foto perfil Juan David Martínez Gordillo

Juan David Martínez Gordillo

Bogotá, Colombia
Conferencista | Capacitador | Coach | Docente | Moderador | Panelista
¡Evalúame!

Juan David es economista bilingüe de la Universidad del Rosario en Colombia y cuenta con más de 8 años de experiencia en el sector público, multilaterales, sector privado y academia. Fue Coordinador de Proyectos Big Data en el Departamento Nacional de Planeación y trabajó como Consultor para el Programa de Naciones Unidas para el Desarrollo (PNUD). Recientemente, se desempeñó como Data Scientist (Científico de Datos) dirigiendo proyectos de data science en empresas de tecnología en Estados Unidos y Colombia en las industrias de Computer Software, Fintech y Edtech. Actualmente, Juan David es Senior Data Scientist en Microsoft y es Profesor en ... Ver más

0.0

Juan David Martínez Gordillo

Bogotá, Colombia

Conferencista | Capacitador | Coach | Docente | Moderador | Panelista


Juan David es economista bilingüe de la Universidad del Rosario en Colombia y cuenta con más de 8 años de experiencia en el sector público, multilaterales, sector privado y academia. Fue Coordinador de Proyectos Big Data en el Departamento Nacional de Planeación y trabajó como Consultor para el Programa de Naciones Unidas para el Desarrollo (PNUD). Recientemente, se desempeñó como Data Scientist (Científico de Datos) dirigiendo proyectos de data science en empresas de tecnología en Estados Unidos y Colombia en las industrias de Computer Software, Fintech y Edtech. Actualmente, Juan David es Senior Data Scientist en Microsoft y es Profesor en ... Ver más


imagen portada Ciencia de los datos aplicada a producto: Predicción y modelado de comportamientos del consumidor

Ciencia de los datos aplicada a producto: Predicción y modelado de comportamientos del consumidor

De 1 Hora a 2 Horas

Decisiones de producto que antes se basaban en intuición, experiencia o costosos estudios de mercad...

"Research results preview: Tweeting for Peace: Experimental Evidence from the 2016 Colombian Plebiscite". Presented at Colombian Big Data Society Conference. July 27, 2017

Computational Sentiment Analysis on Twitter with R webinar

Data science applied to product: Prediction and modeling of consumer behaviors

  • D

    DaCodes

    • Trayectoria de un Data Scientist
    • 2021
    • Conferencia impartida a los empleados de la empresa DaCodes que describió las lecciones y de una carrera en ciencia de datos en múltiples entornos industriales.
  • e

    eTraining.co

    • Ciencia de los Datos Aplicada a Producto: Predicción y Modelado de Comportamientos del Consumidor
    • 2021
  • U

    Universidad del Rosario

    • Ciencia de los datos aplicada a producto: Predicción y modelado de comportamientos del consumidor
    • 2020
    • Decisiones de producto que antes se basaban en intuición, experiencia o costosos estudios de mercado, ahora son guiadas por experimentos, algoritmos predictivos y modelos estadísticos. Esta charla busca hacer una introducción a cómo metodologías basadas en datos pueden ayudar a desarrollar mejores productos y anticipar los comportamientos de los clientes.
  • U

    Universidad del Rosario

    • El camino hacia una beca internacional
    • 2019
  • F

    Facultad de Economía - Universidad del Rosario

    • Ciencia de datos aplicada al análisis económico
    • 2018
    • Conferencia dirigida a estudiantes de pregrado y posgrado, del departamento de economía de la Universidad del Rosario, sobre cómo utilizar las habilidades y metodologías de la ciencia de datos en la profesión de economía. Además, se brindó orientación y posibles trayectorias profesionales a los estudiantes que buscan convertirse en profesionales de la ciencia de datos.
  • B

    Big Data Colombia

    • Trae tus Propios Proyectos de Datos
    • 2018
  • B

    Big Data Colombia

    • Research results preview: Tweeting for Peace: Experimental Evidence from the 2016 Colombian Plebiscite
    • 2017
  • E

