In this chapter, we have seen how effectively we can use and combine the LDA algorithm and NLP libraries, such as Stanford NLP, for finding useful patterns from large-scale text. We have seen a comparative analysis between TM algorithms and the scalability power of LDA.

Netflix is an American entertainment company founded by Reed Hastings and Marc Randolph on August 29, 1997, in Scotts Valley, California. It specializes in, providing, streaming media and video-on-demand, online and DVD by mail. In 2013, Netflix expanded into film and television production, as well as online distribution. Netflix uses a model-based collaborative filtering approach for real-time movie recommendations for its subscribers.

In the next chapter, we will see two end-to-end projects: an item-based collaborative filtering for movie-similarity measurements, and a model-based movie-recommendation engine with Spark to recommend movies to new users. We will see how to interoperate between ALS and Matrix Factorization for these two scalable movie recommendation engines.