Web Scraping Project: Visualize and Analyze Starbucks Coffee Store Amenities
Jhonasttan Regalado
11/12/2016
Problem: Multiple clicks are required to list Starbucks Coffee store amenities
Goal: Visualize the Starbucks Coffee Store locations and their amenities
Scope:
- Scrapy for web scraping stores and their amenities via Starbucks Coffee Store Locator website
- R for numerical / visual EDA
- CARTO for visual EDA
- MCA (Multiple Correspondence Analysis) for analyzing commonalities between the amenities
- Map layer for Shiny Citibike Analysis app
The Process:
Scrapy the magic sauce:
Processed data provides store geospatial information and associated amenities
- Regex (regular expressions) for feature engineering amenities data - used as filter criteria in CARTO map layers
- HTML wrapper for store location - CARTO renders fields in HTML format out of the box
- Duplicate rows!! Not an issue for CARTO which treats geospatial locations as unique
View of the CARTO ready data after processing in Scrapy and R is complete:
List of Amenities:
Five most common amenities:
- LB: LaBoulange
- WA: Oven-warmed Food
- LU: Lunch
- DR: Digital Rewards
- XO: Mobile Order and Pay
Five least common amenities:
- DT: Drive-Through
- EM: Starbucks Evenings
- WT: tbd - Walk-T
- FZ: Fizzio Handcrafted Sodas
- hrs24: Open 24 hours per day
MCA (Multiple Correspondence Analysis)
As with PCA and Correspondence Analysis, MCA allows us to analyze the systematic patterns of variations with categorical data.
List the eigen values
## eigenvalue percentage of variance
## dim 1 0.229307542 22.9307542
## dim 2 0.108699691 10.8699691
## dim 3 0.077774909 7.7774909
## dim 4 0.067799432 6.7799432
## dim 5 0.057700086 5.7700086
## dim 6 0.052447829 5.2447829
## dim 7 0.050270614 5.0270614
## dim 8 0.047859791 4.7859791
## dim 9 0.044506750 4.4506750
## dim 10 0.043505613 4.3505613
## dim 11 0.037468709 3.7468709
## dim 12 0.034816182 3.4816182
## dim 13 0.031310764 3.1310764
## dim 14 0.025410279 2.5410279
## dim 15 0.022428156 2.2428156
## dim 16 0.020807183 2.0807183
## dim 17 0.016384901 1.6384901
## dim 18 0.011369265 1.1369265
## dim 19 0.007996684 0.7996684
## dim 20 0.005336062 0.5336062
## dim 21 0.004379128 0.4379128
## dim 22 0.002420429 0.2420429
## cumulative percentage of variance
## dim 1 22.93075
## dim 2 33.80072
## dim 3 41.57821
## dim 4 48.35816
## dim 5 54.12817
## dim 6 59.37295
## dim 7 64.40001
## dim 8 69.18599
## dim 9 73.63666
## dim 10 77.98723
## dim 11 81.73410
## dim 12 85.21571
## dim 13 88.34679
## dim 14 90.88782
## dim 15 93.13063
## dim 16 95.21135
## dim 17 96.84984
## dim 18 97.98677
## dim 19 98.78644
## dim 20 99.32004
## dim 21 99.75796
## dim 22 100.00000View the relationships
Five most common amenities:
- LB: LaBoulange
- WA: Oven-warmed Food
- LU: Lunch
- DR: Digital Rewards
- XO: Mobile Order and Pay
MCA results shows amenities CD (Mobile Payment) and MX (Music Experience) within the cluster having the five most common amenities.
Five least common amenities:
- DT: Drive-Through
- EM: Starbucks Evenings
- WT: tbd - Walk-T
- FZ: Fizzio Handcrafted Sodas
- hrs24: Open 24 hours per day
Follow ups:
- Highlight the Most / Least common amenities when the user hovers over a store location in CARTO
- Create a Map Layer for Shiny Citibike Analysis app, allowing users to locate Starbucks Coffee stores based on amenities and proximity