Recommender Systems are a very important component of most businesses these days, and one of the most prominent applications of machine learning. There are tons of success stories associated with using these systems, but how can you build one yourself? Which machine learning techniques are typically used to design one, and how do they compare with top-of-the-line deep learning methods?
Deep learning, of course, has seen a ton of success in functions like Image and Video Processing using CNNs, and the like. But recommender systems are built on structured data, so how can we leverage deep learning techniques in this case?
In this hack session, Kiran R will go through the existing techniques in recommender systems, starting with heuristic methods, nearest neighbour techniques (user kNN, item kNN, etc.), matrix factorization, SVD, SVD++, etc. and contrasting them with state-of-the-art deep learning recommender system techniques.
So, to summarize, Kiran will cover the below in this awesome knowledge-packed one hour hack session:Feature engineering, Data wrangling and Visualization are all aspects of Data Preparation – one of the most important phases in any standard data mining or machine learning workflow.
Create a train and validation set from an existing publicly available dataset
Run the above mentioned machine learning techniques
Compare them against a deep learning system
Hackers
Kiran drives data sciences & advanced analytics projects across sales & marketing, digital, partner, pricing and e-commerce in his functional role. He also dotted line leads the Information Innovation Center (IIC) in India which has teams like Master Data Management, Business Intelligence & Analytics.
Duration of Hack-Session: 1 hour