An Approach to Pattern Prediction and Early Recognition of Lung Cancer Employing Machine Learning Techniques

Lung cancer remains a major issue despite the fact that early identification is crucial for bettering patient outcomes due to poor prognoses and delayed diagnosis. By creating and implementing a novel machine learning architecture intended for early lung cancer diagnosis, our work aims to address this problem. To do this, an extensive dataset including a broad spectrum of health-related attributes is utilised. These include age, gender, lifestyle choices, exposure to the environment, and a variety of lung cancer symptoms, such as fatigue, shortness of breath, blood in the cough, and discomfort in the chest and weight loss. Strict testing and training protocols are used in the research to search for minute irregularities that could point to early-stage lung cancer. The study specifically focuses on assessing the decision tree's effectiveness (J48). As well as the Support Vector Machine (SVM), achieving a flawless 100% success rate in the early detection of lung cancer. These findings demonstrate how machine learning methods have the potential to significantly enhance lung cancer early detection. Successful early detection might lead to timely therapies that improve patient outcomes and reduce the death rate from the condition. The study also emphasises how important it is to build cancer detection methods and improve prediction accuracy by including a range of health-related traits and symptoms in machine learning models.