    Escuela de Verano - Universidad del Rosario

    • Analítica de Texto
    • 2017
  • U

    Universidad del Rosario

    • Summit on Peacebuilding
    • 2017
    • “Tweeting for Peace: Experimental Evidence from the 2016 Colombian Plebiscite”. Presentado en el Simposio: Political Economy, Conflict and Peace - Summit on Peacebuilding. Agosto 24, 2017
  • C

    CORPOVISIONARIOS

    • Panel de expertos que buscan formular una metodología para la medición del ciberacoso en redes sociales a través de métodos de aprendizaje automático
    • 2017
  • B

    BD Guidance

    • Fundamentos de Data Science
    • 2016
  • B

    BD Guidance

    • Computational Sentiment Analysis on Twitter with R webinar
    • 2016
  • C

    Colombia 4.0

    • Análisis de Sentimiento en R
    • 2016
  • M

    Data Scientist

    • Microsoft
    • 2021 - Actual
  • U

    Profesor

    • Universidad del Rosario
    • 2017 - Actual
  • P

    Data Scientist

    • Platzi
    • 2020 - 2021
  • A

    Data Scientist

    • Alkanza
    • 2018 - 2019
  • P

    Data Scientist

    • Platzi
    • 2019 - 2019
  • D

    Coordinador de Proyectos Big Data

    • Departamento Nacional de Planeación
    • 2017 - 2018
  • P

    Consultor de Ciencia de Datos

    • PNUD
    • 2016 - 2017
  • D

    Master's in Data Science

    • Duke University
    • 2021
  • U

    Economista

    • Universidad del Rosario
    • 2016

Preventing rather than Punishing: An Early Warning Model of Malfeasance in Public Procurement

Abrir enlace

Is it possible to predict corruption and public inefficiency in public procurement? With the proliferation of e-procurement in the public sector, anti-corruption agencies and watchdog organizations in many countries currently have access to powerful sources of information. These may help anticipate which transactions become faulty and why. In this paper, we discuss the promises and challenges of using machine learning models to predict inefficiency and corruption in public procurement, both from the perspective of researchers and practitioners. We exemplify this procedure using a unique dataset characterizing more than 2 million public contracts in Colombia, and training machine learning models to predict which of them face corruption investigations or implementation inefficiencies. We use different techniques to handle the problem of class imbalance typical of these applications, report the high accuracy of our models, simulate the trade-off between precision and recall in this context, and determine which features contribute the most to the prediction of malfeasance within contracts. Our approach is useful for governments interested in exploiting large administrative datasets to improve provision of public goods and highlights some of the tradeoffs and challenges that they might face throughout this process.


Tweeting for Peace: Experimental Evidence from the 2016 Colombian Plebiscite

Abrir enlace

The decades-long Colombian civil war nearly came to an official end with the 2016 Peace Plebiscite, which was ultimately defeated in a narrow vote. This conflict has deeply divided Colombian civil society, and non-political public figures have played a crucial role in structuring debate on the topic. To understand the mechanisms underlying the influence of members of civil society on political discussion, we performed a randomized experiment on Colombian Twitter users shortly before this election. Sampling from a pool of subjects who had been frequently tweeting about the Plebiscite, we tweeted messages that encouraged subjects to consider different aspects of the decision. We varied the identity (a general, a scientist, and a priest) of the accounts we used and the content of the messages we sent. We found little evidence that any of our interventions were successful in persuading subjects to change their attitudes. However, we show that our pro-Peace messages encouraged liberal Colombians to engage in significantly more public deliberation on the subject.


10 steps to get your first job as a Data Scientist

Abrir enlace

Did you know that the unemployment rate in Data Scientists is negative? That is, there are more offers available in the market than people who can supply them. The demand for professionals in the field of data science is growing, and in this post, I will show you 10 infallible steps to get a job in this field even if you have no experience.


