Jose R. Zapata
Data Science Technical Leader
PhD in Information and Communication Technologies
Mercado Libre
Data Science Technical Leader at Mercado Libre. Researcher at GIDATIC and Professor at the Faculty of Information and Communication Technologies (TIC), Universidad Pontificia Bolivariana (UPB).
My interests are based around Data science, MLOps and Audio and Music information technologies, which includes Music information retrieval, Machine Learning, CI/CD/CT, Audio analysis and Data analysis. EN ESPAÑOL
Interests
- Data science
- MLOps
- Music Information Retrieval
- Audio signal processing
- Python
- Guitar
Education
-
PhD in Information and Communication Technologies, 2013
Universitat Pompeu Fabra (UPF), Spain
-
MEng in Telecommunications, 2008
Universidad Pontificia Bolivariana (UPB), Colombia
-
BSc in in Electronic Engineering, 2002
Universidad Pontificia Bolivariana (UPB), Colombia
Skills
Python
Pandas, Numpy, Matplotlib, Plotly, Seaborn, BeautifulSoup, Selenium
Data Science
Python, SQL, Pandas, BigQuery
Machine Learning &
Deep Learning
Deep Learning
Scikit-learn, Pytorch, NLP, Spacy
MLOps
Kedro, MLflow, Deepchecks, DVC, great expectations
Signal Processing
and Frequency Analysis of signals
Programming
SQL, Git, CI/CD, Bash, Linux
Blog
En Español
*
Album Lofi con Python Version Cero
Generacion automatica musica LoFi y videos con Python utilizando técnicas de inteligencia artificial y procesamiento de audio.
Paso a paso en un Proyecto Machine Learning
Checklist y preguntas para realizar un proyecto de machine learning
Visualizacion Datos Coronavirus (COVID19) Mundial con Plotly
Visualizaciones con Python y Plotly de los datos mundiales del corona virus COVID19
Pyspark con Google Colab
Configuracion de Google Colab para usar pyspark
Tutorial de Produccion de Audio con Reaper
Tutorial en Video y material produccion basica de Audio con Reaper
Tips para la Exposición del Proyecto de Grado
Tips para crear las diapositivas y para realizar la exposición del anteproyecto del trabajo de grado
Projects
Research Project that combines exclusive Facebook data (Condor Dataset, Crowdtangle, Ad’s) and public data to analyze the Social Media behavior to determine if there is coordinated non-authentic behavior.
This project in founded by the Social Media and Democracy Research Grants from the Social Science Research Council and access to Facebook data via Social Science One.
Research Project to explore the use of machine learning techniques for computational musicology, digital music archive managment, and music information retrieval.Two main elements are the core of our project: 1. Emphasis on semi-supervised and unsupervised machine learning techniques that minimally rely on the availability of annotated data for a specific task. 2. Traditional Colombian music as the main focus of our study.
Matlab implementation of the Multi Feature Beat Tracker (Information Gain and Regularity), The essentia beat tracker, More details in Multi-Feature Beat Tracking
Open-source C++ library for audio analysis and audio-based music information retrieval. It contains an extensive collection of reusable algorithms which implement audio input/output functionality, standard digital signal processing blocks, statistical characterization of data, and a large set of spectral, temporal, tonal and high-level music descriptors. More details in Essentia: An Audio Analysis Library for Music Information Retrieval
This beat tracking dataset contains 217 excerpts around 40s each, of which 19 are easy and the remaining 198 are hard. This dataset has been designed for radically new techniques which can contend with challenging beat tracking situations like: quiet accompaniment, expressive timing, changes in time signature, slow tempo, poor sound quality etc. More details in Selective Sampling for Beat Tracking Evaluation
Publications
Quickly discover relevant content by filtering publications.
(2021).
Sesquialtera in the Colombian Bambuco: Perception and Estimation of Beat and Meter – Extended version.
Transactions of the International Society for Music Information Retrieval, 4(1), pp.248-262.
(2020).
