Welcome to our Google Colab notebook for analyzing occurrence data! In this notebook, we've conducted a comprehensive analysis of occurrence data, providing insights and visualizations to help you understand the dataset better.

Our notebook consists of several blocks of code, each serving a specific purpose. Here's what you can expect from each block:

This block analyses the dataset and produces a summary, as well as a series of tweets that can be adapted to post on social media

1. Top 10 Occurrence Types by Districts: This code block identifies and visualizes the top 10 occurrence types (natureza) in different districts. It generates a heatmap to show the frequency of these occurrence types, providing a clear representation of common incidents.

1.2 Top 10 Affected Counties: Here, we calculate and visualize the top 10 counties (concelhos) most affected by occurrences. This code block helps you understand which counties experience the highest incident rates.

1.2.1 Top 10 Affected Counties by most affected District: Here, we calculate and visualize the top 10 counties (concelhos) most affected by occurrences. This code block helps you understand which counties experience the highest incident rates.

1.2.2 **Top 10 Affected parrishes int the most affected county in the most affected district **: Here, we calculate and visualize the top 10 parrishes (freguesias) most affected by occurrences. This code block helps you understand which parrishes experience the highest incident rates in the most affected county in the most affected district

2. Incident Counts Bar Chart: This block creates a bar chart showing the number of occurrences by type. You can modify the number of types displayed in the chart.

3. Stacked Bar Chart: We display a stacked bar chart to illustrate the distribution of the top 10 incident types across different districts. This visualization offers insights into the geographic distribution of incidents.

4. Response Time and Closed Incidents: This code segment calculates and visualizes the response time for incidents with the status 'Arrival at TO' and identifies closed incidents.

5. Occurrence Duration Box Plot: We create a box plot to analyze the distribution of occurrence duration for incidents marked as 'Completion.'

6. Response Time Box Plot: This block generates a box plot to examine the distribution of response times for incidents labeled 'Arrival at TO.'

7. Occurrences by District Scatter Plot: In this code block, we produce a scatter plot to explore the temporal evolution of occurrences across different districts. Annotations are added to highlight specific points.

8. Occurrences by Natureza Scatter Plot: Similar to the previous scatter plot, this one visualizes the temporal evolution of occurrences by 'natureza' types.

To make the most of this notebook, follow these steps:

1. Ensure you read all instructions in the notebook and understand the license

2. Execute each code block sequentially by pressing the play button (▶️) or using the Shift+Enter keyboard shortcut.

3. Examine the visualizations and insights generated by each code block to better understand your occurrence data.

4. Feel free to modify parameters, such as date ranges or the number of displayed types, to customize the analysis to your needs.

5. Download any generated visualizations for sharing or further analysis.

Before using this notebook, make sure you have the following:

• A Google Colab environment set up.
• The necessary Python libraries installed (e.g., pandas, matplotlib, seaborn).

This notebook is provided under the MIT License. All derived products are availale under a CC BY-NC-SA 4.0 License

We hope this notebook helps you gain valuable insights from your occurrence data. If you have any questions or need further assistance, please don't hesitate to reach out.

Happy analyzing!

The VOST Portugal Team