peter-seweiha.github.io

Peter Seweiha
Data Scientist & Business Improvement Specialist
LinkedIn
GitHub

Data Science Portfolio

Repository containing a portfolio of Data Science projects and blogs completed for academic, self-learning, and professional purposes presented in the form of Jupyter Notebooks.

My Toolbox

• Python: NumPy, Pandas, Seaborn, Matplotlib, Scipy, scikit-learn, TensorFlow, keras
• Data Processing: Exploratory Data Analysis, Data Visualisation, Feature Engineering, Feature Selection
• Machine Learning: Linear Regression, Logistic Regression, KNN, Random Forest, Gradient Boosting, Support Vector Machine (SVM), Naive Bayes, Principal Component Analysis (PCA), K-means, Artificial Neural Networks
• Power BI: Interactive Dashboards
• Minitab (SPC): Control Charts, Process Capability Study, Measurement System Analysis(MSA), ANOVA, Hypothesis Testing, Time Series Analysis

My Projects

1. Statistical Process Control (SPC) tools
   Here I’m writing some of the most useful statistical Six Sigma tools in python
   • Process Control Charts
     • Individuals and Moving Range Control Charts (I-MR) NoteBook
     • Xbar-R NoteBook
     • Xbar-S NoteBook
   • Process Capability Study NoteBook
   • Normality Test NoteBook
   • Histogram Comparison NoteBook
2. Forcast the no of daily customers for a restaurant chain (Time Series Analysis) NoteBook
   Here I compared the performance of an ARIMA based model, FaceBook prophet, and a traditional XGB regression. Imagine which performed the best!

My Blogs

• How to teach Artificial Intellegence to play Games NoteBook
  Computer learning how to play a game of Frozen Lake using Reinforcement Learning

• Intro. to solving Differential Equations NoteBook
  Quick run through Differential Equations with a few simple examples using code

• My approach to Kaggle’s Titanic dataset NoteBook
  A solution focused on feature engineering and a hyperparameter tuning (top 8% on Kaggle’s Leaderboard!)

• Intro. to Natural Language Processing (NLP) NoteBook
  Classify a restaurant’s customer text reviews into positive/negative

• Intro. to Artificial Neural Networks (ANN) NoteBook
  Bank customers’ churn prediction

• Intro. to Unsupervised Learning - K.means clustering NoteBook
  Categorising Shopping mall customers based on their income and spending habits

• Intro. to Unsupervised Learning - Dimensionality Reduction (PCA) NoteBook
  Predicting if a customer group will like a new wine based on product parameters