Loss and Waste of Food in Colombia

Abrir enlace

With the aim of having an aggregate calculation for Colombia and defining a work agenda, this study estimates the loss and waste of food in the country. In addition, it proposes a series of guidelines for its reduction in each link of the food chain and identifies the relevant actors for its implementation. It is expected that this study will be a reference to initiate various policy actions that will allow the country to advance in the solution of the problem and meet the goal of loss and waste of food according to the provisions of the Sustainable Development Goals by 2030.


5 Ways in Which Big Data Will Enrich Economic Analysis

Abrir enlace

The large volumes of data are a reality that touches the door of companies, the academy and even the public sector in order to significantly improve the efficiency in the handling and analysis of information. Economic studies should be coupled with these new tools to exploit their promising potential and generate deep knowledge from data.


Egresados rosaristas ganadores de becas Fulbright

Abrir enlace


Evaluaciones de los clientes

0.0


Valoración total de la Mente

5 estrellas

0

4 estrellas

0

3 estrellas

0

2 estrellas

0

1 estrellas

0

Dominio del tema

0.0

0.0

Discurso fluido y secuenciado

0.0

0.0

Capta la atención del público

0.0

0.0

Es inspirador e impactante

0.0

0.0

Conferencias, Cursos y/o Talleres


  • imagen portada Ciencia de los datos aplicada a producto: Predicción y modelado de comportamientos del consumidor

    Ciencia de los datos aplicada a producto: Predicción y modelado de comportamientos del consumidor

    De 1 Hora a 2 Horas

    Decisiones de producto que antes se basaban en intuición, experiencia o costosos estudios de mercad...

Videos en acción


  • "Research results preview: Tweeting for Peace: Experimental Evidence from the 2016 Colombian Plebiscite". Presented at Colombian Big Data Society Conference. July 27, 2017
  • Computational Sentiment Analysis on Twitter with R webinar
  • Data science applied to product: Prediction and modeling of consumer behaviors

Eventos y Clientes


  • D

    DaCodes

    • Trayectoria de un Data Scientist
    • 2021
    • Conferencia impartida a los empleados de la empresa DaCodes que describió las lecciones y de una carrera en ciencia de datos en múltiples entornos industriales.
  • e

    eTraining.co

    • Ciencia de los Datos Aplicada a Producto: Predicción y Modelado de Comportamientos del Consumidor
    • 2021
  • U

    Universidad del Rosario

    • Ciencia de los datos aplicada a producto: Predicción y modelado de comportamientos del consumidor
    • 2020
    • Decisiones de producto que antes se basaban en intuición, experiencia o costosos estudios de mercado, ahora son guiadas por experimentos, algoritmos predictivos y modelos estadísticos. Esta charla busca hacer una introducción a cómo metodologías basadas en datos pueden ayudar a desarrollar mejores productos y anticipar los comportamientos de los clientes.
Ver más Eventos y Clientes
  • D

    DaCodes

    • Trayectoria de un Data Scientist
    • 2021
    • Conferencia impartida a los empleados de la empresa DaCodes que describió las lecciones y de una carrera en ciencia de datos en múltiples entornos industriales.
  • e

    eTraining.co

    • Ciencia de los Datos Aplicada a Producto: Predicción y Modelado de Comportamientos del Consumidor
    • 2021
  • U

    Universidad del Rosario

    • Ciencia de los datos aplicada a producto: Predicción y modelado de comportamientos del consumidor
    • 2020
    • Decisiones de producto que antes se basaban en intuición, experiencia o costosos estudios de mercado, ahora son guiadas por experimentos, algoritmos predictivos y modelos estadísticos. Esta charla busca hacer una introducción a cómo metodologías basadas en datos pueden ayudar a desarrollar mejores productos y anticipar los comportamientos de los clientes.
  • U

    Universidad del Rosario

    • El camino hacia una beca internacional
    • 2019
  • F

    Facultad de Economía - Universidad del Rosario

    • Ciencia de datos aplicada al análisis económico
    • 2018
    • Conferencia dirigida a estudiantes de pregrado y posgrado, del departamento de economía de la Universidad del Rosario, sobre cómo utilizar las habilidades y metodologías de la ciencia de datos en la profesión de economía. Además, se brindó orientación y posibles trayectorias profesionales a los estudiantes que buscan convertirse en profesionales de la ciencia de datos.
  • B

