Notice: Undefined offset: 1 in /var/www/indjst.org/article-detail-page.php on line 103
A Beta Regression Model for Himalayan Medicinal Plant Disease Prediction
 
  • P-ISSN 0974-6846 E-ISSN 0974-5645

Indian Journal of Science and Technology

Article

Indian Journal of Science and Technology

Year: 2014, Volume: 7, Issue: 6, Pages: 776–780

Original Article

A Beta Regression Model for Himalayan Medicinal Plant Disease Prediction

Abstract

Picrorhiza kurrooa plants were grown in greenhouse and was tested for powdery mildew disease. The result of temperature and wetness duration on P.kurrooa was studied under controlled-environment to develop a disease prediction model for controlling infection.Plants were kept at a temperature ranging from 31°C to 37°C with Relative Humidity (RH) more than 90% was maintained in greenhouse and it is measured in the terms of wetness duration (in hours) from 5 to 40hr.To determine the relationship between infection index, temperature and wetness duration data for P. kurrooa were analyzed by nonlinear regression model. Beta model was used to provide the best fit to the data for modeling. Infection index on plant increased with increasing wetness duration at optimum temperature. Minimum and maximum temperatures for infection were around 31 and 37°C, respectively. At 35.5°C maximum infection was recorded, and a minimum duration of wetness that is required for germination of the fungus was 5hr. Highest infection index 0.95 was noticed at 35.5°C. Temperature and duration of wetness were recorded for each event and used in the model equation to calculate disease infection index. Regression coefficient (R2 ) between observed infection index and predicted infection index was 0.8103 and coefficient of determination (R) was 0.9. It indicates that the model could reliably predict the disease infection index over a wide range of temperatures and wetness durations. 

Keywords: Beta Regression Model, Picrorhiza kurrooa, Powdery Mildew Prediction System

DON'T MISS OUT!

Subscribe now for latest articles and news.