NLP-Based SMS Spam Detection Using Ensemble Learning

In today's environment, spam text messages are a common issue and a source of frustration for users. This study suggests a novel method for identifying SMS spam that makes use of natural language processing (NLP), multiple model creation, and ensemble learning. The suggested approach entails pre-processing the data using natural language processing (NLP) techniques including lemmatizing, stemming, and stopword removal in order to extract pertinent features from the text data and convert it into numerical representations. The suggested approach's effectiveness was assessed using an actual dataset and contrasted with conventional machine learning algorithms like SVM and Naïve Bayes.The altered data is used to train the numerous classifiers, and an ensemble learning technique is then used to combine their predictions and produce a more accurate outcome. More precision was provided by the final model than by conventional models. In order to lessen the quantity of spam messages that users get, the research's conclusions may be helpful in the development of an efficient spam SMS detector. Furthermore, the model was hosted using the streamlit web app framework on the cloud-based platform Railway, making it instantly accessible to users without the need to install any software.