Xavi Reverté

Holistic Data Scientist with an environmental science background and a passion for driving innovation through Machine Learning

I am seeking opportunities to leverage my expertise in data-driven decision-making across diverse domains

Python | SQL | R | HTML | CSS

September, 2023

“Influence Factors Analysis on Personal Happiness:
A Machine Learning Approach Using Multidimensional Data"

In this groundbreaking project, we explore factors influencing personal happiness using two datasets. Through clustering, statistical analysis, and supervised ML models, we identify similarities among 'Survey' respondents and examine how daily activities impact happiness in the 'Personal' dataset. This end-to-end project reveals surprising connections between actions and emotions, highlighting discrepancies between perceived and actual influences, offering a deep dive into human psychology and challenging conventional perceptions of happiness.

Python | LaTeX | Google Forms | K-Means | Decission Tree | Random Forest | Gradient Boosting | Extreme Gradient Boosting | Support Vector Machine | Linear Regression | Ensemble approach | End-to-end Project

May, 2024

“Navigating the Job Market: Automated LinkedIn Skill Analysis”

Developed an automated, reproductible, versatile and fault-tolerant system that scrapes LinkedIn to analyze the top skills required for desired job positions *Data Scientist*. Comprehensive analysis of the top skills studying their distribution across various dimensions.

Python | Web Scraping | selenium | scrapy | Natural Language Processing | re | nltk | Word-Cloud | Clustering

November, 2023

“Re-Source Collaboration:
Awareness of Rural Depopulation”

Achieved an appropriate project work teams classification by designing a survey and methodology for its segmentation. Employed K-means clustering to categorize respondents and predict participant group assignments based on their centroids.

Python | Google Forms | K-Means | Gradient Boosting | Effective Communication

June, 2023

“CIFAR-10 Image Classification”

Achieved an impressive accuracy rate of 94.72% in an image classification task through transfer learning with pretrained Convolutional Neural Network models, such as VGG19 and Resnet34, using an ensemble approach.

Python | VGG19 | ResNet34 | DenseNet121 | EfficientNet_b0 | CNN | Transfer Learning

July, 2021

“Aquatic Community Network Structure under Disturbance Effects”



Applied Principal Component Analysis to analyze changes in zooplankton composition, identifying three distinct sample groups. Employed multifactorial ANOVA to evaluate changes in its structure, revealing insights into La Pletera wetlands' zooplankton community.

RStudio | Statistics | PCA | Multifactorial ANOVA | Environmental Sciences | Conservation Strategy