    Big Data Colombia

    • Trae tus Propios Proyectos de Datos
    • 2018
  • B

    Big Data Colombia

    • Research results preview: Tweeting for Peace: Experimental Evidence from the 2016 Colombian Plebiscite
    • 2017
  • E

    Escuela de Verano - Universidad del Rosario

    • Analítica de Texto
    • 2017
  • U

    Universidad del Rosario

    • Summit on Peacebuilding
    • 2017
    • “Tweeting for Peace: Experimental Evidence from the 2016 Colombian Plebiscite”. Presentado en el Simposio: Political Economy, Conflict and Peace - Summit on Peacebuilding. Agosto 24, 2017
  • C

    CORPOVISIONARIOS

    • Panel de expertos que buscan formular una metodología para la medición del ciberacoso en redes sociales a través de métodos de aprendizaje automático
    • 2017
  • B

    BD Guidance

    • Fundamentos de Data Science
    • 2016
  • B

    BD Guidance

    • Computational Sentiment Analysis on Twitter with R webinar
    • 2016
  • C

    Colombia 4.0

    • Análisis de Sentimiento en R
    • 2016

Experiencia Profesional


  • M

    Data Scientist

    • Microsoft
    • 2021 - Actual
  • U

    Profesor

    • Universidad del Rosario
    • 2017 - Actual
  • P

    Data Scientist

    • Platzi
    • 2020 - 2021
Ver más Experiencias Profesionales
  • M

    Data Scientist

    • Microsoft
    • 2021 - Actual
  • U

    Profesor

    • Universidad del Rosario
    • 2017 - Actual
  • P

    Data Scientist

    • Platzi
    • 2020 - 2021
  • A

    Data Scientist

    • Alkanza
    • 2018 - 2019
  • P

    Data Scientist

    • Platzi
    • 2019 - 2019
  • D

    Coordinador de Proyectos Big Data

    • Departamento Nacional de Planeación
    • 2017 - 2018
  • P

    Consultor de Ciencia de Datos

    • PNUD
    • 2016 - 2017

Estudios


  • D

    Master's in Data Science

    • Duke University
    • 2021
  • U

    Economista

    • Universidad del Rosario
    • 2016
  • D

    Master's in Data Science

    • Duke University
    • 2021
  • U

    Economista

    • Universidad del Rosario
    • 2016

Artículos


Preventing rather than Punishing: An Early Warning Model of Malfeasance in Public Procurement

Abrir enlace
Is it possible to predict corruption and public inefficiency in public procurement? With the proliferation of e-procurement in the public sector, anti-corruption agencies and watchdog organizations in many countries currently have access to powerful sources of information. These may help anticipate which transactions become faulty and why. In this paper, we discuss the promises and challenges of using machine learning models to predict inefficiency and corruption in public procurement, both from the perspective of researchers and practitioners. We exemplify this procedure using a unique dataset characterizing more than 2 million public contracts in Colombia, and training machine learning models to predict which of them face corruption investigations or implementation inefficiencies. We use different techniques to handle the problem of class imbalance typical of these applications, report the high accuracy of our models, simulate the trade-off between precision and recall in this context, and determine which features contribute the most to the prediction of malfeasance within contracts. Our approach is useful for governments interested in exploiting large administrative datasets to improve provision of public goods and highlights some of the tradeoffs and challenges that they might face throughout this process.


Tweeting for Peace: Experimental Evidence from the 2016 Colombian Plebiscite

Abrir enlace
The decades-long Colombian civil war nearly came to an official end with the 2016 Peace Plebiscite, which was ultimately defeated in a narrow vote. This conflict has deeply divided Colombian civil society, and non-political public figures have played a crucial role in structuring debate on the topic. To understand the mechanisms underlying the influence of members of civil society on political discussion, we performed a randomized experiment on Colombian Twitter users shortly before this election. Sampling from a pool of subjects who had been frequently tweeting about the Plebiscite, we tweeted messages that encouraged subjects to consider different aspects of the decision. We varied the identity (a general, a scientist, and a priest) of the accounts we used and the content of the messages we sent. We found little evidence that any of our interventions were successful in persuading subjects to change their attitudes. However, we show that our pro-Peace messages encouraged liberal Colombians to engage in significantly more public deliberation on the subject.


