MZMarcos Zárate

Marcos Zárate

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Marcos Zárate has a total of 33 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • The Multiplex Classification Framework: Optimizing Multi-Label Classifiers Through Problem Transformation, Ontology Engineering, and Model Ensembling

    by Mauro Andrés Nievas Offidani, Facundo Roffet, Claudio Augusto Delrieux, María Carolina González Galtier, Marcos Daniel Zarate

    Classification is a fundamental task in machine learning. While conventional methods—such as binary, multiclass, and multi-label classification—are effective for simpler problems, they may not adequately address the complexities of some real-world scenarios. This article introduces the Multiplex Classification Framework, a novel approach developed to tackle these and similar challenges through the integration of problem transformation, ontology engineering, and model ensembling. The framework offers several advantages, including adaptability to any number of classes and logical constraints, an innovative method for managing class imbalance, the elimination of confidence threshold selection, and a modular structure. Two experiments were conducted to compare the performance of conventional classification models with the multiplex approach. Our results demonstrate that the multiplex approach can improve classification performance significantly (up to 10% gain in overall F1 score), particularly in classification problems with a large number of classes and pronounced class imbalances. However, it also has limitations, as it requires a thorough understanding of the problem domain and some experience with ontology engineering, and it involves training multiple models, which can make the whole process more intricate. Overall, this methodology provides a valuable tool for researchers and practitioners dealing with complex classification problems in machine learning.


  • Open Access

    SES Dashboard: A linked data platform to manage data on Southern Elephant Seals

    by Marcos Zárate, Elena Eder, Claudio Delrieux, Mirtha Lewis, Gustavo Nuñez, Dario Ceballos

    In this paper we present SES Dashboard, a linked data platform developed to manage information of Southern Elephant Seals (Mirounga leonina Linnaeus, 1758) from Península Valdés colony (Patagonia, Argentina), collected over more than two decades of research. Southern elephant seals spend most of their time foraging at sea, alternating with two short and highly synchronized haul-outs ashore for breeding and molting. On land, the distribution of the colony extends along ≈350 km, while at sea individuals travel thousands of kilometers The information recorded is mainly based on georeferenced data from censuses, bioregisters (trajectories at sea and hydrographic profiles), and bibliography based on the above data. SES Dashboard was released in 2021 and published following the principles of Linked Open Data to improve its visibility, to make it shareable, extensible and easily reusable through a SPARQL endpoint.

  • Open Access

    Light and temperature records of the seawater associated with southern elephant seal dives during foraging trips in South Atlantic and Pacific Oceans

    by Elena Eder, Marcos Zárate, Mirtha Lewis

    The dataset comprises geolocalised records of dive and surface interval durations, light level and temperature of the seawater during the post-resting and post-moulting tracks of 13 immature southern elephant seals, Mirounga leonina. It describes an unpublished open access version of the original data with records of light level and temperature of the water column using the Darwin Core standard (DwC) through ArOBIS, guaranteeing compliance with the FAIR principles, encompassing a wide time scale (2005, 2006 and 2007) and geographic range in the South Atlantic and Pacific Oceans (South West [-58.75, -81.29], North East [-37.60, -28.65]). Seals were simultaneously equipped with affordable light–temperature loggers (LTLs) and satellite tags. The LTLs recorded light level and temperature of the water column at 30-s intervals during dives and light–time records were applied to estimate dive parameters of diurnal records from 06:00 to 17:00 h, since movements up and down the water column are reflected by changes in light level. For that, the minimum light level reached at the surface of a dive was determined experimentally with diurnal dive simulations at sea using the LTLs devices before deployment. The dataset also includes variation of light and temperature of records between 17:00 to 06:00 h. Data can be used to identify temperature changes associated with seawater masses as drivers of the distribution of other taxa of interest and variation of light level in the seawater (light attenuation) could be linked to concentrations of phytoplankton assemblages as an index of primary productivity. This dataset provides unpublished data of the duration of dives and surface intervals and associated records of light level and temperature variations along the movements throughout the seawater of 13 immature southern elephant seals in the Southern Hemisphere. The location data were generated by satellite tags and the light and temperature data were recorded with light-temperature loggers (LTLs), both devices deployed on individuals simultaneously and uploaded following the Darwin Core standard and compliance with the FAIR principles.

