Machine Analytics has a broad and deep expertise in a range of machine learning techniques and tools along with various optimization methods, such as Expectation Maximization (EM), that are employed in the learning process.
Generative Methods
- Naïve Bayesian Classifier (NBC)
- Hidden Markov Models (HMM) and extensions (Hierarchical, Factorial)
- Mixture Gaussian models
- Bayesian Belief Networks
- Latent Dirichlet Allocation (LDA)
- Probabilistic Latent Semantic Analysis
Discriminative Methods
- Linear and Logistic Regression
- Support Vector Machine (SVM)
- Artificial Neural Networks
- Nearest Neighbor
Clustering Methods
- Subspace methods (SVD, LSA, PCA, ICA), SVM
- Hierarchical and k-Means
Deep Learning
- Restricted Boltzmann Machine (RBM) and Deep Belief Network (DBN)
- Convolutional Neural Network (CNN)
- Stacked Auto-Encoders
- Deep Recursive Neural Network (RNN)
- Deep Fusion: Application of above deep learning techniques to solve data fusion problems.