Course details

Powering Recommendation Engines: Recommendation Engines

Powering Recommendation Engines: Recommendation Engines


Overview/Description
Expected Duration
Lesson Objectives
Course Number
Expertise Level



Overview/Description

This 13-video course explores recommendation engines, systems which provide various users with items or products that they may be interested in by observing their previous purchasing, search, and behavior histories. They are used in many industries to help users find or explore products and content; for example, to find movies, news, insurance, and a myriad of other products and services. Learners will examine the three main types of recommendation systems: item-based, user-based or collaborative, and content-based. The course next examines how to collect data to be used for learning, training, and evaluation. You will learn how to use RStudio, an open-source IDE (integrated development environment) to import, filter, and massage data into data sets. Learners will create an R function that will give a score to an item based on other user ratings and similarity scores. You will learn to use R to create a function called compareUsers, to create an item-to-item similarity or content score. Finally, learn to validate and score by using the built-in R function RMSE (root mean square error).



Expected Duration (hours)
1.1

Lesson Objectives

Powering Recommendation Engines: Recommendation Engines

  • Course Overview
  • describe what a Recommendation Engine does, how it can be used, and the types and reasons they are used
  • compare the different types of Recommendation Engines and how they can be used to solve different recommendation problems
  • describe the process of collecting data and why data sets that can be used for learning, training, and evaluating a Recommendation Engine are needed
  • use R to import, filter, and massage data into data sets
  • describe how Similarity and Neighborhoods can be used to score users and items against another user or a new item
  • create an R function that will score a user against another user to compare their similarity
  • create an R function that will give a score to an item a user has not seen before based on other users' ratings and similarity scores
  • create an R function that finds similar users and finds products they liked which would be good to recommend to the user
  • use R to create an Item to Item similarity, or content, score to Recommend similar items
  • evaluate a Recommendation Engine by using known data and metrics to calculate the accuracy of recommendations
  • validate and score a Recommendation System using R and an evaluation data set
  • describe the types and interfaces required to build a Recommendation System
  • Course Number:
    it_dsprendj_01_enus

    Expertise Level
    Beginner