Implementation Of The Random Forests Method On The Retweet Classification Model Based On Content
DOI:
https://doi.org/10.34818/eoe.v9i3.17991Abstract
Abstract Twitter as one of the biggest social media on the internet has been used as the center of information exchangeon mainstream media. As this paper was written Covid-19 information sporadically propagated through twitter. To help spread validated information to the masses we need to understand which factors are relevant and support the information diffusion. In this paper author tried to find similarities between tweetsby using TF-IDF, author also applied content features from tweet’s meta-data to random forests classifier to predict which tweets users might retweet. The result of the shows that by using content features, machinelearning models can predict retweets from users. The proposed method of combining content features fromtwitter metadata and TF-IDF leads to a better model than the stand-alone features with 69.97% of accuracy. Keywords: Retweet Prediction, TF-IDF, Random ForestReferences
[1] M. Li, X. Wang, K. Gao, and S. Zhang, “A survey on information diffusion in online social networks: Models and methods,” Inf., vol. 8, no. 4, 2017, doi: 10.3390/info8040118.
[2] J. Kim, J. Bae, and M. Hastak, “Emergency information diffusion on online social media during storm Cindy in U.S.,” Int. J. Inf. Manage., vol. 40, no. February, pp. 153–165, 2018, doi: 10.1016/j.ijinfomgt.2018.02.003.
[3] G. Piao and W. Huang, “Regression-enhanced random forests with personalized patching for covid-19 retweet prediction,” CEUR Workshop Proc., vol. 2881, pp. 13–16, 2020.
[4] H. Bunyamin and T. Tunys, “A comparison of retweet prediction approaches: The superiority of random forest learning method,” Telkomnika (Telecommunication Comput. Electron. Control., vol. 14, no. 3, pp. 1052–1058, 2016, doi: 10.12928/TELKOMNIKA.v14i3.3150.
[5] S. N. Firdaus, C. Ding, and A. Sadeghian, “Retweet: A popular information diffusion mechanism – A survey paper,” Online Soc. Networks Media, vol. 6, pp. 26–40, 2018, doi: 10.1016/j.osnem.2018.04.001.
[6] T. B. N. Hoang and J. Mothe, “Predicting information diffusion on Twitter – Analysis of predictive features,” J. Comput. Sci., vol. 28, pp. 257–264, 2018, doi: 10.1016/j.jocs.2017.10.010.
[7] N. Shoeibi, N. Shoeibi, P. Chamoso, Z. Alizadehsani, and J. M. Corchado, “Similarity approximation of Twitter Profiles,” no. June, pp. 1–18, 2021, doi: 10.20944/preprints202106.0196.v2.
[8] I. Daga, A. Gupta, R. Vardhan, and P. Mukherjee, “Prediction of likes and retweets using text information retrieval,” Procedia Comput. Sci., vol. 168, pp. 123–128, 2020, doi: 10.1016/j.procs.2020.02.273.
[9] F. Resyanto, Y. Sibaroni, and A. Romadhony, “Choosing The Most Optimum Text Preprocessing Method for Sentiment Analysis: Case:iPhone Tweets,” Proc. 2019 4th Int. Conf. Informatics Comput. ICIC 2019, pp. 2–6, 2019, doi: 10.1109/ICIC47613.2019.8985943.
[10] F. Pedregosa et al., “Scikit-learn: Machine learning in Python,” J. Mach. Learn. Res., vol. 12, no. May 2014, pp. 2825–2830, 2011.
[11] S. N. Firdaus, C. Ding, and A. Sadeghian, “Retweet Prediction based on Topic, Emotion and Personality,” Online Soc. Networks Media, vol. 25, no. August, p. 100165, 2021, doi: 10.1016/j.osnem.2021.100165.