Sesquialtera in the Colombian Bambuco: Perception and Estimation of Beat and Meter.
International Society for Music Information Retrieval Conference (ISMIR 2020).
(2019).
ACMUS - Advancing Computational Musicology: Semi-supervised and Unsupervised Segmentation and Annotation of Musical Collections.
International Society for Music Information Retrieval Conference (ISMIR 2019).
(2016).
Tempo Estimation.
In Music Data Analysis: Foundations and Applications (pp. 493-510), Claus Weihs, Dietmar Jannach, Igor Vatolkin, Guenter Rudolph (Eds.), Chapman & Hall/CRC, 2016, ISBN 9781498719568.
(2014).
Multi-Feature Beat Tracking.
IEEE/ACM Transactions on Audio, Speech, and Language Processing, Vol. 22, No. 4, P. 816-825.
(2014).
Essentia: an open source library for audio analysis.
ACM SIGMM Records, Vol. 6, No. 1, P. 18-21.
(2013).
Essentia: An Audio Analysis Library for Music Information Retrieval.
14th International Society for Music Information Retrieval Conference (ISMIR 2013), P. 493-498, Curitiba, Brazil.
(2013).
Essentia: an open-source library for sound and music analysis.
21st ACM international conference on Multimedia (ACM Multimedia 2013, MM ‘13), P. 855-858, Barcelona, Spain.
(2013).
Comparative evaluation and combination of automatic rhythm description systems.
Music Technology Group (MTG), Universitat Pompeu Fabra.
(2013).
Using voice suppression algorithms to improve beat tracking in the presence of highly predominant vocals.
IEEE The 38th International Conference on Acoustics, Speech, and Signal Processing , ICASSP 2013, P. 51-55 , Vancouver, Canada.
(2012).
Selective Sampling for Beat Tracking Evaluation.
IEEE Transactions on Audio, Speech and Language Processing, Vol. 20, No. 9, P. 2539-2548.
(2012).
Assigning a confidence threshold on automatic beat annotation in large datasets.
13th International Society for Music Information Retrieval Conference (ISMIR 2012), P.157-162, Porto, Portugal.
(2012).
Improving Beat Tracking in the presence of highly predominant vocals using source separation techniques: Preliminary study.
The 9th International Symposium on Computer Music Modeling and Retrieval (CMMR 2012), London, UK.
(2012).
On the automatic identification of difficult examples for beat tracking: towards building new evaluation datasets.
IEEE The 37th International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2012), P. 89-92 Kyoto, Japan.
(2011).
Comparative Evaluation and Combination of Audio Tempo Estimation Approaches.
42nd AES International Conference, Semantic Audio (Audio Engineering Society), P.198-207, Ilmenau, Germany..
(2008).
Efficient Detection of Exact Redundancies in Audio Signals.
Audio Engineering Society Convention 125, Paper = 7504, San Francisco, EE.UU.
(2004).
Control de Rango Dinamico en Audio con Logica Difusa.
3ra Conferencia Iberoamericana en Sistemas, Cibernética e Informática CISCI 2004, Orlando, EE.UU.
Talks
Teaching
Current and past, postgraduate and undergraduate courses at the Universidad Pontificia Bolivariana (UPB)
| Postgraduate Courses | │ | UnderGraduate Courses |
|---|---|---|
| Python for Data Science | │ | Machine Learning |
| Audio Data Mining | │ | Signals and Systems |
| Principles of Image Processing | │ | Data Mining |
| Audio Processing | │ | Principles of Audio |
| Signal Processing | │ | Audio Applications |
| Telecommunications Principles | │ | Telecommunications Theory & Lab |
| Matlab | │ | Tx Lines and antennas |
Experience
Data Science Technical Leader
Data Scientist Sr
Full Professor
Phd. Student
Thesis: Comparative evaluation and combination of automatic rhythm description systems,
advisor. Emilia Gomez
Phd. Internship
Advisor. Fabien Gouyon, Matthew E.P. Davies
Assistant Professor