Toolkit 0 & 1

tl_2024_55_bg.cpg

@@ -0,0 +1 @@
+UTF-8

tl_2024_55_bg.prj

@@ -0,0 +1 @@
+GEOGCS["GCS_North_American_1983",DATUM["D_North_American_1983",SPHEROID["GRS_1980",6378137,298.257222101]],PRIMEM["Greenwich",0],UNIT["Degree",0.017453292519943295]]

tool0&1.py

@@ -0,0 +1,83 @@
+import geopandas as gpd
+import requests
+import pandas as pd
+import time
+import os
+
+# Load shapefile 
+block_groups = gpd.read_file("tl_2024_55_bg/tl_2024_55_bg.shp").to_crs(epsg=4326)
+
+# Reproject to a local projected CRS 
+block_groups_proj = block_groups.to_crs(epsg=3071)
+
+# Compute centroids 
+block_groups_proj['centroid'] = block_groups_proj.geometry.centroid
+
+# transform centroids back to WGS84 for yelp
+centroids_wgs84 = block_groups_proj.set_geometry('centroid').to_crs(epsg=4326)
+block_groups['lat'] = centroids_wgs84.geometry.y
+block_groups['lon'] = centroids_wgs84.geometry.x
+
+# Yelp API 
+YELP_API_KEY = "bDAon8ZKyViqMGdQp07QOu9X9trtnPOcJGpUxX02u1proTwYqM6DmJqOCzIrHac1e2A-WhrYHfuMQJ3fX7tNpY4icycxFX_VJEYqvCuZm4ptZdhGlYaPpKlXXLH8Z3Yx"
+headers = {"Authorization": f"Bearer {YELP_API_KEY}"}
+yelp_url = "https://api.yelp.com/v3/businesses/search"
+
+yelp_categories = [
+    "grocery",
+    "restaurant",
+    "pharmacy",
+    "clinic",
+    "hospital",
+    "school",
+    "gasstation",
+    "bank",
+    "hair",
+    "gym",
+    "coffee",
+    "daycare",
+    "laundry"
+]
+# Tool 1 (get services per block centroid) 
+services = []
+
+N = 50  # Number of block groups sampled
+
+for idx, row in block_groups.iterrows():
+    for category in yelp_categories:
+        print(f"🔍 GEOID {row['GEOID']} — {category}")
+
+        params = {
+            "term": category,
+            "latitude": row["lat"],
+            "longitude": row["lon"],
+            "radius": 1000,
+            "limit": 5
+        }
+        try:
+            response = requests.get(yelp_url, headers=headers, params=params)
+            if response.status_code == 200:
+                businesses = response.json().get("businesses", [])
+                for b in businesses:
+                    services.append({
+                        "GEOID": row["GEOID"],
+                        "category": category,
+                        "name": b["name"],
+                        "rating": b["rating"],
+                        "lat": b["coordinates"]["latitude"],
+                        "lon": b["coordinates"]["longitude"]
+                    })
+            else:
+                print(f"Yelp error {response.status_code} for GEOID {row['GEOID']}")
+        except Exception as e:
+            print(f"Request failed for GEOID {row['GEOID']}: {e}")
+
+        time.sleep(0.5) 
+
+    if idx >= N - 1:
+        break
+
+# Save to csv
+services_df = pd.DataFrame(services)
+services_df.to_csv("milwaukee_multiservice_yelp.csv", index=False)
+print(f"Done. Saved {len(services_df)} results to milwaukee_multiservice_yelp.csv.")

tool0&1plots.py

@@ -0,0 +1,33 @@
+import pandas as pd
+import matplotlib.pyplot as plt
+import seaborn as sns
+
+services_df = pd.read_csv("milwaukee_multiservice_yelp.csv")
+
+#Plot 1: Top 10 Most Common Service Categories
+
+category_counts = services_df['category'].value_counts().reset_index()
+category_counts.columns = ['category', 'count']
+
+plt.figure(figsize=(10, 6))
+sns.barplot(x='count', y='category', data=category_counts.head(10), palette="viridis")
+plt.title("Top 10 Most Common Service Categories in Milwaukee")
+plt.xlabel("Number of Services")
+plt.ylabel("Service Category")
+plt.tight_layout()
+plt.show()
+
+
+# Plot 2: Distribution of Services per Block Group (GEOID)
+
+geo_counts = services_df['GEOID'].value_counts().reset_index()
+geo_counts.columns = ['GEOID', 'service_count']
+
+plt.figure(figsize=(10, 6))
+sns.histplot(geo_counts['service_count'], bins=20, kde=True, color="skyblue")
+plt.title("Distribution of Services Across Block Groups")
+plt.xlabel("Number of Services per Block Group")
+plt.ylabel("Number of Block Groups")
+plt.tight_layout()
+plt.show()
+