📊 Data Science & AI Portfolio

---

👨‍💻 About Me

Hi, I’m Bruno Reis, a software developer with 15+ years of experience, currently pursuing a Bachelor’s Degree in Data Science & Artificial Intelligence at Instituto Infnet, Brazil.

This repository showcases my learning journey in data analysis, statistics, SQL, and machine learning foundations, combining my background in programming with a new focus on data-driven problem solving.

---

📂 Featured Projects

🏘️ Data Quality Assessment - King County (TP2)

Description: Comprehensive data quality diagnosis of the kc_house_data.csv dataset. The project focuses on distinguishing between actual data errors and natural outliers within the real estate market.

• Key Techniques: Comparative statistical outlier detection (IQR, MAD, Z-Score), referential integrity verification, and temporal logical consistency validation.

• Stack: Python, Pandas, NumPy, Seaborn.

• Highlight: Demonstrated that the MAD method is ~9x more sensitive than Z-Score for long-tail data distributions (real estate).

• 📎 Project README

---

🏋️‍♂️ Data Analysis with Python – Gym Members

• Description: Exploratory Data Analysis (EDA) of a fictional gym membership dataset (1,000 members).
• Tools: Python, Pandas.
• Highlights:
  • Descriptive statistics & distribution analysis.
  • Behavior analysis (training days, visits, group classes, services).
  • Key insights: avg. 2.68 training days/week, 36 favorite drinks, most popular class BodyPump.
• 📎 Project README

---

🎓 Student Performance Analysis with SQL

• Description: SQL-based analysis of student performance across Math, Reading, and Writing.
• Tools: SQL, Matplotlib (for visualizations).
• Highlights:
  • Score segmentation & demographic analysis.
  • Correlation between Math, Reading & Writing.
  • Test preparation courses improve performance across all subjects.
• 📎 Project README

---

🎵 Spotify Tracks Dataset – Data Analysis

• Description: Analysis of Spotify music tracks dataset from Kaggle.
• Dataset: Spotify Tracks Dataset
• Tools: Python, Pandas, Matplotlib, Seaborn.
• Highlights:
  • Temporal insights: analysis of release years and popularity over time.
  • Genre-based comparisons (distribution and top genres).
  • Visualization of continuous variables (e.g., danceability, energy).
  • Key findings on how musical features vary by genre and popularity.
• 📎 Project README

---

🏬 Superstore Sales Analysis with SQL

• Description: Sales and profitability analysis of the Superstore dataset using SQL queries.
• Tools: SQL (PostgreSQL), Pandas (for result handling), Matplotlib (for charts).
• Highlights:
  • Time-based sales trends with DATE_TRUNC().
  • Regional sales ranking and performance breakdown.
  • Customer purchase history tracking with window functions.
  • Top 5 customers with highest profit growth (using LAG() and month-over-month comparison).
• 📎 Project README

---

🛠 Tech Stack

Area	Tools & Technologies
Programming	Python (Pandas, Matplotlib, Seaborn), SQL
Data Handling	Data cleaning, exploratory data analysis, descriptive statistics
Visualization	Matplotlib, Seaborn
Soft Skills	Analytical thinking, problem-solving, data storytelling
Background	15+ years in software development

---

🚀 Next Steps

• Expand with projects in Machine Learning and Neural Networks.
• Apply predictive modeling to datasets (e.g., customer churn, education outcomes).
• Build interactive dashboards and advanced data visualizations.

---

✍️ Author: Bruno Reis
📅 Portfolio started in 2025