, owned by Google LLC, is the most widespread video platform. It has gained much popularity in Sri Lanka within the last couple of years. Youth has an interest of becoming full-Time content creators, but lacks technical and marketing knowledge. In this paper, we present a computer aided analysis to provide insights of trending YouTube videos in Sri Lanka. The system gets YouTube trending video data of one and a half months in year 2021 as input and analyze them to report information on video title, description and thumbnail which are the main parameters of YouTube algorithm when deciding on a trending video. Particularly, the system uses YOLO and RetinaNet for detecting objects in thumbnails. Moreover, Sri Lankan video content mainly include Sinhala language which is the mostly used language in the country. We used Sinhala language processing techniques for reporting various information. In summary, using the provided information, a YouTuber can decide on the: (1) colors and objects to include in the thumbnail, (2) title and the description by considering word count and symbols, (3) number of tags, words and phrases to include in a tag and (4) date and time to publish the video. The results indicate that the proposed approach can significantly support an accurate decision of YouTube video upload with a little computational effort.
KW - Data mining
KW - Data visualisation
KW - Text analysis
KW - Video analysis
KW - YouTube analysis
UR - http://www.scopus.com/inward/record.url?scp=85134006589&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85134006589&partnerID=8YFLogxK
U2 - 10.1109/ICIPRob54042.2022.9798745
DO - 10.1109/ICIPRob54042.2022.9798745
M3 - Conference contribution
AN - SCOPUS:85134006589
T3 - 2022 2nd International Conference on Image Processing and Robotics, ICIPRob 2022
BT - 2022 2nd International Conference on Image Processing and Robotics, ICIPRob 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2nd International Conference on Image Processing and Robotics, ICIPRob 2022
Y2 - 12 March 2022 through 13 March 2022
ER -