SMS Spam Detection using Supervised Learning

Abstract

Over the last decade, the growth of short message services has been rising. These text messages are more powerful for corporations than even SMS. This is because about 80 percent of sms remain unopened while 98 percent of smartphone users read theirs by the end of the day. Spam, which refers to any irrelevant text messages sent via mobile networks, has also gained popularity. For consumers, they are seriously irritating. Due to the geographical material, use of abbreviated words, the current Spam Detection techniques are more challenging than e-mail spam detection techniques , unfortunately very few of the existing research addresses these challenges. Much of the current research that has attempted to filter Spam has focused on features that were manually found. This paper aims to solve these concerns. Filtering is one of the most effective strategies among the methods developed to stop spam. Days of machine learning techniques are now used to process the spam SMS automatically at a very good rate. The goal of this research is to differentiate between ham and spam messages by developing an accurate and responsive model of classification that provides good accuracy with a low false positive rate