• No products in the cart.

ratings 

Working in a hands-on learning environment, led by our Recommendation Systems expert instructor, students will learn about and explore: you should be able to read code in a programming language such as Python or Java .you should understand an SQL query, and you should have a basic understanding of higher math and statistics.Figures and code listings that explain concepts can get you only so far.

PRIVATE
Course Access

Unlimited Duration

Last Updated

January 25, 2021

Students Enrolled

Total Video Time

EXPIRED

Posted by
Certification

Course Description

Are you envious when Amazon recommends its products or when Netflix is spot-on with a recommendation for a user? Then here’s your chance to learn how to add these skills to your repertoire. Reading this course will give you an understanding of what recommender systems are and how to apply them in practice. To make a recommender work, many things need to perform in concert. You need to understand how to collect data from your users and how to interpret it, and you need a toolbox of different recommender algorithms so you can choose the best one for your particular scenario. Most importantly, you need to understand how to evaluate whether your recommender system is doing its job well. All this and more is hidden within this course.

Profile Photo
3.86 3.857142857142857
424

Studens

About Instructor

More Courses by Insturctor

Course Curriculum

    • Real-life recommendations 00:00:00
    • Taxonomy of recommender systems 00:00:00
    • Machine learning and the Netflix Prize 00:00:00
    • The MovieGEEKs website 00:00:00
    • Building a recommender system 00:00:00
    • How (I think) Netflix gathers evidence while you browse 00:00:00
    • Finding useful user behavior 00:00:00
    • Identifying users 00:00:00
    • Getting visitor data from other sources 00:00:00
    • The collector 00:00:00
    • What users in the system are and how to model them 00:00:00
    • Why adding a dashboard is a good idea 00:00:00
    • Doing the analytics 00:00:00
    • Personas 00:00:00
    • MovieGEEKs dashboard 00:00:00
    • User-item preferences 00:00:00
    • Explicit or implicit ratings 00:00:00
    • Revisiting explicit ratings 00:00:00
    • What are implicit ratings? 00:00:00
    • Calculating implicit ratings 00:00:00
    • How to implement implicit ratings 00:00:00
    • Less frequent items provide more value 00:00:00
    • What’s a non-personalized recommendation? 00:00:00
    • How to make recommendations when you have no data 00:00:00
    • Implementing the chart and the groundwork for the recommender system component 00:00:00
    • Seeded recommendations 00:00:00
    • What’s a cold start? 00:00:00
    • Keeping track of visitors 00:00:00
    • Addressing cold-start problems with algorithms 00:00:00
    • Those who doesn’t ask, won’t know 00:00:00
    • Using association rules to start recommending things fast 00:00:00
    • Why similarity? 00:00:00
    • Essential similarity functions 00:00:00
    • k-means clustering 00:00:00
    • Implementing similarities 00:00:00
    • Collaborative filtering: A history lesson 00:00:00
    • Calculating recommendations 00:00:00
    • Calculating similarities 00:00:00
    • Amazon’s algorithm to precalculate item similarity 00:00:00
    • Ways to select the neighborhood 00:00:00
    • Finding the right neighborhood 00:00:00
    • Ways to calculate predicted ratings 00:00:00
    • Prediction with item-based filtering 00:00:00
    • Cold-start problems 00:00:00
    • A few words on machine learning terms 00:00:00
    • Collaborative filtering on the MovieGEEKs site 00:00:00
    • What’s the difference between association rule recs and collaborative recs? 00:00:00
    • Levers to fiddle with for collaborative filtering 00:00:00
    • Pros and cons of collaborative filtering 00:00:00
    • Business wants lift, cross-sales, up-sales, and conversions 00:00:00
    • Why is it important to evaluate? 00:00:00
    • How to interpret user behavior 00:00:00
