Lucas Gomez Tobon

I hold a Master’s degree in Economics from Universidad de los Andes. My thesis, titled “Perreology: Leveraging Music in Statistical Analysis”, utilized Convolutional Neural Networks (CNNs) to derive vector representations of music. This methodology informed the development of two applications. The first predicts the likelihood of a song ranking among the top 50 tracks in Colombia with a recall of 69%. The second application shows that the musical embeddings of the monthly top 50 songs in Colombia can explain up to 82% of the variability in the Consumer Confidence Index.

A demo of the predictive application is available here.

I currently serve as an Adjunct Professor at Universidad de los Andes, offering courses on machine learning and game theory. Additionally, I hold the position of Deputy Director of the Data Mining Department at Quantil.

news

Oct 31, 2023	Mentored an educational intervention in a Mexican indigenous community as part of the REDNACECYT-MD4SG Summer of Science program
Oct 9, 2023	The research paper titled ‘Modelling Underreported Spatio-Temporal Crime Events’ has been accepted for presentation at the XV Latin American Congress of Statistical Societies (CLATSE), where I will be the speaker.
Sep 22, 2023	Launched the tidyX package to facilitate text preprocessing for running NLP models.
Jul 26, 2023	Juan Carlos Echeverry, former Minister of Finance and former President of Ecopetrol, discussed our research 'Modelling Underreported Spatio-Temporal Crime Events' on Caracol Radio

selected publications

2023

Modelling underreported spatio-temporal crime events

Álvaro J. Riascos Villegas, Jose Sebastian Ñungo, Lucas Gómez Tobón, and 2 more authors

PLOS ONE, 2023

Abs Bib HTML TALK PDF

Crime observations are one of the principal inputs used by governments for designing citizens’ security strategies. However, crime measurements are obscured by underreporting biases, resulting in the so-called “dark figure of crime”. This work studies the possibility of recovering “true” crime and underreported incident rates over time using sequentially available daily data. For this, a novel underreporting model of spatiotemporal events based on the combinatorial multi-armed bandit framework was proposed. Through extensive simulations, the proposed methodology was validated for identifying the fundamental parameters of the proposed model: the “true” rates of incidence and underreporting of events. Once the proposed model was validated, crime data from a large city, Bogotá (Colombia), was used to estimate the “true” crime and underreporting rates. Our results suggest that this methodology could be used to rapidly estimate the underreporting rates of spatiotemporal events, which is a critical problem in public policy design.
@article{Modelling-underreported-crime, title = {Modelling underreported spatio-temporal crime events}, volume = {18}, doi = {10.1371/journal.pone.0287776}, number = {7}, journal = {PLOS ONE}, author = {Riascos Villegas, Álvaro J. and Ñungo, Jose Sebastian and Gómez Tobón, Lucas and Dulce Rubio, Mateo and Gómez, Francisco}, year = {2023}, dimensions = {true}, talk = {<iframe width="560" height="315" src="https://www.youtube.com/embed/ae74kxqFQVY?si=25mVurC1VKV4ye0I" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen></iframe>} }

2022

Perreology: Leveraging Music in Statistical Analysis

Lucas Gómez Tobón

2022

HTML TALK