Classification of TV Programs According to Their Topics Using the XLNet Model

Authors

  • Quang-Long Vo Ho Chi Minh City University of Technology and Education, Vietnam
  • Ngoc-Hung-Anh Nguyen Posts and Telecommunications Institute of Technology, Vietnam
  • Minh-Son Tran University of Technology HUTECH, Vietnam
  • Thu-Hà Tran Ho Chi Minh City University of Technology and Education, Vietnam

Corressponding author's email:

thuha@hcmute.edu.vn

DOI:

https://doi.org/10.54644/jte.69.2022.1146

Keywords:

Classification, TV channel, Artificial intelligence, XLNet, Deep learning

Abstract

While observing the behavior of TV viewer, the HCMUTE researchers concluded that TV spectators rather watch the content of the TV channel than the channels themselves. The latters are just the containers / carriers of the precious TV information inside. Therefore to enable TV viewers to access directly the content, a new TV user graphic interface is offered to them: the TV program list is ordered by the topics of the broadcasted programs but not the channels carrying them. Therefore, viewers can now choose a program according to topics of interest, not via a intermediate step by selecting channel with dummy name / number. To classify all the broadcast programs, the researchers propose a sequence-to-sequence Model to organize these programs into one of five predefined topics / themes: feature films, news, music, sports and synthesis. The XLNet network is incorporated in the classification. TV channels currently playing the program with the selected theme will be displayed so that viewers can quickly access the wish content.

Downloads: 0

Download data is not yet available.

Author Biographies

Quang-Long Vo, Ho Chi Minh City University of Technology and Education, Vietnam

Vo Quang Long was born in 1996. Majoring in Electronics and Communication, graduated in 2019 from Ho Chi Minh City University of Technology and Education. Email longvo2396@gmail.com , Contact phone 0777116785

Ngoc-Hung-Anh Nguyen, Posts and Telecommunications Institute of Technology, Vietnam

Nguyen Ngoc Hung Anh was born in 1989 in Long An, He graduated in  major in Information Systems at  Posts and Telecommunications Institute of Technology, in  Ho Chi Minh City, 2020. Graduated with a Master's Degree in 2022; Email hunganhbc@gmail.com. Phone 0902717074

Minh-Son Tran, University of Technology HUTECH, Vietnam

Son Minh Tran was born in 1973. He received the PhD degree in Telecommunication in 2004 and became an assistant professor at the Budapest University of Technology, Hungary till 2006. He then worked as a research fellow on video compression algorithms and enhanced features for digital video broadcasting at the ARTEMIS Department at TELECOM & Management SudParis, France. He spent 10 years doing research on video security in Nagra France, Group Kudelski, leader in DTV security. Rejoining Ho Chi Minh  University of Technology and Education as senior research fellow since 2018, he continues his activity on video processing. He has a portfolio of 14 granted and pending patents on the Secured Video Transmission/ Processing and watermarking.

Thu-Hà Tran, Ho Chi Minh City University of Technology and Education, Vietnam

Tran Thu Ha was born in 1966, holds in PhD Industrial electronics, automation - telecommunications in 1996 at Kiev Polytechnic University, Ukraine. After that, she worked at the  Ho Chi Minh city University of Technology and Education since 1997 as lecturer at Electrical and electronics department. She got a title Assoc. Prof. PhD in 2011. She spent the time for researching and teaching in the fields: in electronics industrial for using neural network, fuzzy logic in auto-control, sliding control, PID, robot, voice control, IOT noise cancellation. In the area of telecommunications: conducting research some problems on teletex, image processing, noise cancellation in signal transmission systems; research on Signal processing, watermarking video security, video processing.

References

XLNet–Người khổng lồ thay thế vị trí của BERT, 06/07/2019, URL: https://trituenhantao.io/kien-thuc/xlnet-nguoi-khong-lo-thay-the-vi-tri-cua-bert/

Jacob Devlin, Ming-Wei Chang, Kenton Lee, Kristina Toutanova, BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Zhilin Yang, Zihang Dai, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le, XLNet: Generalized Autoregressive Pretraining for Language Understanding, Đại học Carnegie Mellon, Nhóm trí tuệ AI của Google.C. Waldenor, Is OTT Disrupting Television? Master Thesis, Stockholm, June 7th 2013.

Giới thiệu vể chuyên đổi câu trong xử lý ngôn ngữ tự nhiên Transformers In NLP | State-Of-The-Art-Models (analyticsvidhya.com)

Mã nguồn của mạng XLNET https://github.com/zihangdai/xlnet.J. C. Whitaker, Interactive TV Demystified, 2001 McFraw-Hill, ISBN 0-07-136325-4

A. Punchihewa, Tutorial on IPTV and its latest developments, ICIAFS January 2011. DOI: https://doi.org/10.1109/ICIAFS.2010.5715633

L. Bringuier, White Paper OTT Streaming – 2nd edition, September 2011, Anevia

T. Ohanian, Over-the-Top Considerations: Functionalities and Technologies, Cisco Systems, NAB 2014.

D. Riddlesden, A. D. Singleton, Broadband speed equity: A new digital divide, 2014, Applied Geography, Elsevier. DOI: https://doi.org/10.1016/j.apgeog.2014.04.008

Trang web chính thức của dự án OTT: https://ott.atlassian.net

Y. Cui, B. Li, K. Nahrstedt, ”Layered Peer-to-Peer Streaming”, Proc. NOSSDAV’03, Monterey, USA, pp. 162-171, Jun 2003. DOI: https://doi.org/10.1145/776322.776348

Lịch phát sóng các đài truyền hình VN: Lịch phát sóng truyền hình hàng ngày 24/7 nhanh nhất | LichPhatSongTiVi

Mô hình phân tích đoạn văn tiếng Việt: PhoBERT: The first public large-scale language models for Vietnamese – VinAI Research

Xử lý ngôn ngữ tự nhiên dùng mô hình đa ngôn ngữ: XLM-RoBERTa: The alternative for non-english NLP | by Branden Chan | deepset-ai | Medium

Downloads

Published

28-04-2022

How to Cite

Vo, Q. L., Nguyen, N. H. A., Tran, M. S., & Tran, T. H. (2022). Classification of TV Programs According to Their Topics Using the XLNet Model. Journal of Technical Education Science, 17(2), 8–16. https://doi.org/10.54644/jte.69.2022.1146

Issue

Vol. 17 No. 2 (2022)

Section

Research Article

Categories

License

Copyright (c) 2022 Journal of Technical Education Science

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Copyright © JTE.

Crossref
Scopus
Google Scholar
Europe PMC