    • What to measure 00:00:00
    • Before implementing the recommender… 00:00:00
    • Types of evaluation 00:00:00
    • Offline evaluation 00:00:00
    • Offline experiments 00:00:00
    • Implementing the experiment in MovieGEEKs 00:00:00
    • Evaluating the test set 00:00:00
    • Online evaluation 00:00:00
    • Continuous testing with exploit/explore 00:00:00
    • Business wants lift, cross-sales, up-sales, and conversions 00:00:00
    • Why is it important to evaluate? 00:00:00
    • How to interpret user behavior 00:00:00
    • What to measure 00:00:00
    • Before implementing the recommender… 00:00:00
    • Types of evaluation 00:00:00
    • Offline evaluation 00:00:00
    • Offline experiments 00:00:00
    • Implementing the experiment in MovieGEEKs 00:00:00
    • Evaluating the test set 00:00:00
    • Online evaluation 00:00:00
    • Continuous testing with exploit/explore 00:00:00
    • Descriptive example 00:00:00
    • Content-based filtering 00:00:00
    • Content analyzer 00:00:00
    • Extracting metadata from descriptions 00:00:00
    • Finding important words with TF-IDF 00:00:00
    • Topic modeling using the LDA 00:00:00
    • Finding similar content 00:00:00
    • Creating the user profile 00:00:00
    • Evaluation of the content-based recommender 00:00:00
    • Pros and cons of content-based filtering 00:00:00
    • Sometimes it’s good to reduce the amount of data 00:00:00
    • Example of what you want to solve 00:00:00
    • A whiff of linear algebra 00:00:00
    • Matrix 00:00:00
    • What’s factorization? 00:00:00
    • Constructing the factorization using SVD 00:00:00
    • Adding a new user by folding in 00:00:00
    • How to do recommendations with SVD 00:00:00
    • Baseline predictors 00:00:00
    • Temporal dynamic 00:00:00
    • Constructing the factorization using Funk SVD 00:00:00
    • Root Mean Squared Error 00:00:00
    • Gradient descent 00:00:00
    • Stochastic gradient descent 00:00:00
    • And finally, to the factorization 00:00:00
    • Adding biases 00:00:00
    • How to start and when to stop 00:00:00
    • Doing recommendations with Funk SVD 00:00:00
    • User vector 00:00:00
    • Funk SVD implementation in MovieGEEKs 00:00:00
    • What to do with outliers 00:00:00
    • Keeping the model up to date 00:00:00
    • Faster implementation 00:00:00
    • Explicit vs. implicit data 00:00:00
    • Evaluation 00:00:00
    • Levers to fiddle with for Funk SVD 00:00:00
    • The confused world of hybrids 00:00:00
    • The monolithic 00:00:00
    • Mixing content-based features with behavioral data to improve collaborative filtering recommenders 00:00:00
    • Mixed hybrid recommender 00:00:00
    • The ensemble 00:00:00
    • Switched ensemble recommender 00:00:00
    • Weighted ensemble recommender 00:00:00
    • Linear regression 00:00:00
    • Feature-weighted linear stacking (FWLS) 00:00:00
    • Meta features: Weights as functions 00:00:00
    • The algorithm 00:00:00
    • Implementation 00:00:00
    • Learning to rank an example at Foursquare 00:00:00
    • Re-ranking 00:00:00
    • What’s learning to rank again? 00:00:00
    • The three types of LTR algorithms 00:00:00
    • Bayesian Personalized Ranking 00:00:00
    • Ranking with BPR 00:00:00
    • Math magic (advanced wizardry) 00:00:00
    • The BPR algorithm 00:00:00
    • BPR with matrix factorization 00:00:00
    • Implementation of BPR 00:00:00
    • Doing the recommendations 00:00:00
    • Evaluation 00:00:00
    • Levers to fiddle with for BPR 00:00:00
    • Algorithms 00:00:00
    • Context 00:00:00
    • Human-computer interactions 00:00:00
    • Choosing a good architecture 00:00:00
    • What’s the future of recommender systems? 00:00:00
    • User profiles 00:00:00
    • context 00:00:00
    • Algorithms 00:00:00
    • Privacy 00:00:00
    • Architecture 00:00:00
    • Surprising recommendations 00:00:00

    Course Reviews

    No Reviews found for this course.

    © 2021 Ernesto.  All rights reserved.  
    X