The Odd Pack

Around 500 Million farmers around the world earn below $2 a day yet 45 % of all vegetables produced are wasted due to overproduction. These are seemingly contradictory statements. According to United nations if we could reduce the overproduced crop and increase the underproduced crop farmers will gain a better revenue and people will be less hungry. According to Food and agricultural organization main reason why farmers over or under produce is instead of growing crops with precise knowledge on vegetable output of a particular season farmers tend to grow based upon assumptions derived from crop output and prices on immediately prior season.Due to this farmers grow the crops which gave highest revenue in past season.But the problem is lot of farmers in that particular region thinks in the same way and grows the same crop. Resulting in an overproduction and hence an drop in price. To avoid this using the power of Azure machine learning we developed a predictive analytics app which can tell farmers before growing a crop what vegetables will give them best revenue when they are harvested.And this forecast change dynamically according to the amount of crops farmers grow in the current season. We do this using a model which we built using arima model,multi linear regression and a neural network.we use past data like yield,demand,weather data and current seasons seed sales as inputs. Although price forecasting has been done previously for crops like corn and wheat we are the first to do this for vegetables. We tested this along with sri lanka agricultural department on a set of 138 farmers for a period of two agricultural seasons. It resulted in a 35-150 % increase in revenue for the farmers. With farmers feedback we implemented the app in three local languages and have included a feature in which when a farmer snaps a photo of a diseased leaf ,details including disease name and treatment options are shown to the farmer. This is done using azure cognitive services. In addition there's a disease notification system which notifies farmers when plant diseases related to them are spreading in their region and lists preventive action that they can take.

Our team consists of three people coming from different backgrounds coming together to make the world a better place.Our Leader Prasanna Higgoda is a Medical Student in Grigore T Popa University of Medicine and Pharmacy,Iasi, Romania. We have Raluca Paula who's also a medical student who happens to have a background in statistics and therefore well versed in machine learning models and neuroscience .Raluca is a Medical Student at Grigore T Popa University of Medicine and Pharmacy,Iasi, Romania. Our Third Member Nissanka Seneviratne is a Microsoft Student Partner from the island nation of Sri Lanka. He's a software engineering Graduate from Curtin University of Technology , Australia and is currently undergoing postgraduate studies at Sri Lanka Institute of Information Technology and British Computer Society.