Stella Sposito

Data Scientist

|

About

Stella's Image

My name defines a lot of who I am. I like to behave like a star in the solar system: with intensity, brightness, and authenticity. I learned to be a Scientist when I graduated in Biology, discovering my critical, researching, and questioning side. I learned to be a Data Scientist in my postgraduate studies, where I realized that I could use my scientist side to look at analyses with different eyes and obtain important information through data.

Projects

Sentiment Analysis API with LLM and Vector Search

This project implements a sentiment analysis pipeline using Retrieval-Augmented Generation (RAG) powered by Large Language Models (LLMs) and a vector database (ChromaDB).

Wine

More Info

Sentiment Analysis API with LLM and Vector Search

This project implements a sentiment analysis pipeline using Retrieval-Augmented Generation (RAG) powered by Large Language Models (LLMs) and a vector database (ChromaDB).

Predicting Customer Churn: A Data-Driven Approach

Analysis of factors influencing customer churn for a telecommunications company and identification of factors leading to customer retention.

Churn

More Info

Predicting Customer Churn: A Data-Driven Approach

Analysis of factors influencing customer churn for a telecommunications company and identification of factors leading to customer retention.

Credit Risk Analysis

This project applies machine learning techniques to predict the likelihood of customer default based on credit data.

Titanic

More Info

Credit Risk Analysis

This project applies machine learning techniques to predict the likelihood of customer default based on credit data.

A Data-Driven Exploration of iFood Customers

A detailed analysis using statistical and visual techniques to uncover patterns and insights in the food delivery market.

Ifood Analysis

More Info

A Data-Driven Exploration of iFood Customers

A detailed analysis using statistical and visual techniques to uncover patterns and insights in the food delivery market.

Sentiment Analysis of Airline Reviews

Classifying customer sentiment across airlines using NLP and Machine Learning models, with insights into the impact of reviews on each airline's Net Promoter Score (NPS).

Airlines

More Info

Sentiment Analysis of Airline Reviews

Classifying customer sentiment across airlines using NLP and Machine Learning models, with insights into the impact of reviews on each airline's Net Promoter Score (NPS).

Trend Analysis of ENEM Questions

A trend analysis of ENEM exam questions, visualizing word proximity to classify questions as physics, chemistry, or biology with Random Forest model.

Enem

More Info

Trend Analysis of ENEM Questions

A trend analysis of ENEM exam questions, visualizing word proximity to classify questions as physics, chemistry, or biology with Random Forest model.

Publications

Special Student in Algorithms and Data Structures: My Learning Journey

In this post, I share practical study tips that helped me balance work, study, and growth. Hope it inspires you to keep going, even when it’s tough!

Trade-off Bias-Variance

This analogy compares two students to bias and variance in machine learning. It explains how models, like students, need to balance learning from training data while generalizing to new challenges.

Understanding Parameters vs. Hyperparameters in Machine Learning

Parameters are learned during model training, while hyperparameters are set beforehand to guide the learning process. Just like an athlete needs both well-developed muscles and a solid training plan, a model needs tuned parameters and well-chosen hyperparameters to perform well.

Transforming Categorical Variables

Here, I discuss some common approaches to transform categorical variables into numerical ones, such as mapping strings to numbers, Label Encoding, and One Hot Encoding, along with practical examples. To check out more publications like these, visit my Linkedin.

Temporal resource partitioning and stochastic colonization explain the co-occurrence of gall-inducing insects in the super-host plant Copaifera langsdorffii Desf. (Fabaceae)

This was my first scientific paper published in a journal. Through this experience, I enhanced my skills in data visualization using R and creating statistical graphics.

Contact Me