Towards efficient sharing and discovery of foodborne diseases information

foodmorneA high level meeting on Open Data in Agriculture took place on 28 September 2015 in Amsterdam, Netherlands. The participants of the event represented organisations like the Global Forum on Agricultural Research (GFAR), the Food and Agriculture Organisation of the UN (FAO), Land Portal Foundation, Wageningen UR, Open Data Institute (ODI) and Institute of Development Studies, UK (IDS). 

GODAN

Most of these organisations are members of GODAN, a network which supports the proactive sharing of open data to make information about agriculture and Open Data in Agriculture 1nutrition available, accessible and usable to deal with the urgent challenge of ensuring world food security. GODAN is a rapidly growing group, currently with over 160 partners from national governments, non-governmental, international and private sector organisations. 

Presenting the Food Safety use case of OpenMinTeD

Nikos Marianos from Agro-Know participated in the event to present the Food Safety use case of OpenMinTeD, collect requirements and discuss how it can be supported by GODAN and DFID. This event is important for OpenMinTeD, since the involved organisations are part of the stakeholder groups of the Agriculture/Biodiversity research community of OpenMinTeD.

Overview of the Food Safety use case

The Food Safety use case is focused on using TDM technologies to achieve efficient sharing and discovery of foodborne diseases information. The problem of discovery of information for foodborne diseases, food alerts, outbreaks and recalls has been stressed as a problem with major importance by global initiatives like the Global Food Safety Partnership (GFSP) that works on improving the capacity building for food safety.

 The efficient discovery of information about foodborne diseases, food alerts, recalls and outbreaks available at diverse sources can be enabled by the use of

  1. a common semantic vocabulary (e.g. FoodEx2#) that will operate as the backbone for harmonizing the information is needed. As the amount of information is not possible to be manually annotated with such multilingual semantic vocabulary, efficient text mining methods that will automatically extract vocabulary/ontology terms from such information

  2. text mining tools that will extract structured data from trusted web sources containing information about foodborne diseases, outbreaks and recalls

  3. text mining tools that can extract the secondary data from publications and reports such as table with results, images and genetic information (DNA of bacteria)

  4. tools that will transform the annotated information for foodborne diseases to linked data maximizing the processing of the information by machines. 

  5. of text mining tools that will enable the automatic linking of food outbreaks’ reports with food recall data.

Addressing the above data challenges will allow scientists to find foodborne diseases information in order to analyse and correlate the data and to create notifications about food safety issues. Apart from the data and technological enablers the most important prerequisite for applying state of the art text analysis techniques is the availability and openness of information regarding the foodborne diseases, outbreaks, food alerts and recalls.

Open Data in Agriculture 2Outcomes

The potential global societal and economic impact of the use case was recognised by the participants and they expressed their interest in adopting the OpenMinTeD Food Safety use case in the context of the GODAN Theory of Change. The use of TDM solutions in other use cases to solve their identified challenges was also welcomed, showing that the food and agriculture community has much to gain from the outcomes of OpenMinTeD project.