Personal blog about Data and AI related topics

API REST for detecting if a text correspond to a fake news or to a legitimate one. It's served using Flask and uses a fine-tuned BERT model.

ML web app that can be deployed through Docker to analyse the sentiment of tweets.

Forecasting of the dataset located at https://www.kaggle.com/datasets/sumanthvrao/daily-climate-time-series-data

Exercise to use simple ML models to perform sentiment analysis and show they are still relevant to the task

Code to prevent credit card fraud based on the data located at https://www.kaggle.com/datasets/mlg-ulb/creditcardfraud

Training of a machine translation system from English to German from the ground up. It uses the dataset contained at https://github.com/multi30k/dataset/tree/master/data/task1/raw

Implicit Collaborative Filtering trained on the RetailRocket dataset based on the ideas from arXiv:2009.08950

MovieLens 100K dataset exploration and recommender system building

Resolution of the classification problem stated at https://www.kaggle.com/uysimty/keras-cnn-dog-or-cat-classification using Keras and ConvNets

Data Discovery and Modeling of the set located at https://www.kaggle.com/arashnic/hr-analytics-job-change-of-data-scientists

General project to learn about Darts module in Python

Quick and dirty forecasting of the time series datasets located at https://www.kaggle.com/shenba/time-series-datasets?select=sales-of-shampoo-over-a-three-ye.csv

Attempt at the Kaggle competition based on https://www.kaggle.com/c/house-prices-advanced-regression-techniques/overview

Analysis of the information located at https://www.kaggle.com/crowdflower/twitter-airline-sentiment combined with the official dataset at https://www.transtats.bts.gov/Fields.asp?gnoyr_VQ=FGJ

Analysis made to discover patterns in baby names on the NY state. Data to be located at https://health.data.ny.gov/Health/Baby-Names-Beginning-2007/jxy9-yhdk

Attempt at the Kaggle competition https://www.kaggle.com/c/titanic/ based on the famous Titanic dataset

Project based on getting data from public sources and combine it into a single DDBB to explore

Code in R to describe and analyse employee attrition and predict it using statistical models.

Cleaning, preparation and analysis of the data from https://archive.ics.uci.edu/ml/datasets/Automobile

Scrapes the website https://datosmacro.expansion.com/ to get a dataset describing the evolution of the 3 main indicators for the petroleum price between 2000-01 and 2020-10.