This session covers creation of a recommender using Apache Mahout for a consumer Web application. After attending this session, application developers will be able to notice the need for using recommenders in their application and will be able to start planning and implementing them for their specific use cases.

Outline

Recommendation engines, or recommenders, are widely used by many applications for suggesting objects users may like. For example, an online shopping site will suggest products users may like depending on what they have bought and/or visited earlier. This session covers creation of a recommender for a consumer Web application.

We will talk about recommenders in the context of a specific real world use case, covering:

1. What is a recommender,
2. How to identify the essential input for a recommender, i.e. users and items,
3. Designing the recommender
4. Deploying it as part of a Web application
5. Tuning the recommender

We’ll be using Apache Mahout to create parts of the recommender. Apache Mahout is a machine learning library which also plays nice with Apache Hadoop for processing. We will, however, be focusing more on how to use it without Hadoop for this session.

Speaker bio

Viraj is a Software Architect at GS Lab. For the last 7 years while at GS Lab, he has worked in the area of Web Applications. His current area of focus is Data Analytics, Design/Development of Scalable Web Applications and exploring Data Analytics use cases in Web based products. As a part of this, he created a recommender for a social news reader product. Prior to this, he has worked in Web Applications Security, developing Web-based attacks for an enterprise Web security assessment product.

He is a Computer Science M.Tech. from IIT-Kharagpur. Before that, he did his M.Sc. in Mathematics from University of Pune.