Machine-Learning-Projects

Welcome! This repository features a collection of machine learning models that I have built and practiced. Through these projects, I have implemented and explored a wide range of core machine learning concepts, including:

• Data Preprocessing & Feature Engineering

  • Standard scaling and normalization.
  • Label encoding for categorical variables.
  • Text preprocessing, including stopwords removal and stemming.
  • TF-IDF vectorization for text-based features.
  • Outlier analysis and removal (e.g., using the IQR rule to identify outliers, followed by counting them or replacing them with the median).
  • Feature selection using correlation matrices (e.g., identifying and removing highly correlated columns).

• Handling Imbalanced Datasets

  • Undersampling techniques.
  • SMOTE (Synthetic Minority Over-sampling Technique) for oversampling.

• Model Optimization

  • Hyperparameter tuning using techniques like RandomizedSearchCV/GridSearchCV.

• Algorithms Built From Scratch

  • Linear Regression
  • Logistic Regression
  • Lasso Regression
  • Support Vector Machine (SVM) Classifier
  • K-Nearest Neighbors (KNN) Classifier

• Unsupervised Learning & Similarity Metrics

  • K-Means Clustering (applied in the customer clustering project).
  • Cosine Similarity (applied in the movie recommender system).

• Model Evaluation & Testing Metrics

  • accuracy_score and cross_val_score
  • confusion_matrix and classification_report
  • $R^2$ score and Mean Absolute Error (MAE)

...and a few more topics!