THE ANATOMY
Before we dive into exploring specific algorithms, we first need to examine how recommender systems fit into the broader landscape of data science.
Data science, itself, is an interdisciplinary field of methodologies and algorithms to extract knowledge or insight from data. Within the vast space of data science lies the popular field of artificial intelligence (AI), which is the ability of machines to simulate intellectual tasks. A prominent sub-field of artificial intelligence is machine learning, among other sub-fields such as perception, and search and planning. Recommender systems fall under the banner of machine learning and to some extent data mining.
Figure 1: Visual representation of data-related fields and sub-fields
Machine learning applies statistical methods to improve performance based on previous experience. While the programmer is responsible for feature selection and setting the model’s hyperparameters (algorithm learning settings...