ABOUT THE COURSE :Recommender Systems have been a prevalent area of research for a long time. They have been applied to various dimensions, ranging from marketing, education, social media, financial services, and more. Recommender systems have changed the way people find products, information, and even other people. Recommender systems discover information items (e.g., people, products) that are likely to be of interest to users. Such systems study patterns of behavior to know what someone will prefer from among a collection of things he has never experienced. At a high-level, recommendation systems are pieces of software equipped with data mining and machine learning tools that aim to recommend products or information to users, based on certain preferences. The proposed course aims to cover the following aspects of recommender system with a focus of developing such systems in Web based environment.1. Theoretical foundations 2. Data preprocessing and preparation3. Algorithms 4. Performance evaluation INTENDED AUDIENCE : Students/Industry ParticipantsPREREQUISITES : At least pursuing BTechINDUSTRY SUPPORT : Ecommerce Companies
Overview
Syllabus
Week 1:
Introduction
Business value of Recommender System
A conceptual framework for understanding recommender system
Types of recommender system
Week 2:Data for recommendation: Explicit Vs Implicit data collection
Scales of measurement
Statistical and machine learning foundations for recommender system
Data preprocessing
Week 3:Introduction to collaborative filtering
Collaborative filtering approaches: Memory based and model based
Memory based collaborative filtering foundations: Distance and similarity measures
User based collaborative filtering; Item based collaborative filtering
Week 4:Model based collaborative filtering foundations: matrix factorization, UV decomposition, Singular value decomposition Model based collaborative filtering techniques: SVD, SVD++ etc
Week 5:Content based recommender system foundations
Examples with text data
Feature engineering: Feature extraction, feature selection, dimensionality reduction
Week 6:Content based recommender system examples with few supervised machine learning techniques
Week 7:Evaluation of recommender systems: Online and offline evaluation, metrics such as RMSE, AME, Good Item MAE, Good predicted item MAE, Precision, Recall, F1 Measure, NDCG, Average Reciprocal Rank, Top@N Measure.
Week 8:Overview of other types of recommender systems such as trust based, social network based, and context aware systems
Introduction
Business value of Recommender System
A conceptual framework for understanding recommender system
Types of recommender system
Week 2:Data for recommendation: Explicit Vs Implicit data collection
Scales of measurement
Statistical and machine learning foundations for recommender system
Data preprocessing
Week 3:Introduction to collaborative filtering
Collaborative filtering approaches: Memory based and model based
Memory based collaborative filtering foundations: Distance and similarity measures
User based collaborative filtering; Item based collaborative filtering
Week 4:Model based collaborative filtering foundations: matrix factorization, UV decomposition, Singular value decomposition Model based collaborative filtering techniques: SVD, SVD++ etc
Week 5:Content based recommender system foundations
Examples with text data
Feature engineering: Feature extraction, feature selection, dimensionality reduction
Week 6:Content based recommender system examples with few supervised machine learning techniques
Week 7:Evaluation of recommender systems: Online and offline evaluation, metrics such as RMSE, AME, Good Item MAE, Good predicted item MAE, Precision, Recall, F1 Measure, NDCG, Average Reciprocal Rank, Top@N Measure.
Week 8:Overview of other types of recommender systems such as trust based, social network based, and context aware systems
Taught by
Prof. Mamata Jenamani