  • Open Access

    Light and temperature records of the seawater associated to southern elephant seal dives during foraging trips in South Atlantic and Pacific Oceans

    by Elena B. Eder, Marcos Zárate, Mirtha N. Lewis

    The dataset encloses geolocalized records of dive and surface interval durations, light level and temperature of the seawater during the postresting and postmolting trackings of 13 immature southern elephant seals, Mirounga leonina. It describes an unpublished open access version of the original data with records of light level and temperature of the water column using the Darwin Core standard (DwC) through ArOBIS guaranteeing compliance with the FAIR principles, encompassing a wide time scale (2005, 2006 and 2007) and geographic range in the South Atlantic and Pacific Oceans (South West [-58.75, -81.29], North East [-37.60, -28.65]). Seals were simultaneously equipped with affordable light–temperature loggers (LTLs) and satellite devices. The LTLs recorded light level and temperature of the water column at 30-s intervals during dives, and light–time records were applied to estimate dive parameters of diurnal records from 06:00 am to 05:00 pm, since movements up and down the water column are reflected by changes in light level. For that, the minimum light level reached at the surface of a dive was determined experimentally with diurnal dive simulations at sea using the LTLs devices before deployment. The dataset also includes variation of light and temperature of records between 17 pm to 06:00 am. Data can be used to identify temperature changes associated to seawater masses as drivers of the distribution of other taxa of interest, and variation of light level in the seawater (light attenuation) could be linked to concentrations of phytoplankton assemblages as an index of primary productivity. This dataset provides unpublished primary data of the duration of dives and surface intervals, and associated records of light level and temperature variations along the movements throughout the sea water of 13 immature southern elephant seals in the southern Hemisphere. Data were recorded with light-temperature loggers (LTLs) and uploaded following the Darwin Core standard and compliance with the FAIR principles.

  • BiGe-Onto: An ontology-based system for managing biodiversity and biogeography data1

    by Marcos Zárate, Germán Braun, Pablo Fillottrani, Claudio Delrieux, Mirtha Lewis

    Great progress to digitize the world’s available Biodiversity and Biogeography data have been made recently, but managing data from many different providers and research domains still remains a challenge. A review of the current landscape of metadata standards and ontologies in Biodiversity sciences suggests that existing standards, such as the Darwin Core terminology, are inadequate for describing Biodiversity data in a semantically meaningful and computationally useful way. As a contribution to fill this gap, we present an ontology-based system, called BiGe-Onto, designed to manage data together from Biodiversity and Biogeography. As data sources, we use two internationally recognized repositories: the Global Biodiversity Information Facility (GBIF) and the Ocean Biogeographic Information System (OBIS). BiGe-Onto system is composed of (i) BiGe-Onto Architecture (ii) a conceptual model called BiGe-Onto specified in OntoUML, (iii) an operational version of BiGe-Onto encoded in OWL 2, and (iv) an integrated dataset for its exploitation through a SPARQL endpoint. We will show use cases that allow researchers to answer questions that manage information from both domains.


  • Harmonizing Big Data with a Knowledge Graph: OceanGraph KG Uses Case

    by Marcos Zárate, Carlos Buckle, Renato Mazzanti, Mirtha Lewis, Pablo Fillottrani, Claudio Delrieux


  • Open Access

    Linked Open Biodiversity Data (LOBD): A semantic application for integrating biodiversity information

