Olive Fly Infestation Prediction Using Machine Learning Techniques

Environmental Impact on Predicting Olive Fruit Fly Population Using Trap Measurements

“…This fact explains why the J48 algorithm, as shown by the constructed tree in Figure 2, can't classify any instance to the value "5 to 6". Figure 6 shows the comparison of the best classifier presented herein against the best classifier of [13] in terms of recall. It is obvious that the Classification Trees (CT) have a far better recall compared to the SMO algorithm by almost 10%.…”

“…Machine learning techniques have been used to detect oil spills on the surface of the sea by scanning radar images [7], to automatically identify species by sound [8] and to monitor flood protection systems [9]. Machine learning techniques have also been applied in numerous agriculture processes such as the prediction of when a cow should be culled in a dairy herd [10], the estimation of soil moisture [11], the estimation of a cow's oestrus [12] and the prediction of olive fruit fly infestation using information about olive tree health as well as trap measurements [13].…”

“…The aim of this paper is to predict future olive fruit fly trap measurements, and by extension olive fruit fly infestations/outbreaks, using machine learning algorithms. Our approach differentiates from previous work [13] by constructing a feature vector that consists of environmental factors e.g. temperature, instead of the olive tree health, as well as trap measurements.…”

SEAMLESS SOFTWARE DEVELOPMENT FOR SYSTEMS BASED ON BAYESIAN NETWORKS - An Agricultural Pest Control System Example

“…Once the BN structure has been established, the probabilities are estimated (QE) based on a database of cases, completing the construction of the BN model. This expert modelling process has been successfully applied to determine the need of applying a treatment for the olive's fly (dacus olae) (Sagrado and Águila, 2007). The SSS workflow produces the software model that has to be designed, coded and debugged.…”