📧 [email protected] | 💼 LinkedIn | 🚀 GitHub

Hello! I’m a Data Science Professional with a solid foundation of 2 years as a Business Analyst, where I’ve specialized in delivering impactful data insights to senior leadership and guiding data-driven decisions across dynamic industries. With expertise in SQL, Python, and Tableau, and an MS in Data Science, I excel in data preprocessing, visualization, and applying machine learning algorithms to turn complex datasets into actionable insights. I’m also skilled at managing cross-functional projects and effectively communicating data insights to non-technical stakeholders.

Actively looking for a full-time role as a Data Scientist / Data Analyst

Python, data ETL, Transformer, Streamlit github

• Led a team of 3 graduate students for a World Bank project, contributing to the digital transformation efforts for unbanked women in Nigeria.
• Developed a web application to automate the extraction and transformation of financial data into CSV format, enhancing data accessibility
• Boosted data extraction efficiency by 600% using advanced machine learning models to process over 200 unstructured documents.
• Designed and implemented an interactive visualization dashboard with Streamlit, showcasing financial insights through exploratory data analysis (EDA) and statistical testing.
• Presented findings to over 100 World Bank employees, receiving positive feedback from a non-technical audience.

Python, LLM, Transformer, Streamlit github

• Developed a web app using Streamlit for Artist-specific Lyric Generation & Lyric Analysis.
• Fine-tuned GPT-2 model for lyrics generation, and trained on the lyrics from top artists.
• Built a lyrics analyzer (Lyric Summarization, Keyword Extraction,...) using transformer models including BERT and BART.

Python, Binary Classification, Statistical Modeling github

• Applied machine learning algorithms to the CDC's Behavioral Risk Factor Surveillance System (BRFSS) dataset to predict heart disease likelihood.
• Focused on data cleaning, feature selection, and identifying key health indicators impacting heart disease risk.
• Handled imbalanced binary classification data with the SMOTE oversampling method.
• Developed statistical prediction models including MLP, Logistic Regression, XGBoost, and Random Forest with the best F1 macro score of 0.8.

Python, Time Series, SARIMA, LSTM github

• Developed a time-series model to predict traffic volume on I-94 in Minneapolis-St Paul using multivariate data, including weather and holidays.
• Handled the missing value in hourly traffic volume data with various methods such as interpolation and data resampling.
• Built ARIMA, SARIMA, OLS model, and LSTM model to predict the traffic volume.
• Evaluated models based on RMSE, analyzed seasonality, and provided insights to enhance urban planning and reduce congestion.

Tableau link

• Developed interactive Tableau dashboard to analyze U.S. road accidents, providing key insights on road safety by identifying critical factors from a dataset of 30,000 observations and 400 variables from NHTSA.

R, Binary Classification, Statistical Modeling github

• Conducted exploratory data analysis to identify key numerical and categorical variables influencing song popularity on R.
• Developed binary classification models including KNN, Tree-based, and Logistic Regression to classify popular songs on Spotify based on audio features with the best accuracy of 0.923 (KNN).

• Programming Languages: Python, R, SQL, C++
• Tools: PyTorch, Transformers, TensorFlow, Streamlit, Sklearn, Pandas, MySQL, Oracle SQL, MongoDB, Linux Shell Script, NLTK, spaCy
• Cloud Platforms: AWS (Cloud Web Builder, SageMaker), Google Cloud Platform (Vertex AI, Google BigQuery)
• Data Visualization: Tableau, Power BI