| Summary: | The sheer volume of information available on the internet far exceeds our ability to consume it. There is so much content competing for our time that it is not practical for users to sieve through the information themselves. In order to keep users on their platform, companies have to provide personalized recommendations that are highly relevant to their interests.
While there are many well established techniques for providing recommendations to users, the complexity involved is prohibitive for smaller internet platforms that do not have the engineering expertise. Furthermore, many of these internet platforms have a large enough content base such would benefit from a recommendation engine.
In this project, I have set out to identify techniques for recommendation engines that would be applicable to ecommerce and online media and content platforms. I will discuss the key concepts and workings of six different recommendation techniques. These techniques will be implemented and evaluated against the popular movie lens data set. Finally, the findings will be used in developing a recommendation engine suited for ecommerce and online media and content platforms.
|
|---|