Towards a Global Data Ecosystem for Agriculture and Food

Towards a Global Data Ecosystem for Agriculture

and Food

A story Who we are Open data in agri-food Cases from our work A common data ecosystem?

6

A story

1

Κρήτη (Crete)• Largest & most populous of Greek islands• Significant part of economy & cultural heritage of

Greece while retaining own local cultural traits (poetry, music, dialect)

• Once center of Minoan civilization (circa 2700–1420 BC)

…the earliest recorded civilization in Europe

Minoan civilisation

• Named after King Minos

• A king of Crete, son of Zeus and Europa

Minoans: enemies with Athens

• Every nine years, King Minos of Crete made King Aegeus of Athens to pick seven young boys and seven young girls to be sent to his palace, the labyrinth, to be eaten by the monster Minotaur (half man, half bull)

Theseus prince of Athens

princess Ariadne, daughter of Minos

so the myth is about navigating through

a labyrinth

our story: helping people navigate through information

6

Who we are

2

We find, connect and deliveragriculture & food information worldwide

(for the past 8 years)

So yes, we consider ourselves pioneers

We help organizations build a better future for all, using open agriculture and

food data

We work with major institutions and initiatives

We are a flexible, intuitional team of experts that delivers

Expert in strategy design Designs the most

efficient solutions

Expert in customer needs identification

and analysis

The Strategist

The Architect

The Perfectionist

Nikos Manouselis Giannis

Stoitsis

Babis Thanopoulos Expert in data and

technology solutions

Kostas Kastrantas

The Doer

Blending domain experts, computer engineers and information scientists

We Lead We MakeWe Support

The elements

Big Data Indexing

Sentiment Analysis & Opinion Mining

Big Data Processing

Information Extraction

Information Retrieval

Summarization Social Network Analysis

Large-scale Schema Matching

Hessian-Free Non-Linear Optimization

Argumentative Ontology Alignment

Semantic Annotation

Ontology Evolution

Ontology Population

Deep Kernelized Multi-task Learning

Convolutional Networks

Cascade Learning Machines

Data-intensive R&D linking basic with applied research in real-life settings

open stack of software for big data analytics & text/data mining

6

Open Data in Agri-Food

2

6

Cases from our work

3

AGRIS

8 million bibliographic

records350.000 visitors per month

Serving >200 data providers20% of the total FAO

visits

We aggregate and serve agricultural scientific information to maximise the impact of AGRIS

TAPipedia

A G20 initiative

GWPP Global Water Pathogen Project

Global Scientific Knowledge Hub for Water Pathogens

UNESCO, Procter & Gamble, Gates Foundation

VitisAll science about Greek viticulture

Mac

hine

Lea

rnin

gTo

olbo

x

High-Throughput Genotyping

VITIS Open Linked Portal

DATASETS

2016 2017 2018 2019 2020

Image Recognition / Convolutional Networks

BIBLIOGRAPHIC

Big DataIndexing

InformationRetrieval

+ PHENOTYPIC + GEOLOCATIONAL + GENETIC + SATELLITE IMAGING

High-Throughput Phenotyping

Identification

QuantificationFeature Extraction / Image Analysis

Prediction Sequence Alignment / HMMs

DataToolbox

Classification

InformationExtraction Summarization

ModellingToolbox

Large-scale Schema Matching

SemanticAnnotation

OntologyPopulation

Notifications: SMS, email based on company’s food profile

Analytics: provide Food Safety analytics per product, ingredient and per region

Supplier check: check if the current or a new supplier has been involved in a food incident in the past

Regulations: access regulations and standards for the product and ingredients per region

Decision support: take decisions about new ingredients that a company would like to use in a product

6

Towards a common data ecosystem?

4

Doing business with open data“new businesses and new business models are beginning to emerge: Suppliers, aggregators, developers, enrichers and enablers”“key link in the value chain for open data is the consumer…direct relevance to the choices individuals make as part of their day-to-day lives”

“For a data revolution to happen, agriculture and food need a fabric of interoperable and interplaying infrastructure layers that will make data sharing and exchange as natural to us as it is to use the road or rail infrastructure to move from one country to another.”

Agroknow’s response to A global data ecosystem for agriculture & food, September 2016

The right people together• Open Harvest’16 & Chania Declaration: public and private

sector want to work more closely together on data sharing• GODAN Data Ecosystem for Agri-food WG to bring senior

data infrastructure people together (co-led by Agroknow & Syngenta)

• Engage people developing data infrastructure pieces– CTO meeting is planned– follow up paper on Technology Principles for Enhanced

Data Interoperability & Discoverability (?)

Sharing scientific knowledge• Agri-food industry investing into Linked R&D Data

platforms• EU and CGIAR build platforms based on Linked

Open Data and Big Data Analytics for their scientific data and outputs

• Agroknow, INRA, WUR & FAO to coordinate the development of a 10y roadmap for a scientific data infrastructure in agriculture and food

How to get involved?• If you are building data platforms or infrastructures– join the GODAN WG (talk to me)

• If you are producing, managing or using scientific data– follow GODAN for an announcement on roadmap

development events (and talk to me)• If you are funding such work– share with us your priorities, interests and planned

interventions to (at least) avoid overlaps and double-funding (you can also talk to me)

4

Thank you