UnSupervised Machine Learning

Here we only have Features but don’t have any Target column in our dataset, hence its known as UnSupervised ML.

It’s more complex as compared to Supervised ML, and also needs some level of manual intervention once unsupervised modelling is done to further investigate the results.

• Common UnSupervised ML Models are:
  1. Dimensionality Reduction (used to reduce number of features by deriving new features from old)
     • PCA (Principal Component Analysis)
     • ICA (Independent Component Analysis)
     • SVD (Singular Value Decomposition)
     • NMF (Non-Negative Matrix Factorization)
     • LDA
     • t-SNE
     • UMAP
     • Isomap
     • MDS
  2. Clustering (Used to group similar observations together into clusters based on some similarity metric like we use in KNN model)
     • Hierarchal Clustering
       • Divisive
       • Agglomerative
     • Density Based Clustering
       • DBSCAN
       • OPTICS
     • Centroid/Partition Based Clustering (Non_Overlapping Clusters)
       • K-means (MiniBatch Variation)
       • K-medoids
       • K-mode
     • Fuzzy Clustering (Overlapping Clusters)
       • C-means
  3. Recommender Systems (Used by organizations to recommend useful content to their customers based on their features and preferences. e.g. movie recommendations by Netflix, Amazon Prime and Hotstar)
     • Content Based
     • Collaborative Filtering
       • Memory Based
         • User Based
         • Item Based
       • Model Based
         • Dimensionality Reduction
         • Matrix factorization
         • Autoencoders
  4. ARM (Association Rule Mining) [ Used in Market Basket Analysis (MBA) ]
     • Apriori Algo
     • Eclat
     • FP-growth
  5. Anomaly Detection (Finding instances that are deviating from the general/usual patterns e.g. outliers, malicious traffic to a website or server)