Courses studied
3 enrolledProjects
2 publishedANTIBIOTIC RESISTANCE
This is a binary classification model that detect either resistant or Non-resistant to antibiotics in healthcare services. Modules/Libraries Used; I. Pandas – Data handling II. Scikit-learn – Model building II. Joblib – Saving and loading the model Dataset; - A 5,000-row realistic dataset was used to build the ML model. - The dataset contains 8 columns: 7 features and 1 target variable. Model Algorithm; - DecisionTreeClassifier() was used to train the model.Model Evaluation; Since this is a classification model, the following evaluation metrics were used: I. Accuracy Score (measures overall correctness) = 99.9% II. Classification Report (provides detailed performance per class): - Precision (out of predicted positives, how many were correct) = 100% - Recall (out of actual positives, how many were correctly identified) = 100% - F1-Score (balance between precision and recall) = 100%
MLM-101anemiaclasspredictionApp
This model was trained with DecisionTreeClassifier () algorithm. It operates by mapping function of the features and predicts the suitable or best output for an unseen input. The input are numerical values obtain from clinical laboratory. Since all the features were not categorical, StandardeScaler () is used in ColumnTransformer () to scaled the numerical values. When evaluating the model performance, accuracy_score was used and the score is 99.56%.
MLM-101