See how we’ve applied AI, machine learning, and data analysis to solve business problems and deliver measurable gains, whether in efficiency, revenue, or customer experience
These results demonstrate our expertise in driving impactful outcomes for businesses
Explore Our Case Studies
IMDb Movie Reviews Sentiment Analysis with NLP (Python)
In this project, we developed several machine learning models (e.g. Logistic Regression, Naïve Bayes) to classify movie reviews as positive or negative using the IMDb Dataset of Movie Reviews.
Our workflow included data preprocessing, visualization, feature selection, model building and tuning. We concluded by visualizing feature contributions and explaining predictions for logistic regression, the best-performing model.
Sentiment analysis in NLP offers several benefits, such as understanding customers, improving service, and staying competitive. In filmmaking, analyzing IMDb reviews provides actionable insights to improve content, refine marketing, and understand audience preferences.
A/B Testing for LunarTech Homepage's CTA Button (Python)
In this project, we conducted an A/B test for LunarTech using proxy data similar to the company’s real data. LunarTech is a platform offering courses, bootcamps, and career support to help students land their ideal data role.
We tested two versions of LunarTech homepage’s CTA button with a control and experimental group, aiming to identify which version performs better based on the click-through rate (CTR) metric. This helped determine whether to implement the new button.
For LunarTech, CTR is important as it reflects user engagement and assesses the effectiveness of CTA buttons in driving sign-ups and conversions.
365 Data Science Subscription Purchase Prediction (Python)
In this project, we analyzed 365 Data Science student engagement metrics to train machine learning models (e.g. Logistic Regression, K-Nearest Neighbors) to predict whether students would upgrade their free plan to a paid one.
This classification task involved a heavily imbalanced dataset, as most 365 Data Science students retained their free plan. However, addressing this imbalance was not required to achieve successful results.
By identifying potential customers, businesses can optimize targeted advertising and exclusive offers, ensuring efficient budget allocation and increased revenue.
365 Data Science Customer Segmentation in Marketing (Python)
In this project, we analyzed real customer data from an onboarding survey for the 365 Data Science platform to perform customer segmentation. This is critical for businesses to understand customer behavior and develop personalized marketing strategies.
The process involved data cleaning, exploration, feature engineering, clustering algorithm implementation, and result interpretation. We applied two common clustering techniques: K-Means and Hierarchical Clustering.
By segmenting a sample representing the entire population, we identified patterns in acquisition channels and geographic influences. With these insights, we collaborated with the 365 Data Science marketing team to implement data-driven strategies.
365 Data Science Customer Engagement (SQL & Tableau)
In this project, we built a three-page Tableau dashboard with key metrics and visualizations to track student engagement with the 365 Data Science platform from January 1 to October 20, 2022.
We retrieved data from the 365 Data Science database using MySQL, which includes real-life records on students, purchases, and courses. Finally, we answered key questions to help the company better understand customer registration, engagement, and demographic patterns.
This helps refine marketing strategies, tailor course offerings, and boost user retention, ultimately driving business growth.