/Social-Spam-Detection

Spam detection task on twitter dataset

Primary LanguageJupyter Notebook

Social-Spam-Detection

Spam detection studied on the Twitter dataset by building 3 case studies:

  1. Case 1 - Using all numerical features
  2. Case 2 - Selecting top 7 features by using SelectKBest package from SKlearn
  3. Case 3 - Performing PCA and selecting the features explaining 95% variance

Grid search CV is employed to find the optimal hyperparameters for the following classification algorithms -

  1. Naive Bayes
  2. KNN
  3. SVM
  4. Decision Tree
  5. Random Forest
  6. Multi Layer Perceptron