Best Paper at the International Conference on Case-Based Reasoning (2019)
Eoin Kenny, a Ph.D. candidate in computer science at the Insight Centre for Data Analytics, recently received an award for best paper at the International Conference on Case-Based Reasoning (2019).
The work emanated from the VistaMilk research project involving Insight, Teagasc, Tyndall, TSSG and ICBF. The main aim of this initiative is to promote smart agriculture in Ireland which both increases profits and improves sustainability of farming practices. We used an artificial intelligence technique called case-based reasoning (CBR) to augment an existing decision support application called PastureBase Ireland (PBI) used by Irish dairy farmers. The system predicts grass growth on Irish dairy farms in advance which allows greater utilisation of grass as a food source for cattle and lessens the carbon costs of importing meal. Continual meetings over several months between Insight and Teagasc were a crucial part of the research that allowed us access to the necessary materials and domain knowledge to design a workable system.
The primary novelties of our work include (1) a novel way to clean datasets which is coined “Bayesian Case-Exclusion”, (2) showing that a simple CBR system can both deliver sufficient accuracy and provide human understandable explanations in this domain, and (3) demonstrating that a practical application of Bayesian statistics can help the system adapt to climate change and improve accuracy.
The paper marks arguably the most notable collaboration between these institutes to date. Ideally, the next step is user testing of the system on real farmers which can be launched on the PBI application. For future work, we hopeto (1) improve the Bayesian-Case Exclusion idea by including more accurate weather data, (2) collaborate with the people at Teagasc by using their mechanical model for grass growth (called MoSt) to assist the CBR system, and (3) generate missing datapoints using a relatively new artificial intelligence technique known as generative adversarial networks.
Reference for Best Paper
Kenny, E.M., Ruelle, E., Geoghegan, A., Shalloo, L., O’Leary, M., O’Donovan, M. and Keane, M.T., 2019, September. Predicting Grass Growth for Sustainable Dairy Farming: A CBR System UsingBayesian Case-Exclusion and Post-Hoc, Personalized Explanation-by-Example (XAI). In International Conference on Case-Based Reasoning (pp. 172-187). Springer, Cham. This publication has emanated from research conducted with the financial support of (i) Science Foundation Ireland (SFI) to the Insight Centre for Data Analytics under Grant Number 12/RC/2289 and (ii) SFI and the Department of Agriculture, Food and Marine on behalf of the Government of Ireland to the VistaMilk SFI Research Centre under
Twitter: Eoin @EoinKNNy, Supervisor @keanema