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
AI Stock Portfolio Selector — Random Forest on IBOVESPA 📈🤖
ML model that reads company fundamentals and classifies stocks as Buy / Hold / Sell — building a data-driven portfolio from Brazil's IBOVESPA index (77 companies, price data Dec 2012–Apr 2021).
8 classifiers benchmarked; Random Forest won on precision and was tuned with GridSearchCV. Result: the model's portfolio returned +4.6% in a quarter while the IBOVESPA fell ~3%.
Capstone project of the Data Science program at Digital House Brasil. Stack: Python, scikit-learn, pandas.
#machinelearning #datascience #python #fintech #randomforest
The key validation question for this kind of panel dataset is whether the benchmark used a time-aware split with fundamentals lagged to their actual publication dates. A random split can leak future information. Adding walk-forward results, a simple baseline, turnover and...
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
Trending
retrofuturism
Share your favorite designs from the past exploring the distant future. Or your own take on the aesthetic 🚀
halftone
Dots. Texture. Grain. Let's see your designs using the halftone technique.
Midjourney
Creatives are showing off what they're making with Midjourney.
colorgrade
Video editors sound off: How do you enhance the mood of your video in color grading?
Related posts
E-Commerce Intelligence & Predictive Analytics Platform
Built a Python-based analytics system to uncover patterns in retail product data. Used Decision Tree Classifier for product trend analysis, K-Means clustering for customer segmentation, and the Apriori algorithm for market basket analysis (association rule mining). Included data cleaning/preprocessing and visualization of key insights using Pandas, NumPy, and Scikit-learn.
I analyzed a large retail dataset to uncover “Hidden Gem” products items receiving highly positive customer feedback despite having relatively low visibility.
Using Python and Pandas, I cleaned and combined product, customer review, and metadata from multiple CSV and JSON sources. I then applied VADER sentiment analysis to thousands of customer reviews to measure customer sentiment beyond traditional star ratings.
To identify overlooked products, I developed a custom Hidden Gem Score combining average rating, customer sentiment, and review volume. The analysis identified approximately 2.8% of products as strong hidden gems, highlighting products with strong customer satisfaction but comparatively limited engagement.
Key skills: Python · Pandas · Data Cleaning · Sentiment Analysis · NLP · Data Visualization · Feature Engineering · Business Analytics
