Data Scientist · Sportska analitika · Valensija

Pretvaram podatke u jasne odluke

Ja sam most između analize podataka i vrhunskog sporta: pretvaram metrike performansi u strategije koje prave razliku na terenu.

Pogledaj projekte Kontakt

Python
R
SQL
Power BI
Machine Learning

O meni

Podaci sa kontekstom staze, puta i takmičenja

Ja sam Mario Álvarez, student Data Science na UPV-u sa stvarnim iskustvom u profesionalnom biciklizmu. Kombinujem analizu podataka sa vrhunskim sportskim performansama, kako bih svaku metriku pretvorio u odluku koja poboljšava rezultate.

Pogledaj više

UPV

Obrazovanje

Osnovne studije Data Science, Universitat Politècnica de València.

Elita

Iskustvo

Stvarni rad unutar profesionalnog biciklističkog tima.

Data + Sport

Pristup

Analitika primenjena na poboljšanje sportskih performansi.

Iskustvo

Dva sveta, isti jezik: performanse

Takmičio sam se na najvišem nivou i danas primenjujem isti kriterijum na podatke: razumem šta broj znači za sportistu jer sam bio taj sportista.

Pogledaj više

Rad

Projekti

Automatski izbor mojih javnih repozitorijuma na GitHub-u.

Specijalizacije

HTML
Power BI
Python
R
SQL

VIS

HTML

This repository contains coursework, practical exercises, and study materials for a university data visualization course. It includes solved programming tasks using R and Python libraries like ggplot2 and Plotly, alongside practice dashboards built in Tableau.

Pogledajte na GitHub-u

EVALUACION-DE-MODELOS-

HTML

This repository contains practical assignments for a university model evaluation course, covering topics such as cost-sensitive classification, fairness metrics on the COMPAS dataset, and model interpretability using partial dependence plots. The exercises and analyses are implemented using Python and R across Jupyter Notebooks and R Markdown reports.

Pogledajte na GitHub-u

TRABAJO_MDP

An academic data analysis project in R that explores and characterizes a physicochemical crystal dataset. It covers data cleaning, preprocessing, and dimensionality reduction through Principal Component Analysis (PCA) using packages such as FactoMineR, factoextra, and ggplot2.

Pogledajte na GitHub-u

TRABAJO_CORREDORES

Python

A Streamlit web application that helps manage cycling teams by scraping race and rider data from ProCyclingStats and FirstCycling to analyze upcoming events and cyclist specialties. It is built with Python using BeautifulSoup, Selenium, Pandas, and the Gemini API to process data and extract UCI points regulations from PDFs.

Pogledajte na GitHub-u

COFIDIS

Python

Developed as an academic project, this tool scrapes UCI rankings and race data from ProCyclingStats to calculate race difficulty scores tailored to the Cofidis cycling team's calendar. It is built with Python, using Selenium and BeautifulSoup for web scraping, Pandas for data processing, and Streamlit for the interactive interface.

Pogledajte na GitHub-u