Summary

This is the last chapter of the analytics domain. So far, you have learned a lot of concepts that can help us build amazing analytics applications. In this chapter, you learned how to make a recommendation engine for an e-commerce product. In the baseline approach, we used the concept of TF-IDF and cosine similarity. In the revised approach, we built a book recommendation system that used the concept of correlation. In the best approach, we used the KNN algorithm to build a recommendation engine that used a collaborative-filtering-based approach. We looked at the advantages and disadvantages of all the approaches. You also learned about the architecture of the recommendation system. All these topics will help you understand and build your own recommendation system. You can also build a computer vision-based recommendation engine. This kind of recommendation engine really changes the way content is recommended to the users. So don't hesitate to build new types of recommendation systems...