I want to search


Other Summer Internship Opportunities

Below are internship opportunities offered by DataONE leaders and/or partners through other channels. These are NOT DataONE funded internship opportunities and the application and selection process will vary by position. Stipends, duration and criteria may also differ. Please contact the internship mentor listed for full details.

DataONE Summer Internship Program opportunities are advertised here.

Entity Linking Management Framework

Mentor / Contact: Deborah McGuinness (dlm@cs.rpi.edu)
Other Mentor(s): James McCusker (RPI)
Necessary Prerequisites: Knowledge of Java and python web frameworks, RDF/Linked Data, and web service concepts.
Desirable Skills / Qualifications: Knowledge of Natural Language Processing (NLP) concepts and algorithms
Expected Outcomes: A generalized framework for deploying and managing entity linking tools based on dynamic or pre-defined target ontologies
Project Description: Currently, existing entity linking solutions have a diverse set of APIs and management strategies. The Earth Sciences Ontology Repository (ESOR) linking platform allows for a fixed set of target ontologies to be indexed at deployment, without the ability to extend or limit those ontologies through the access API. We will build a consistent API that will allow users to submit requests against particular APIs, and to allow Entity Linking (EL) implementers to access and store ontologies and their local indexing approaches on demand.

Ontology-based Prospective Provenance

Mentor / Contact: Deborah McGuinness (dlm@cs.rpi.edu)
Other Mentor(s): James McCusker (RPI) John Erickson, Paulo Missier
Necessary Prerequisites: Knowledge and experience in ontology engineering and modeling using OWL 2
Desirable Skills / Qualifications: Knowledge of the recommended provenance language for the web and the PROV-O Ontology and prospective provenance vocabularies and standards, including workflow specifications.
Expected Outcomes: A generalized theory of ontological prospective provenance with mappings to commonly used prospective provenance vocabularies. If successful, publication should also be possible
Project Description: The project rests on a simple hypothesis: prospective provenance defines classes of activities that haven't happened yet. As such, a large fraction of prospective provenance vocabularies can be reduced (in the computer science sense) to expressions of subclasses of prov:Activity, prov:Agent, and prov:Entity. We will evaluate this hypothesis by developing mappings of a number of prospective provenance vocabularies and standards to the core recommended ontology for provenance on the web (PROV-O) vocabulary, where instances in prospective provenance become classes in PROV-O. We will also evaluate for usefulness and ease of expression to determine if there are practical as well as theoretical benefits to such a mapping.