Spam Classification Using Machine Learning: Performance Evaluation and Insights
Jitha Janardhanan, Rosita Kamala, Simna Shanavas · 2025
The article discusses about using machine learning methods for determining unwanted messages. The expansion of portable phone applications has driven to an uncommon rise in SMS spam. Although considered reliable and efficient in many parts of the world, recent data clearly shows a significant increase in the volume of mobile phone spam is consistently expanding. Sorting through SMS fake is also late errand to resolve this kind of issue. The algorithms such as RandomForest,Logistic Regression and SVM are utilized to recognize spam and ham messages. The effectiveness each channel in identifying spam and legitimate messages ham is too inspected. One imperative way to reduce the issue of spam in versatile communication is to create proficient spam location algorithms.The libraries like NLTK is used for preprocessing, and distinctive calculations, , and tokenizing are all carried out.. The algorithms achieved an accuracy of 96%,97%,98% with SVM, Logistic Regression and RandomForest