Projects
A selection of projects that I'm happy to share
Churn Effect by Average Annual Temperature
![Churn Effect by Average Annual Temperature](/images/projects/noaa.png)
Tableau Dashboard to explore the churn effect by the average annual temperature in a fictitious company.
Bank Maven BI Analysis - Tableau Dashboard
![Bank Maven BI Analysis - Tableau Dashboard](/images/projects/maven.png)
BI analysis done with Tableau Desktop Professional.
Natural Language Processing with Transformers
![Natural Language Processing with Transformers](/images/projects/BERT.png)
NLP application using a BERT transformer model.
Appointment Scheduler
![Appointment Scheduler](/images/projects/Appointment Scheduler.png)
Done using JavaFx, the purpose of the application is to assist a global consulting organization in managing the appointments of their customers with the firm's contacts. The application allows to add, edit and delete customers and appointments, and has different kind of alerts that prevent errors such as scheduling an appointment in a weekend. Click above for more information and a video demo.
Integrating Spark and Google Colab
![Integrating Spark and Google Colab](/images/projects/medium.png)
This was a project regarding an interesting dataset of cancelled flights. My end goal was to create a step-by-step guide to process, train, execute and evaluate the dataset combining Spark and Google Colab.
Predicting user’s demographics
![Predicting user’s demographics](/images/projects/Predicting Users Demographics.png)
Using a dataset from China’s largest third-party mobile data platform, TalkingData, the objective was to predict user’s demographics characteristics which in return gives us valuable information that can help millions of developers and brand advertisers pursue data-driven marketing efforts which are relevant totheir users and catered to their preferences.
Determining Telco Churning
![Determining Telco Churning](/images/projects/Churn.png)
Training and testing data from 7042 customers from a Telco with eight different supervised models, the code predicted customer churn with a 78 % accuracy. It also determined the most important features that influence churn.
Author detection
![Author detection](/images/projects/Author detection.png)
The goal of this project is to train a model with texts from different authors,and determine the accuracy of the model when predicting from which author a new text belongs.