Cristian Pulido
Logo PhD Student in Systems and Computer Engineering
Logo Logo Visiting Ph.D. Research Intern at Université Laval / CERVO
Logo Data Science Researcher

I am a mathematician with a strong background in applied modeling and computational analysis. My current work focuses on the integration of uncertainty-aware predictive models into decision-making processes for complex systems.

Through my doctoral research, I focus on integrating uncertainty into predictive models to improve decision-making in emergency response systems. I study how incomplete information and intervention-induced behavioral changes affect planning and resource allocation. At DataLab, I contribute to the development of data-driven tools for assessing health risks, optimizing interventions, and supporting public sector decision-making.

Security & Crime Prediction · Animal Health & Epidemiology · Decision Support Systems · Network Modeling

Curriculum Vitae

Education
  • Universidad Nacional de Colombia, Bogotá D.C., Colombia.
    Universidad Nacional de Colombia, Bogotá D.C., Colombia.
    Department of Systems and Industrial Engineering
    Ph.D. Student
    Aug. 2023 - present
  • Université Laval (Quebec, Canada) / CERVO Research Centre Université Laval (Quebec, Canada) / CERVO Research Centre
    Université Laval (Quebec, Canada) / CERVO Research Centre
    Visiting Ph.D. Research Intern (ELAP Mobility Stay)
    Jan. 12, 2026 – May 7, 2026
  • Universidad Nacional de Colombia, Bogotá D.C., Colombia.
    Universidad Nacional de Colombia, Bogotá D.C., Colombia.
    Department of Mathematics
    M.Sc. in Applied Mathematics
    Feb. 2018 - Nov. 2020
  • Universidad Distrital Francisco José de Caldas, Bogotá D.C., Colombia.
    Universidad Distrital Francisco José de Caldas, Bogotá D.C., Colombia.
    Faculty of Science and Education
    B.Sc. in Mathematics
    Feb. 2011 – Apr. 2017
Experience
  • DataLab – Universidad Nacional de Colombia
    DataLab – Universidad Nacional de Colombia
    Researcher in Predictive Analytics
    Dec. 2019 – Present
  • Universidad Antonio Nariño (UAN)
    Universidad Antonio Nariño (UAN)
    Adjunct Professor (Lecturer 2022–2024, Adjunct from 2024)
    Feb. 2022 – Dec. 2025
  • ViveLab Bogotá
    ViveLab Bogotá
    Researcher in Emerging Technologies
    Jun. 2019 – Sep. 2019
  • Fundación Universitaria de Ciencias de la Salud
    Fundación Universitaria de Ciencias de la Salud
    Research Assistant – Medical Imaging
    Mar. 2018 – Nov. 2019
Honors & Awards
  • Best Student Paper Award - 7th Int. Conference on Behavioral and Social Computing (BESC)
    2020
  • Emerging Leaders in the Americas Program (ELAP) Scholarship – Government of Canada. Selected for the ELAP scholarship funded by Global Affairs Canada to carry out a 4-month doctoral research stay at Université Laval (Quebec, Canada) from Jan. 12, 2026 to May 7, 2026.
    2025
Selected Publications (view all )
Quantifying Predictive Uncertainty in Crime Forecasting using Conformal Prediction
Quantifying Predictive Uncertainty in Crime Forecasting using Conformal Prediction

Cristian Pulido, Francisco Gómez

Accepted at 19° Congreso Colombiano de Computación (CCCesp 2025) SecurityUncertainty

This work presents a model-agnostic framework using Conformal Prediction to generate calibrated prediction intervals for crime hotspot forecasting. Applied to Chicago data, it reveals spatial variations in prediction confidence, highlighting the value of uncertainty in decision-making.

Quantifying Predictive Uncertainty in Crime Forecasting using Conformal Prediction

Cristian Pulido, Francisco Gómez

Accepted at 19° Congreso Colombiano de Computación (CCCesp 2025) SecurityUncertainty

This work presents a model-agnostic framework using Conformal Prediction to generate calibrated prediction intervals for crime hotspot forecasting. Applied to Chicago data, it reveals spatial variations in prediction confidence, highlighting the value of uncertainty in decision-making.

A Question-Driven AI Framework for Early-Stage Territorial Decision Support
A Question-Driven AI Framework for Early-Stage Territorial Decision Support

Cristian Pulido*, Oscar Sánchez*, Diana Aldana, Juan Diego Murcia, Sarah Coral, Francisco Gómez (* equal contribution)

Accepted at 19° Congreso Colombiano de Computación (CCCesp 2025) Spotlight Decision Support

This work introduces a modular AI framework combining guiding questions with LLM-based assistants to support early-stage territorial planning in Colombia. Two assistants—traditional RAG and agentic RAG—were evaluated, showing improved response quality and usefulness.

A Question-Driven AI Framework for Early-Stage Territorial Decision Support

Cristian Pulido*, Oscar Sánchez*, Diana Aldana, Juan Diego Murcia, Sarah Coral, Francisco Gómez (* equal contribution)

Accepted at 19° Congreso Colombiano de Computación (CCCesp 2025) Spotlight Decision Support

This work introduces a modular AI framework combining guiding questions with LLM-based assistants to support early-stage territorial planning in Colombia. Two assistants—traditional RAG and agentic RAG—were evaluated, showing improved response quality and usefulness.

Quantifying perception of security through social media and its relationship with crime
Quantifying perception of security through social media and its relationship with crime

Luisa Fernanda Chaparro, Cristian Pulido, Jorge Rudas, Jorge Victorino, Ana María Reyes, Camilo Estrada, Luz Ángela Narváez, Francisco Gómez

IEEE Access 2021 Social Sensing

This study proposes a machine learning model to quantify the Perception of Security on Twitter using sentiment analysis and crime-related filtering. Results reveal meaningful distinctions from traditional content-counting methods and dynamic links with specific crimes like robbery.

Quantifying perception of security through social media and its relationship with crime

Luisa Fernanda Chaparro, Cristian Pulido, Jorge Rudas, Jorge Victorino, Ana María Reyes, Camilo Estrada, Luz Ángela Narváez, Francisco Gómez

IEEE Access 2021 Social Sensing

This study proposes a machine learning model to quantify the Perception of Security on Twitter using sentiment analysis and crime-related filtering. Results reveal meaningful distinctions from traditional content-counting methods and dynamic links with specific crimes like robbery.

All publications