Investigate models based on unsupervised learning

[h1alexbel]

Let's investigate models like Anomaly detection and text transformers that are based on unsupervised learning

[h1alexbel]

for classic ML unsupervised learning algorithms let's try Clustering. Some of the application contain these models:

• Hiearchical clustering - connectivity based clustering
• K-means clustering - centroid based clustering
• Principal Component Analysis (PCA)
• Singular Value Decomposition (SVD)
• Independent Component Analysis (ICA)
• Gaussian Mixture Models (GMM) - model based clustering
• Density-Based Spatial Clustering of Applications with Noise (DBSCAN) - densitiy based clustering

let's study them and pick one of them, that mostly applicable for our use-case

[h1alexbel]

Let's try Isolation Forest as classic one. Gaussian Mixture Models can be used as a sub contender.

[0pdd]

@h1alexbel the puzzle #139 is still not solved.

[h1alexbel]

for transformers we trying BERT and SRdT (custom one)

[0pdd]

@h1alexbel 2 puzzles #139, #154 are still not solved.