911 Emergency Calls Data Analysis

911紧急呼叫数据分析

Tools: Python, Pandas, NumPy, Matplotlib, Seaborn

Final Project

NoteBook Link - Recommended to view in Google Colab.

This project analyzes 911 emergency call data from Kaggle to explore temporal and spatial patterns in emergency response. The dataset includes call descriptions, locations, timestamps, and emergency types (EMS, Fire, Traffic). By extracting time-based features (hour, weekday, month), the analysis uncovers trends in call volume over time. Visualizations such as count plots, line charts, and heatmaps highlight patterns by location, time, and reason. The project also applies grouping, aggregation, and clustering techniques to identify high-activity periods and areas.

Applied techniques

  • Datetime feature engineering: Extracted Hour, Month, Day_of_Week, and Date from timestamp data

  • Text preprocessing and parsing: Extracted Reason from the title field using string splitting

  • Aggregation and grouping: Used groupby() to compute call counts over time and by categorical variables

  • Data visualization: Created countplots, line plots, and regression plots to explore time-based trends

  • Heatmap and clustering: Built heatmaps and clustermaps to analyze high-frequency call periods across hours and weekdays/months