German researchers have found the Dutch University’s ontology of measurements to be the best worldwide performer. From apps with personalized recipes to systems that predict the shelf-life of fresh fruit and vegetables, programs only work if computers know how to handle the underlying data.

Ontology is like a dictionary for computers, giving meaning to words and numbers and their relationships, helping software programs unlock, connect and interpret data from different sources.

This makes it possible to develop smart applications and software solutions, enabling companies to make informed decisions and allowing processes to be automated in a smart way.

The independent German survey from May 2017 found that the ontology of measurements by Wageningen Food & Biobased Research is the best performer internationally.

Jan Top, senior scientist for Food Informatics at Wageningen Food & Biobased Research, explains how widely applicable Wageningen’s OM is. “Product specification in supply chains; when exchanging physical data about products, parties need to be sure that they explicitly define the units and quantities used. This ontology can be a shared reference to incorporate the standards in their software,” he says.

“Research and development; We have developed a (prototype) extension of Excel that supports semantic annotation of data. This means that users can describe, select and combine spreadsheet data using terms they are familiar with.”

“In our applications, we focus on the agrifood domain, but this ontology is generic and can also be used in any other domain where data on the physical properties of objects are shared.”

Ontologies come in all shapes and sizes. One is the ontology of measurements (OM), which contains thousands of terms on quantities and units pertaining to things like length, weight and energy value – from meters and inches to spoons and milliliters per liter.

The external evaluators concluded that OM by Wageningen Food & Biobased Research offers the most application possibilities of all the existing variants. They evaluated seven ontologies for quantities and units on the basis of criteria such as consistency, completeness and ease of use.

The results were then combined with the faculties necessary to convert software units or compare the results of scientific experiments with each other.

“Ontology is the first step toward unlocking data,” adds Top. “But that unlocking only makes sense if the knowledge needed to make good decisions with the data is modeled as well. We have built many different concepts and relationships into our ontology, making it suitable for a wide range of applications both within and beyond the agrifood sector.”

”Wageningen’s OM is used in smart software which can, for instance, adapt recipes to someone’s personal requirements, or translate them from English into Dutch – including converting ounces and other terms from English cuisine to grams, for example.

“The ontology, which is available as open source software, is essential in making calculations and predictions based on data, which often originates in a combination of sources. An example of another relevant theme is food security; for instance, what is the impact of rising energy prices on food in different countries?”

Combining ontologies
Wageningen experts often combine OM with other ontologies. “We do this for instance when developing apps which adapt recipes to consumers with food allergies,” explains Top.

Combinations of ontologies are also used to predict the quality of fresh fruit and vegetables in the chain or to lend farmers in Europe a hand in finding information about useful innovations.

Another example is research into the growth in global protein demand and how this is related to the changing use of biomass. Wageningen Food & Biobased Research has also developed a method with associated software called ROC+, which allows groups of experts to efficiently create a new ontology by themselves.