    by Marcos Zárate, Paula Zermoglio, John Wieczorek, Anabela Plos, Renato Mazzanti

    Scientists frequently collect biological and environmental information over years and store it in database systems to answer their own research questions without exposing it in repositories that make it easy to find and retrieve. While in recent years the community working on biodiversity informatics has made significant strides by creating common shared vocabularies such as the Darwin Core (DwC, Wieczorek et al. 2012) and publishing mechanisms such as the Integrated Publishing Toolkit (IPT, Robertson et al. 2014), integration is largely limited to the aggregation of datasets and full interoperability has still not been achieved. In this context, The Semantic Web (SW) aims to represent information in a way that, in addition to the human-centered display purposes, it can be used autonomously by machines for integration and reuse across applications. From the biodiversity informatics point of view, interoperability and links among data sources would allow integration of information that is otherwise disconnected, enabling scientists to answer broader questions. These considerations provide strong motivations to formulate a web application considering the semantic interoperability that may provide answers to questions such as the following: (Q1) Is it possible to complement taxonomic, bibliographic and environmental information of a particular species without relying on specific Application Programming Interfaces (APIs)? (Q2) How to relate occurrences of species with environmental variables within a specific region? (Q3) What are the bibliographic references associated with a given species? (Q1) Is it possible to complement taxonomic, bibliographic and environmental information of a particular species without relying on specific Application Programming Interfaces (APIs)? (Q2) How to relate occurrences of species with environmental variables within a specific region? (Q3) What are the bibliographic references associated with a given species? With questions such as these in mind, we present the design of a proof-of-concept application: Linked Open Biodiversity Data (LOBD). LOBD uses Linked Data (LD) (Heath and Bizer 2011) to complement species occurrence information previously extracted from GBIF and converted to Resource Description Framework (RDF) (Zárate et al. 2020) with information about the taxa in question from different RDF datasets, such as Wikidata, NCBI Taxonomy, Springer Nature SciGraph and OpenCitation corpus. A simplified view of the architecture is shown in Fig. 1. To achieve semantic interoperability, we use the SPARQL query language, which allows us not to depend on specific APIs to retrieve information. The application consists of three modules: General information, where the Wikidata endpoint is used to retrieve additional information about the selected species, including links to other databases and information about the species extracted from National Center for Biotechnology Information (NCBI) Taxonomy. Bibliography, where all publications related to the species are retrieved and extracted from OpenCitation. Environment, where users can plot species on a map and add layers related to marine regions as well as environmental layers (e.g., temperature, salinity, etc). General information, where the Wikidata endpoint is used to retrieve additional information about the selected species, including links to other databases and information about the species extracted from National Center for Biotechnology Information (NCBI) Taxonomy. Bibliography, where all publications related to the species are retrieved and extracted from OpenCitation. Environment, where users can plot species on a map and add layers related to marine regions as well as environmental layers (e.g., temperature, salinity, etc). For the development of the application, we use the Shiny framework for R, access to SPARQL endpoints is done through the SPARQL package, marine regions are obtained from marineregion.org and the environmental layers are extracted from Bio-ORACLE. The data used for this article were collected by the Center for the Study of Marine Systems at the National Patagonian Sci-Tech Centre (CCT CENPAT-CONICET), and are published and available through the GBIF network. Linked Data is a powerful tool for scientists, as it allows generating new approaches to biodiversity informatics, which can help to address the data integration challenges. Users would benefit from complementing the current prevalent use of vocabularies that are not ontologically defined (like DwC) for sharing biodiversity data. Although this application is a proof of concept, it shows that with little effort, it is possible to achieve greater interoperability between datasets that were not initially represented as LD.


  • Open Access

    Data Interoperability Between Elements of the Global Ocean Observing System

    by Derrick Snowden, Vardis M. Tsontos, Nils Olav Handegard, Marcos Zarate, Kevin O’ Brien, Kenneth S. Casey, Neville Smith, Helge Sagen, Kathleen Bailey, Mirtha N. Lewis, et al.

    The data management landscape associated with the Global Ocean Observing System is distributed, complex, and only loosely coordinated. Yet interoperability across this distributed landscape is essential to enable data to be reused, preserved, and integrated and to minimize costs in the process. A building block for a distributed system in which component systems can exchange and understand information is standardization of data formats, distribution protocols, and metadata. By reviewing several data management use cases we attempt to characterize the current state of ocean data interoperability and make suggestions for continued evolution of the interoperability standards underpinning the data system. We reaffirm the technical data standard recommendations from previous OceanObs conferences and suggest incremental improvements to them that can help the GOOS data system address the significant challenges that remain in order to develop a truly multidisciplinary data system.


  • OceanGraph: Some Initial Steps Toward a Oceanographic Knowledge Graph

    by Marcos Zárate, Pablo Rosales, Germán Braun, Mirtha Lewis, Pablo Rubén Fillottrani, Claudio Delrieux