Advertisement

News Collection and Analysis on Public Political Opinions

  • Zhi-Qian Hong
  • Fang-Yie LeuEmail author
  • Heru Susanto
Conference paper
Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 97)

Abstract

With the fast development of news media and freedom of speech in Taiwan, some news is not objectively reported. In fact, in order to attract people’s attention and increase the click rates of news, many journalists did not convey the exact meanings of news, even distorting news meanings or adding some subjective criticisms or opinions. As a result, news confusions come out one after the other. Based on the analysis of political opinion news, this study would like to analyze certain political characters, such as candidates during a certain period of time, for example, the election period. Last year (2018), Kaohsiung-city-mayor election was held in December.

We develop a news gathering and analytical scheme, named Focused News Collection and analytical System (FNCaS), which predicts which candidate might be the winner. By analyzing the possible outcomes for readers through big data analysis techniques and deep learning approaches after some amount of news were gathered. The purpose is to reduce the time for readers to absorb news essentials, and to conclude the possible results of the analyses immediately, aiming to improve the efficiency that people access to news contents and understand the implications behind it. Our conclusion is that the FNCaS has capability in collecting news immediately and analyzing some amount of news of focused domains efficiently.

Keywords

News Public political opinion Deep learning Artificial intelligence LSTM 

References

  1. 1.
    Turing, A.M.: Computing machinery and intelligence. Mind 59(236), 433–460 (1950)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Rong, X.: Word2vec parameter learning explained. eprint arXiv: 1411.2738, November 2014. https://arxiv.org/pdf/1411.2738.pdf. Accessed 27 June 2019
  3. 3.
    Sherstinsky, A.: Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network. eprint arXiv:1808.03314, August 2018. https://arxiv.org/pdf/1808.03314.pdf
  4. 4.
    Ramos, J.: Using TF-IDF to determine word relevance in document queries, January 2003. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.121.1424&rep=rep1&type=pdf. Accessed 4 June 2019
  5. 5.
    Gensim (2009). Accessed 5 June 2019Google Scholar
  6. 6.
    Kingma, D.P., Ba, J.L.: Adam: a method for stochastic optimization. eprint arXiv:1412.6980, December 2014. https://arxiv.org/pdf/1412.6980.pdf. Accessed 10 June 2019

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Department of Computer ScienceTunghai UniversityTaichung CityTaiwan
  2. 2.Research Center for InformaticsThe Indonesian Institute of SciencesJakartaIndonesia
  3. 3.Information Management DepartmentTunghai UniversityTaichung CityTaiwan

Personalised recommendations