20160914 dcu ei_the_datas_project_dr._andrew_mc_carren
TRANSCRIPT
} PhD in statistics
} Worked for a number of Multi-nationals
} Recommenced Family business
} Sold it in 2014
} Now datas.
} Current state of availability of agricultural data –multiple, disparate sources
} Real Time Analytics are not available
} Over-reliance on gut instinct
} Results coming to late.
} Lifting the mist, creating knowledge
datas.
“…organizations have no guide to what data they actually have in the present and how to find it.”
Tom Davenport, professor at Babson College, research fellow at the MIT Center for Digital Business, senior adviser to Deloitte Analytics.
“I have so many reports to review, with endless amounts of data displayed. Its nearly impossible to optimise my thought process and give the right guidance to business”
MD of the Kepak Meat division.
datas.
Stage 11. Problem recognition2. Review of previous findings
Stage 23. Modelling4. Data Collection5. Data Analysis
Stage 36. Results presentation
© DATAS @DCU
datas.
WWWClient DB
Data Warehouse
Predictions
ExtractingValidatingTransforming
Business Intelligence
Data visualisation
Data miningPredictive tools
© DATAS @DCU datas.
} Integrating data from multiple web sources: Eurostat, U.S. Department of Agriculture, UN Comtrade
} All features relevant to Agriculture– trade, price, production…
} Designing unique new features and predictive algorithms
} Creating useful, actionable knowledge.
datas.
} Software to solve a specific problem or answer a question
} Faster decision making
} Give users more power to determine their future
© DATAS @DCU
datas.
} Traditional Time Series approach◦ Garch, Egarch, structural equation modelling◦ Frequentist approach, Bayesian Estimation
} Machine Learning◦ Neural Networks, Deep Learning
} Discrete Wavelet Transforms
} Bayesian Networks
datas.