Yelp User Recommendation
FireWorks
Ying Chen, Yadnesh Gotey, Thea Huang, Xiaoqiu Yu, Chloe Zhan
Goal
We have found the Best Match restaurant for Yelp users
based on their previous activities.
01
02
04
03
CONTENTS
Live Demo
How it works!
Google Analytics
SAS Overview
DEMO
1-1899
6
CF is a technique used by recommender systems.
A method of making automatic predictions (filtering) about the interests of a user by collecting preferences or taste information from many users (collaborating).
Collaborative Filtering
How it Works!
Dataset
User_id | User_name | Yelping_since | User_averageStars | isElite | userCountLV |
U 1 | Michael | 2008-04 | 3.82 | Elite | 18 |
Business_id | Rest_name | Rest_stars | Rest_ReviewCount | Full_address | Category |
R 393 | Rocco’s NY Pizzeria | NV | 238 | 10860 W Charleston Blvd Las Vegas, NV 89135 | Pizza |
Review_star |
4 |
User Info
User
Restaurants Info
Restaurants
Review
8
STEP 1
STEP 3
STEP 2
STEP 4
STEP 5
Matrix
Predict Ratings
Similarity
Find the top 6 “Match”
Step by Step
Make an APP!
9
=>
Rating DataFrame
Rating Matrix
Step1. Matrix
10
Step2. Similarity
For a giver user pair, with their rating vectors, we could know their similarities.
11
Step3. Predict Rating
①
Similarity Matrix
Rating Matrix
Weighted Matrix
②
Predict Matrix
12
Step4. Best Match
Best Match!
13
Step5. Make the APP!
server.R
ui.R
14
15
Go further!
Shinyapps.io
Deploy your APPs!
16
More Info
THANK YOU!
?