The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
I developed a machine learning classification project to predict student exam performance using demographic, academic, and behavioral factors. The project included data exploration and preprocessing, feature analysis, model development, and evaluation of multiple classification algorithms.
I compared four machine learning approaches—Logistic Regression, Decision Tree, Random Forest, and Gradient Boosting—using accuracy, precision, recall, F1 score, and ROC-AUC to evaluate performance.
Gradient Boosting produced the strongest overall results, achieving 96.50% accuracy, a 98.21% F1 score, and a 93.73% ROC-AUC.
The project demonstrates my ability to take a structured dataset through the machine learning workflow, compare competing models, evaluate results using multiple performance metrics, and translate model outputs into meaningful insights about the factors associated with student performance.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
I just finished building my own open-source farm bot called Farmbt 🌱
It's a team of 13 AI agents. I asked them one question, and they came back with a ranked list of opportunities nobody has fully cracked yet.
You describe a problem area in plain words, and the agents take it from there:
🔎 Researchers search the web in parallel and pull out findings, each with its source
✅ A verifier re-opens the pages and checks every claim against the real text
🗳️ A gap hunter runs three independent passes, and only gaps that show up repeatedly survive
🥊 A critic tries to prove each gap is already solved
🛠️ Solver agents design fixes, check them against existing products and papers, and write an experiment plan
The hard part wasn't getting AI to produce answers. It was making it not make things up. So every claim is graded by evidence, confidence ratings are capped by what the source actually supports, and everything is saved, so a run that hits a rate limit picks up where it stopped.
It runs on free-tier models today and gets better with every model upgrade, with no code changes.
First test: crop-disease detection apps for smallholder farmers. 39 sources, 16 findings and 5 opportunities, ending in a concrete first experiment to run.
Open source: github.com/Samthnalyst1234/Farmbt
If you need an AI pipeline that's reliable enough to trust, for research, automation or anything in between, let's talk 👋
The step where a verifier re opens every page is a solid guard against hallucinations, and persisting state lets the run survive rate limits without missing a beat
WhatsApp AI assistant for an energy company's employees. It answers questions from company documents (RAG with Azure OpenAI and Cognitive Search), adds source links, verifies who is asking before answering and understands voice messages. Python/Flask, Twilio, PostgreSQL, Docker.
Built a Customer Churn Prediction project using Python, Pandas, NumPy, and Scikit-learn. I cleaned and preprocessed the data, performed exploratory analysis, trained classification models, and evaluated performance using metrics such as Accuracy, Precision, Recall, F1-score, Confusion Matrix, and ROC analysis.
The goal was to identify customers at risk of leaving and uncover patterns that could help improve retention strategies.
