Guillem Escriba Molto's Projects
We will see how we can apply association rules to infer and analyze the basket and transactions in a supermarket.
Explore the intricacies of language use on Twitter with detailed bigram analysis across multiple languages and insights into data processing performance using cloud computing and AWS.
Here we have the code, models, and implementation of an app able to predict car prices based on several factors and its relevance explained thanks to XAI (Explanatory AI).
The main aplication of the Isolation Forest (non-linear ML algorithm) is to detect both, outliers and anomalies, in this notebook we will see how to apply it to the detection of abnormal tyroid of patients.
In this repository we can find a web application for Data Science using a github dataset to predict the salary of Software Engineers and Programmers with an EDA as well as ML techniques.
Dive into the world of 3D graphics with this project, featuring versatile mesh manipulation, camera control, and light positioning.
In this repository we'll apply an Exploratory Data Analysis and Feature engineering to analyze and extract insights of a electronic devices database.
A robust tool for parsing, language filtering, cloud uploading, and benchmarking Twitter data, developed for comprehensive tweet analysis.
A comprehensive guide to understanding and implementing Hidden Markov Models with practical applications in robot navigation and text improvement, using the pgmpy library in Python.
The following repository contains the required code (Matlab) and data to reproduce a predictive model of Taipei housing prices based on different variables such as size(m2), distance to stations (m) and number of convenience stores nearby. To do so we use Statistic techniques and linear regressors.
Explore the nuances of 3D lighting with Phong and Gouraud shading, light manipulation, and mesh control in this graphics lab.
A collection of interactive computer graphics tasks and tools, featuring customizable shapes, gradients, and filters.
In this repository we can see how to use the LSTMs analyzing and infering over sequential data as well as the potential of using RNNs in Cryptography.
Data mining is an essential part of the job of a Data Scientist, in this repository we can find some examples of those activities such as data preparation and analysis, association rules, near duplicates, data streams and forecasting.
The following repository contains the code of an implementation of a Mobile Net (CNN) to analyze the SVHN dataset using the minimum amount of parameters as well as preserving the accuracy.
Use of SQL to create a database based on the NBA and several queries to test it.
Using n-grams we will detect near duplicates in a real case scenario, a Twitter Dataset during the lockdown in Catalonia.
This repository contains different practices and implementations of network science and management of network's data. The practices are about how to work with graphs and networks to analyze its data and obtain the major insights of them.
Overview of a database management using SQL, queries, procedures and triggers to do several operations.
This repository contains projects related to Object Oriented Programming with Java. Essentially there are 3 main projects and its corresponding classes and tests. Two related to World map representation, and another one of an online bookstore.
Here we can find how to apply optimization techniques to improve the results of ML algorithms such as Linear Regression and PCA as well as using regularization and SVD for Image Denoising.
In this repository we can see how to optimize different functions by using multi-threads and synchronization as well as some basics of Operative Systems.
In this repository we will review how Network Science techniques can be applied to Spam detection. The main focus of the work is using PageRank to compute the spam gain of different webpages.
A detailed Business Plan developed for a university course, showcasing strategic planning in market analysis, financial modeling, and operational strategies. This repository serves as a documentation and template for academic and educational purposes in the field of business studies oriented towards ICT.
Advanced processing and analysis of language data from Mastodon streams, featuring real-time filtering, stateful operations, and hashtag tracking using DynamoDB and AWS.
The following repository contains a detailed explanation of how the Heap's Law affects to the use of a reservoir to processing a data stream, in this case, movie lines.
Using as documents a database of tweets, we have extracted all the information and created the corresponding index and ranking functions to create a functional search engine with web analytics.
For our Deep Learning final project, we enhanced a SRGAN network for Image Super Resolution, achieving a PSNR of 29.23 with 4x upscaling. We utilized a pre-trained VGG-19 for the Discriminator and developed a Weighted Dense Residual Network, adjusting weights in each connection to optimize results.
The main goal of this project is to select and target potential customers in order to offer them a new contract and choose the ones who are more likely to accept it. To do so, we will infer and predict over the dataset using ML techniques such as SVMs, Decision Trees, Ensamble Models,...