Competition Registration is now Live
Hackathon Project Submission Deadline:
Event Organisers and Sponsors
Bringing together academia and industry we have joined power with :
Overview
Big data will enable animal health and veterinary science to embark on information technology-driven discoveries to face daunting challenges such as adaptation to climate changes, mitigation of environmental impacts, and developing epidemiological models on current and emerging diseases. With that, the rapid development of artificial intelligence (AI) models is expected to revolutionise human and veterinary medicine, whilst having a multitude of beneficial uses in animal health science.
The 2024 Competition will run from
the 2nd October 2024 to 10th December 2024
Objectives
The competition is seeking applications that address novel uses of the LEO data on livestock, and which leverage approaches in AI, data sciences, and statistics. Such applications can address any of the following objectives:
- Develop prediction models, visualization tools and infographics
- Identify hidden patterns
- Explore the potential of artificial intelligence
- Create chatbots and virtual consultancy (for farmers, vets, consortia/associations and scientists)
- Promote the development of big open data use in livestock
- Stimulate innovation development in animal health AI
- Reinforce collaboration between academia, industry and society within and across countries
- Strengthen multidisciplinary collaboration
Specific proposals (including but not limited to):
- Identify hidden patterns
Utilize big data analytics to identify hidden patterns, trends and correlations in livestock health exploiting the LEO data repository.
- Innovative visualization tools and infographics
Propose innovative visualization tools and infographics to present Livestock data in an accessible, actionable and meaningful format (by breed, farm, geographic area, etc.)
- Predict welfare in dairy cattle
Create models to predict welfare in dairy cattle based on productive and reproductive performances data, environmental factors, and genetic information stored in LEO data.
- Risk of diseases in dairy cattle by using AI and machine learning
Explore the potential of AI and machine learning to estimate the risk of diseases in dairy cattle exploiting LEO data.
- Development of chatbots
Develop chatbots and virtual consultancy for different types of stakeholders (farmers, vets, consortia/associations and scientists) querying the LEO data.
Who can apply?
Any individual or team that has an interest in (but not limited to) animal and veterinary sciences, digital health, data visualization, machine learning and AI models.
FREE ENTRY
Competition Timeline
2nd October, 2024
Online webinar
2nd October, 2024
Competition Registration opens
14th November, 2024
Competition closes- Submission deadline
15th November, 2024
Evaluation phase opens. Appointed panel of Judges reviews submissions
15th November, 2024
Participant voting opens
20th November, 2024
Participant voting closes
21st November, 2024
Winning Projects Notified
10th December 2024
Hackathon Event at Surrey University, UK
The winners will also be invited to attend and present at the LEO event. Date and Location TBD
The Data
The Livestock Environment Open Data Project (LEO) has created a unique worldwide linked open data on Italian livestock that contains more than 15 billion triplets (ID-Data-Measure) generated, among others, including:
30,074,112 dairy and beef cattle
6,242,945 sheep
1,328,356 goats
Data are collected routinely on 18,194 livestock farms.
Data is available on the following:
- Dairy and Beef Cattle
- Sheep and Goats
- Pigs
- Horses
- Buffaloes
- Rabbits
- Field (e.g., inseminations, pregnancies, weights)
- Laboratory (e.g., urea, fatty acids) (dairy cattle milk spectra available upon request)
- Precision Farming (e.g., milk yields)
- Climate (e.g., daily, hourly)
- Health and Wellness (e.g., sperm analysis, health checks)
- Genetics (e.g., EBVs)
- Calculated (e.g., biodiversity, aridity)
Evaluation and Awards
- The first three winning teams will receive a cash award of €3,000, €2,000 and €1,000, respectively.
- One member of each winning team will be invited to present their model and results at our Event at Surrey University (Guildford, UK) which will take place in December 2024.
- Selected abstracts will be published free of charge on the vHive website.
- Invitation to be the first author on the joint competition publication, if desired.
The winner will be determined based on:
Novelty of the research question or AI application
Utility– usefulness of the results/analysis for animal health productivity etc.
