Research Article
Effectiveness of Machine Learning in Predicting Maize Sowing and Harvesting Time
@INPROCEEDINGS{10.4108/eai.7-12-2021.2315118, author={Srilatha Toomula}, title={Effectiveness of Machine Learning in Predicting Maize Sowing and Harvesting Time}, proceedings={Proceedings of the First International Conference on Combinatorial and Optimization, ICCAP 2021, December 7-8 2021, Chennai, India}, publisher={EAI}, proceedings_a={ICCAP}, year={2021}, month={12}, keywords={ict in agriculture machine learning experimental \& descriptive research design component}, doi={10.4108/eai.7-12-2021.2315118} }- Srilatha Toomula
Year: 2021
Effectiveness of Machine Learning in Predicting Maize Sowing and Harvesting Time
ICCAP
EAI
DOI: 10.4108/eai.7-12-2021.2315118
Abstract
The study aimed to analyse the efficiency of the system made using a machine learning algorithm in predicting the sowing and harvesting time. Further to investigate the opinions of experts and agriculture farmer against the output generated by the system for maize cultivation. The system was designed in such a way to predict sowing time and harvest time using variables: Location, Temperature, Soil Type, Stem Weight, No. of Kernels in an ear, Kernel Weights and Water Content. Results will be displayed in Likert scale whether to sow the maize seeds or not and whether to harvest or not. For this study, opinion-based data was collected from 124 farmers, 113 experts in the agricultural department and compared against 117 data produced by the system. From the analysis, it was understood that there is no significant difference among the opinions for sowing and harvesting of maize. It can be perceived that system opinion matches 97.7% with Farmers Opinion and 96.7% with expert opinion. The estimated R square value is 0.955, meaning the forecasting accuracy of the system is almost 95.5% and the system-generated output has the efficiency equivalent to farmers and experts. For predicting system output using opinion is given by; System Opinion = 0.021 + (0.954×Farmer Opinion) + (0.022×Expert Opinion).
