Calss 2: Spark ALS

---

What you should do before the class

1. Please read the below materials

Spark ALS sample code

Spark CFMF documentation

Matrix factorization techniques for recommender systems

Algorithmic Acceleration of Parallel ALS for Collaborative Filtering: Speeding up Distributed Big Data Recommendation in Spark

Large-Scale Parallel Collaborative Filtering for the Netflix Prize

Collaborative filtering for implicit feedback datasets

2. Please prepare the answers to the questions in the next session. Please use slides to convey your ideas when answering the questions. you can find the slides template at the home page

---

Please present the answer to the below questions

1. How is ALS differ from the SVD/SGD/common CFMF? (10 minutes)
2. What's the difference between implicit and explicit data? How is Spark ALS treating them differently? (10 minutes)
3. What is cold start problem in recommendation engine? How can you deal with it? (10 minutes)
4. what are the major functions of the spark ALS API? (10 minutes)
5. One noteworthy point about spark ALS not included in the above questions. (5 minutes x 4)

---

Spark installation

Install on mac

Install on windows

Install on linux

Install on Azure Notebook

Set up spark in jupyter notebook

pyspark in jupyter

Create conda environment kernel in jupyter notebook

Install Spark in Docker Container

For people who have difficulties installing spark on windows, please try to see whether this can solve your problem.

Quick-start Apache Spark Environment Using Docker Containers

---

Additional question

1. Why SGD compared to GD?