Projects

BigMartSales Data Analysis

Notebook analyzing the analyticsvidhya dataset. It's also available on Github.

Skills used in the analysis:

Initial Data Analysis
Exploratory Data Analysis
Data Cleaning
Tableau
Linear Regression Model Interpretation and Assumptions

PUBG Finish Placement Prediction using LightGBM

Notebook of the Kaggle competition "PUBG Finish Placement Prediction (Kernels Only)". The idea is to predict the final position of the players in this popular battle royale game based on their match data, such as how many heals they used or how far they walked. It's also available on Github and Kaggle.

Skills used in the process:

Exploratory data analysis.
Feature engineering (Feature creation).
Outlier analysis.
Gradient boosting with LightGBM.

Connectionist Bench (Sonar, Mines vs. Rocks) Classification Problem

Notebook training an ANN using UCI dataset as part of an I+D homework for data mining class. It's also available on Github.

Skills used in the process:

Artificial Neural Network (ANN) design
Hyperparameter choosing for classification task with ANN

Titanic Prediction Using Ensamble Learning

Notebook of the Kaggle's competition "Titanic: Machine Learning from Disaster". It's also available on Github and Kaggle.

Skills used in the final solution:

Feature engineering (missing values, feature extraction)
Ensamble learning (hard voting classifier implementation)
Modeling and hyperparameter tuning (XGBoost, Random Forest, Support Vector Machines)
Cross-validation test implementation
Tableau