Need– addressing a gap in knowledge, or meeting a health or industry need
Instructional Webinar Proceedings
In case you missed our launch webinar on the 2nd October. Here is the recording introducing the organisers, summarising the Hackathon, and providing a tutorial on how to use the LEO data:
Registration and Submission
First, please complete the Hackathon Registration Form. Each member of your team MUST be included on this form to participate in the competition. Teams of one are permitted.
Then, download and use the template linked in the button below to create a Project Summary Document for submission.
Participant requirements:
- Any individual or team can participate. One individual must be designated as team leader and all team members must register for the instructive webinar.
- Each individual can be involved in only one team/entry.
- Companies or individuals are encouraged to participate.
- Competition participants agree to have their models and results included in a possible joint publication on which they will be first/co-authors if they desire.
- If you have a model that is proprietary, or you would prefer not to share widely, you are still encouraged to participate but will not be eligible to win, or be an author on any joint publication. All data you submit to “AI Hackathon for Livestock” (not including code) is considered public and may therefore be used in the domain.
- The organisers reserve the right to disqualify any participants from the competition at any time.
- A member of the winning team is expected and encouraged to present their results at the December 2024 Event in Surrey, UK.
- Competition participants agree that all submissions are original ideas, that adhere to scientific principles, and do not infringe on any copyright or intellectual property rights.
Modelling requirements
- Any modelling strategy is permitted.
- All data which were publicly and freely available are permitted and encouraged for use in model development. Data that became publicly available after February 1, 2024, is not permissible.
Evaluation
- Participants will be provided with the LEO data for use in developing and training their models.
- By the submission deadline participants will submit:
- A summary of the results from their proposed project in a two-page Word document (template will be provided).
- While not included in the results submission, participants should be prepared to provide their well-organized code and any data they have used upon request of the organizers. The organizers must be provided sufficient information to re-run the model themselves and validate the results if desired. Participants who desire to keep their model proprietary will not be required to provide this information, but they will not be considered for the awards.
Upload your Proposal
Please submit your proposal here. Only proposals submitted using this form will be counted.
Format of the proposal is specified in the word template provided.
Files will only be accepted in PDF format.
Please save and upload Project Summary files using the format Team_Name_Date, and any supplementary files as Team_Name_Supplement.
If you are an individual rather than a team please submit the form using the format Full_Name_Date.
Do I need to be part of a team, or can I work independently?
You can work as either a team or you can work independently it is entirely up to you.
Can I participate in multiple projects?
No, each participant can join and submit one and only one project
Can I join the competition after the 2nd of October?
Yes, participants can join the competition at any time till the closing day (14th November 2024)
Can I join the competition even if I didn’t register for the online webinar on the 2nd of October?
Yes, participants can join the competition at any time till the closing day (14th November 2024), even if they didn’t register for the online webinar on the 2nd of October.
Are there any restrictions applied on the number of members forming a team?
No, there is no restriction on the number of members forming a team
I am not working in academia, can I join the competition?
Yes, participants outside academia can join the competition.
Can I present a project that is not included in the specific proposals?
Yes, any project idea is welcome
I have difficulties on downloading the data
If you face any difficulties on downloading the data, please contact us using the contact form at the bottom of this webpage
I have difficulties on selecting the appropriate data for my project
If you face any difficulties regarding the data, please contact us using the contact form at the bottom of this webpage
During the evaluation process, can I vote for my project?
No, you are not allowed to vote for your project
How can I find the codes for species and breeds of interest?
Code explanation is available on the LEO github repository. Species codes are available here https://opendata.leo-italy.eu/lodview/controlled-vocabulary/livestock-species.html and breed codes here https://opendata.leo-italy.eu/lodview/controlled-vocabulary/breed.html; For example, if you are interested for the Italian Holstein Friesian dairy cattle, bovines are coded as C4 https://opendata.leo-italy.eu/lodview/controlled-vocabulary/livestock-species/C4.html and the Italian Holstein Friesian is coded as breed 02 https://opendata.leo-italy.eu/lodview/controlled-vocabulary/breed/breed02.html
How can I efficiently query the LEO data?
There are various examples of Sparql here https://opendata.leo-italy.eu/portale/sparql
Contact Us
If you have any questions about the Hackathon, please contact us by filling out this form. One of the team will reply to you shortly.
