This document discusses analyzing factors that affect restaurant ratings on Google+, develops a regression model to predict ratings, and recommends some options for where Matt Jayson could eat dinner. It finds that being a Japanese, American, or Chinese restaurant can impact ratings. It questions the fit of the regression model and proposes further analyzing the data by restaurant type and location. For dinner, it suggests a popular Indian restaurant or a hipster cafe.
1 of 7
More Related Content
Restaurants in Mountain View
1. QUESTION I WANTED TO
EXPLORE
1. How does the type of restaurant (Chinese,
American, Indian, etc.) affect the rating it gets on
Google+?
2. Where should Matt Jayson eat for dinner?
Thursday, December 12, 13
2. The regression equation is
RATING =
3.30 + 0.00023(Reviews) + 0.250(Cost Value)
+ 0.189(MexicanInd) + 0.214(JapaneseInd)
- 0.173(ChineseInd) - 0.0950(AmericanInd)
Thursday, December 12, 13
3. YES, THE FIT IS A LITTLE QUESTIONABLE
Thursday, December 12, 13
4. INTERESTING FINDINGS
Being a Japanese, American, Chinese style restaurant
increases your odds of possibly getting higher ratings on
Google+
Cost was not a signi?cant player in predicting reviews.
Thursday, December 12, 13
5. WHAT¡¯S NEXT?
Segment the data by restaurant type and analyze separately.
Does the type of restaurant affect the number of reviews (aka
# of customers)?
How does address affect rating and review count?
Thursday, December 12, 13
6. WHERE YOU SHOULD EAT:
1. Join the masses
Amber India Rest.
Cascal
2. Be a hipster
Sweet Leaf Cafe
Marya¡¯s Cafe
Thursday, December 12, 13