9 Selected unsupervised learning algorithms
This chapter covers
- Latent Dirichlet allocation for topic discovery
- Density estimators in computational biology and finance
- Structure learning for relational data
- Simulated annealing for energy minimization
- Genetic algorithm in evolutionary biology
- ML research: unsupervised learning
In the previous chapter, we looked at unsupervised ML algorithms to help learn patterns in our data; this chapter continues that discussion, focusing on selected algorithms. The algorithms presented in this chapter have been included to cover the breadth of unsupervised learning, and they are important to learn because they cover a range of applications, from computational biology to physics to finance. We’ll start by looking at latent Dirichlet allocation (LDA) for learning topic models, followed by density estimators and structure learning algorithms, and concluding with simulated annealing (SA) and genetic algorithms (GAs).