Anikait Kapoor
Department of Computer Science and Engineering, Apex Institute of Technology, Chandigarh University, Mohali, Punjab, India
Debavushan Saikia
Department of Computer Science and Engineering, Apex Institute of Technology, Chandigarh University, Mohali, Punjab, India
Ishaan Dhawan
Department of Computer Science and Engineering, Apex Institute of Technology, Chandigarh University, Mohali, Punjab, India
Download PDF
http://doi.org/10.37648/ijrst.v14i01.002
With the rise in mobile awareness in recent years, the short message service (SMS) industry has generated billions of dollars in revenue. However, this has led to an increase in unwanted commercial advertising or spam sent to regular phones, with parts of Asia having up to 30% of content messages as spam in 2012. One of the challenges in SMS spam filtering, it requires a comprehensive database and the limited usefulness and dialect used in SMS. In this extension, analysts used a real SMS spam database from the UCI Machine Learning store and connected different machine learning methods after preprocessing and extracting markup. The results were compared and the main spam filtering algorithms for the message body were distinguished. The final reconstruction using 10-fold cross-validation appeared to have the primary classifier more than halve the overall error rate compared to the best proof in a paper.
Keywords: supervised learning; classification algorithms; feature engineering; natural language processing (NLP); text classification
Disclaimer: All papers published in IJRST will be indexed on Google Search Engine as per their policy.
