Filtration of Unwanted Messages From Osn User Wall Using Machine Learning
M.S.Sivapriya1, Aditi Tiwari2, Ritesh Kumar Singh3
1M.S.Sivapriya Assistant Professor in the Department of CSE, At SRMIST, Chennai (Tamil Nadu), India.
2Aditi Tiwari, Student in the Department of CSE, From SRMIST, Chennai (Tamil Nadu), India.
3Ritesh Kumar Singh, Student in the Department of CSE, From SRMIST, Chennai (Tamil Nadu), India.
Manuscript received on 01 May 2019 | Revised Manuscript received on 15 May 2019 | Manuscript published on 30 May 2019 | PP: 1358-1362 | Volume-8 Issue-7, May 2019 | Retrieval Number: F3874048619/19©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: The platform to make friends and pass message among each other are becoming a powerful source and tool of communication. Social Networking sites serves as the best platform for entertainment of upcoming generation. OSNs helps the users to connect online to others in order to communicate and share their various experiences in the forms of posts and status. But now-a-days Online Social Networking are facing many problems of posting annoying content on someone else’s profile which make others humiliated after seeing this. In arrangement to eliminate these foul words, machine learning is used that will filter the unbearable word from the present content. The content of social media are amorphous Since the data (textual content) on online media is mainly unstructured and often in casual style, the existing research on message-level offensive language detection cannot detect the accurate offensiveness of the content. In comparison with message-level offensiveness detection, the identification at user level will be more viable but this is under analysis stage. In disposition for removal ofobjectionable words from an OSN user’s wall, a new system will be offered which will have LSF(Lexical Syntactic Feature), the objectionable content will be filtered based on LSF. Different approaches like Bag of Words(Bow) and n-gram will be used through which filtration of bad words will occur. Thus, the focus of the ongoing work is to offer a mode that will filter the unrelated messages and propose a Filtered Wall (FW).
Keyword: Online Social Network(OSN); Offensive words, Lexical Syntactic Feature(LSF), Bag of Words (BoW), n-Gram Algorithms; Data Filtration, Short Text Classification.
Scope of the Article: Machine Learning.