Control on Predatory Comments on Social Media Platform using Full-Text Search Algorithm
B. Mathangi1, Anisha Rajesh2, Disha Dikshita Behera3
1B. Mathangi, Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai, India.
2Anisha Rajesh, Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai, India.
3Disha Dikshita Behera, Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai, India.
Manuscript received on 19 April 2019 | Revised Manuscript received on 26 May 2019 | Manuscript published on 30 May 2019 | PP: 810-814 | Volume-8 Issue-1, May 2019 | Retrieval Number: A9207058119/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Social media is quickly evolving in front of our eyes and it is almost impossible to reject and hide from this new form of media. Not only is it an important part of socialization within peer groups but now it is used to market and motivate people to become a part of a larger community. But, young creators in this community are suffering from child predatory comments from users which consists of un-parliamentary words that affect the society a lot. One such social media platform, YouTube, had recently faced a lot of trouble due to the unavoidable predatorily comments. Since the existing system disables the entire comment section by observing some predatory comments, the restriction of interaction and feedback opinion will affect the creator’s efforts. This project focuses on discarding only those comments that contains profanity and vulgar words,on run time, before its post. This system lets commentators even for negative feedbacks, as comments are one of the most important features in social media platform, that lets both users and viewers to explore more on their subjects. The dataset is trained from a pre-defined database that consists of offensive and slang terms mostly used in media platforms. The predefined set is connected with Elastic Search open-source full text engine that enables run time search. The comment typed by the user is tokenized word by word and checked with the set using Full Text Search algorithm. If found any, the comment is declined from posting further leaving an notification behind, thereby enabling the comment section still active for normal and kid friendly interaction.
Index Terms: Full-Text Search, Elasticsearch, Predatory, Scrutinize.
Scope of the Article: Social Sciences