visualization improvements

app_pages/visualization.py

@@ -20,7 +20,7 @@ def app():
     # Sidebar navigation for graph selection
     graph_choice = st.sidebar.radio(
         "Select a Graph to Display:",
-        ("Job Categories Distribution", "Contract Types Distribution", "Salary Distribution")
+        ("Job Categories Distribution", "Location vs Job Count", "Salary Distribution")
     )
 
     # Job Categories Bar Chart
@@ -30,7 +30,7 @@ def app():
             category_counts = df["Category"].value_counts().head(10).reset_index()
             category_counts.columns = ["Category", "Count"]
             plt.figure(figsize=(10, 6))
-            sns.barplot(data=category_counts, y="Category", x="Count", hue="Category", dodge=False, palette="viridis", legend=False)
+            sns.barplot(data=category_counts, y="Category", x="Count", palette="viridis")
             plt.title("Top 10 Job Categories")
             plt.xlabel("Number of Jobs")
             plt.ylabel("Category")
@@ -38,30 +38,32 @@ def app():
         else:
             st.warning("No 'Category' column found in the dataset.")
 
-    # Contract Types Bar Chart
-    elif graph_choice == "Contract Types Distribution":
-        st.subheader("Contract Types Distribution")
-        if "Primary Contract Type" in df.columns:
-            contract_counts = df["Primary Contract Type"].value_counts().reset_index()
-            contract_counts.columns = ["Contract Type", "Count"]
-            plt.figure(figsize=(8, 5))
-            sns.barplot(data=contract_counts, x="Contract Type", y="Count", hue="Contract Type", dodge=False, palette="crest", legend=False)
-            plt.title("Contract Types Distribution")
-            plt.xlabel("Contract Type")
-            plt.ylabel("Number of Jobs")
+    # Location vs Job Count Bar Chart
+    elif graph_choice == "Location vs Job Count":
+        st.subheader("Top Locations by Job Count")
+        if "Location" in df.columns:
+            location_counts = df["Location"].value_counts().head(10).reset_index()
+            location_counts.columns = ["Location", "Count"]
+            plt.figure(figsize=(10, 6))
+            sns.barplot(data=location_counts, y="Location", x="Count", palette="crest")
+            plt.title("Top 10 Locations by Job Count")
+            plt.xlabel("Number of Jobs")
+            plt.ylabel("Location")
             st.pyplot(plt)
         else:
-            st.warning("No contract types data available.")
+            st.warning("No 'Location' column found in the dataset.")
 
-    # Salary Histogram
+    # Salary Scatter Plot
     elif graph_choice == "Salary Distribution":
-        st.subheader("Salary Distribution")
+        st.subheader("Salary Scatter Plot")
         if "Salary (Numeric)" in df.columns:
-            plt.figure(figsize=(8, 5))
-            sns.histplot(df["Salary (Numeric)"].dropna(), bins=30, color='skyblue')
-            plt.title("Distribution of Salary")
-            plt.xlabel("Salary (EUR)")
-            plt.ylabel("Frequency")
+            filtered_df = df.dropna(subset=["Salary (Numeric)"])
+            filtered_df = filtered_df.reset_index()
+            plt.figure(figsize=(10, 6))
+            sns.scatterplot(data=filtered_df, x="index", y="Salary (Numeric)", alpha=0.7, color="blue")
+            plt.title("Salary Distribution Across Job Index")
+            plt.xlabel("Job Index")
+            plt.ylabel("Salary (EUR)")
             st.pyplot(plt)
         else:
-            st.warning("No numerical salary data available.")
+            st.warning("No 'Salary (Numeric)' column found in the dataset.")