Early Thyroid prediction system
Problem Statement
1. Thyroid disease is highly prevalent in India, affecting over one crore individuals annually, with a notable predilection towards females. 2. Hyperthyroidism and hypothyroidism are the most prevalent thyroid disorders, stemming from irregular thyroid gland function. 3. These disorders significantly impact metabolism, leading to either accelerated or slowed metabolic rates. 4. Despite technological advancements, early detection of thyroid disorders remains crucial for effective management. 5. Leveraging Artificial Intelligence, specifically machine learning algorithms, presents an opportunity to enhance disease detection and improve overall quality of life. 6. This study aims to evaluate the efficacy of various classification algorithms, including Logistic Regression, Random Forest, Decision Tree, NaΓ―ve Bayes, and Support Vector Machine, in predicting thyroid disease presence for better patient outcomes.
Objective
Encoding categorical data
Encoding the Independent Variable
Encoding the Independent Variable
Standardization
Combining data
Handling imbalanced Dataset
Since the dataset is small, will use over-sampling: SMOTE technique to balance the data