10 steps to get your first job as a Data Scientist

Abrir enlace
Did you know that the unemployment rate in Data Scientists is negative? That is, there are more offers available in the market than people who can supply them. The demand for professionals in the field of data science is growing, and in this post, I will show you 10 infallible steps to get a job in this field even if you have no experience.


Ver más Artículos

Preventing rather than Punishing: An Early Warning Model of Malfeasance in Public Procurement

Abrir enlace
Is it possible to predict corruption and public inefficiency in public procurement? With the proliferation of e-procurement in the public sector, anti-corruption agencies and watchdog organizations in many countries currently have access to powerful sources of information. These may help anticipate which transactions become faulty and why. In this paper, we discuss the promises and challenges of using machine learning models to predict inefficiency and corruption in public procurement, both from the perspective of researchers and practitioners. We exemplify this procedure using a unique dataset characterizing more than 2 million public contracts in Colombia, and training machine learning models to predict which of them face corruption investigations or implementation inefficiencies. We use different techniques to handle the problem of class imbalance typical of these applications, report the high accuracy of our models, simulate the trade-off between precision and recall in this context, and determine which features contribute the most to the prediction of malfeasance within contracts. Our approach is useful for governments interested in exploiting large administrative datasets to improve provision of public goods and highlights some of the tradeoffs and challenges that they might face throughout this process.


Tweeting for Peace: Experimental Evidence from the 2016 Colombian Plebiscite

Abrir enlace
The decades-long Colombian civil war nearly came to an official end with the 2016 Peace Plebiscite, which was ultimately defeated in a narrow vote. This conflict has deeply divided Colombian civil society, and non-political public figures have played a crucial role in structuring debate on the topic. To understand the mechanisms underlying the influence of members of civil society on political discussion, we performed a randomized experiment on Colombian Twitter users shortly before this election. Sampling from a pool of subjects who had been frequently tweeting about the Plebiscite, we tweeted messages that encouraged subjects to consider different aspects of the decision. We varied the identity (a general, a scientist, and a priest) of the accounts we used and the content of the messages we sent. We found little evidence that any of our interventions were successful in persuading subjects to change their attitudes. However, we show that our pro-Peace messages encouraged liberal Colombians to engage in significantly more public deliberation on the subject.


10 steps to get your first job as a Data Scientist

Abrir enlace
Did you know that the unemployment rate in Data Scientists is negative? That is, there are more offers available in the market than people who can supply them. The demand for professionals in the field of data science is growing, and in this post, I will show you 10 infallible steps to get a job in this field even if you have no experience.


Loss and Waste of Food in Colombia

Abrir enlace
With the aim of having an aggregate calculation for Colombia and defining a work agenda, this study estimates the loss and waste of food in the country. In addition, it proposes a series of guidelines for its reduction in each link of the food chain and identifies the relevant actors for its implementation. It is expected that this study will be a reference to initiate various policy actions that will allow the country to advance in the solution of the problem and meet the goal of loss and waste of food according to the provisions of the Sustainable Development Goals by 2030.


5 Ways in Which Big Data Will Enrich Economic Analysis

Abrir enlace
The large volumes of data are a reality that touches the door of companies, the academy and even the public sector in order to significantly improve the efficiency in the handling and analysis of information. Economic studies should be coupled with these new tools to exploit their promising potential and generate deep knowledge from data.


Egresados rosaristas ganadores de becas Fulbright

Abrir enlace



Evaluaciones


0.0

5 estrellas

0

4 estrellas

0

3 estrellas

0

2 estrellas

0

1 estrellas

0
Valoración total de la Mente