🦈 Shark Attack Analysis

Caique Fabris | Sara Iriarte Paz | Rubén | Susanna

🌊 Overview

The Shark Attack Analysis project was a team effort developed during the IronHack Data Analytics Bootcamp. Although shark attacks are rare, they can have serious consequences for beachgoers and marine enthusiasts. By studying these incidents, we can better understand the factors that contribute to shark attacks, leading to improved safety measures, increased awareness, and enhanced conservation efforts. This project involved cleaning and analyzing a dataset on shark attacks using Python and data wrangling techniques to uncover meaningful insights.

🔍 Approach

1. Initial Dataset Examination: Investigated patterns in shark attacks, such as higher frequencies in specific locations or during particular activities.
2. Data Cleaning:
   • Cleaned the dataframe by dropping unnamed columns and files that were not useful for data analysis, including PDFs, external links, videos, and other types of media.
   • Renamed columns, addressed typos, and standardized data expressions.
3. Exploratory Data Analysis (EDA): Conducted basic EDA to validate hypotheses and extract key insights after cleaning the dataset.

🛠️ Tools & Technologies

• Python: The primary programming language used.
• pandas: For data manipulation and cleaning.
• Jupyter Notebook: For coding and visualizations.
• Google Colab: Used for collaborative coding and sharing.

📁 Repository Structure

• Main File: shark_attack_team_final.py – This file contains the final analysis of the shark attack dataset.
• Extra Files:
  • shark_attack_value_counts.py – An initial file to extensively evaluate the dataframe and better understand its structure before starting the cleaning process and formulating hypotheses.
  • shark_attack_breeds.py – An additional analysis file that cleans the 'species' column in a different way, necessitating separate handling.
  • shark_attack_presentation.pdf - PDF file containing the project presentation along with graphics for a visual overview of the findings.

🔑 Key Insights

• Validated or disproved insights and hypotheses through EDA.
• Provided final analysis and conclusions on shark attack trends.

🔬 Hypotheses Explored

• Caique Fabris: Investigated the hypothesis that attacks are seasonal and may correlate with shark migration patterns.
• Sara Iriarte Paz: Analyzed the correlation between the size of the shark, the severity of injuries, and the location of the attack.
• Rubén: Explored the relationship between the activities of victims and their ages.
• Susana: Examined the proportional increase in shark attacks on males over the past years.

📦 Deliverables

• Cleaned dataset and code.
• Visualizations and insights derived from EDA.
• Project presentation.

✅ Project Status

This project is now complete. All data utilized in this project was retrieved from: Shark Attack File.

---

Feel free to dive into the code and explore the data! 🏖️🐋