Human Aware Sensor Network Ontology

Printer-friendly version

Initial Ontology development effort

A Like Science Ontology (LSO) was developed to address the ontological need of integrating Jefferson-related data. LSO has evolved into HASNetO described below.

Human-Aware Sensor Network Ontology (HASNetO)

In addition of being a device that outputs numbers, a sensor is an entity with characteristics such as what it measures, in what units, and to what precision, which are necessary to understanding what the numbers it records actually mean. HAS-Net O leverages PROV, OBOE, and VTSO in order to describe the sensors and instruments themselves, and related events such as instrument deployment. This will maintain structured metadata about the sensors and instruments, providing valuable context behind the measurements gathered during the course of the project.

See presentation:
Paulo Pinheiro A Human-Aware Sensor Network Ontology [Download], e-Science Group Meeting, Tetherless World Constellation, Troy, NY, June, 2014.

Access the HASNetO ontology
[current version] of the ontology.


The VSTO-Instrument is a copy of the VSTO ontology version 3 that has been stripped of concepts and relationships that are not directly connected to instruments and datasets. In particular, VSTP-Instrument does not include VSTO concepts that are subsumed by OBOE concepts.

Access the ontology
[current version] to the ontology.

OBOE Enhancements for Scientific Data Collection

OBOE Extensions for Jefferson Project

For the data in the Jefferson Project, we utilize the OBOE (Extensible Observation Ontology) vocabulary to describe the Entities we are measuring, the Characteristics being measured by the sensors, the units in which the data are recorded. OBOE was originally designed for describing ecological data like what the Jefferson Project aims to collect, however our use case covers more concepts than are represented in the exiting OBOE vocabulary. In order to ensure complete coverage of the data from Lake George, I created our own Jefferson Project extension ontologies to add the concepts required to fully describe all of our data.

One of our goals is to allow interoperability between our future data as well as data collected in the past, and so extensions of OBOE concepts had to take both of these types of data into consideration. The Fund for Lake George already had data collected over the last thirty years, which they report on in their State of the Lake document. I reviewed the original data used to generate this report to take an inventory of concepts already measured, and then compared them to what was already available in OBOE to determine extensions.

For the data we anticipate collecting during the course of the Jefferson Project, I had to consult the lists of sensors that we purchased for use. Documentation for the sensors provided Characteristic and Standard units information, and I gathered this in one place to review, again checking for what was already represented in the OBOE ontologies and what still needed to be added. One particular area of focus was addition of terminology for describing weather data - while OBOE already had fairly broad coverage for water quality measurements, weather was not as well represented. I also needed to add more entities of interest to the project, which included the sensors themselves in some cases. Certain models record their own temperature, which may be important to determining data quality if a sensor moves outside of its operating temperature range. The vertical profilers also record the depth at which they take measurements, which is important contextual metadata that is distinct from the total depth of the lake at the same point.

On both fronts, I went through several iterations, consulting with the Darrin Freshwater Institute and IBM's meterologist in order to ensure that concepts were described completely and related accurately. As a result, we now have three ontologies parallel to OBOE's structure with additional coverage of concepts important to the Jefferson Project:

Access the ontologies
[link] for jp-standards extension (v0.4 uploaded 2015/7/25).
[link] for jp-characteristics extension (v0.4 uploaded 2015/7/25).
[link] for jp-entities extension (v0.3 uploaded 2015/7/25).

CCSV Vocabulary

Contextualized Comma Separated Values (CCSV) is an extension to CSV that provides content and context restrictions by using a Turtle preamble on top of the actual CSV data. We have developed this vocabulary to support the automation of processing CCSV files.

Access CCSV ontology
[link] for CCSV vocabulary (v0.1 uploaded 2015/4/27).