/CSE258-Recommender-Systems

Code for UCSD CSE 258 Web Mining and Recommender Systems

Primary LanguageJupyter NotebookMIT LicenseMIT

CSE258-Recommender-Systems

Code for UCSD CSE 258 Web Mining and Recommender Systems

Course Website: http://cseweb.ucsd.edu/classes/fa17/cse258-a/

HW1

Simple Regression and Classification about beer review data

http://cseweb.ucsd.edu/classes/fa17/cse258-a/files/homework1.pdf

Beer data: /data/beer_50000.zip

HW2

  • Classifier Evaluation: TP, FP, TN, FN, Precision, Recall and BER

  • Dimension Reduction: PCA

http://cseweb.ucsd.edu/classes/fa17/cse258-a/files/homework2.pdf

Beer data: /data/beer_50000.zip

HW3

Preview of Assignment 1

http://cseweb.ucsd.edu/classes/fa17/cse258-a/files/homework3.pdf

Data: https://www.kaggle.com/c/cse258-fa17-rating-prediction/data

Assignment 1

Using kNN Method to predict whether the user visited the item or not

Google Local Visit Prediction: https://www.kaggle.com/c/cse158-258-fa17-visit-prediction

Using Latent Factor Model to predict the rating score of given user-item pairs.

Google Local Rating Prediction: https://www.kaggle.com/c/cse258-fa17-rating-prediction

Assignment 2

Amazon Review Helpful Classification

Data: http://jmcauley.ucsd.edu/data/amazon/

Using Logistic Regression and SVM to predict the reviews' helpfulness rated by other user

Features include review time, review Readability and TF-IDF.