Utilization of artificial neural networks in the prediction of the bunches’ weight in banana plants

• Published in: Scientia horticulturae Vol. 155; pp. 24 - 29
• Main Authors: Soares, J.D.R., Pasqual, M., Lacerda, W.S., Silva, S.O., Donato, S.L.R.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 29-05-2013
• Subjects: adventitious shoots; Agriculture; agronomic traits; Artificial neural networks (ANNs); bananas; crop yield; cultivars; field experimentation; fruiting; fruits; Harvest; hybrids; leaves; Mathematic; metrology; neural networks; prediction; tetraploidy; R2; ANN; Mathematic; LOF; MPE; RW; LS; NHB; SBW; NFB; Harvest; NLL; DS; FW; DFR; CV; FBW; Agriculture; Models; Artificial neural networks (ANNs); MSD
• ISSN: 0304-4238; 1879-1018
• DOI: 10.1016/j.scienta.2013.01.026

► Identifying strategies to improve yield prediction can help producers make better management decisions. ► One of the biggest problems in the improvement of banana is the time spent in obtaining a new cultivar. ► RNA can guide the development of more productive cultivars through the prediction of harvest.
Phytotechnical characters observed in field experimental are of phenotypic nature and most of the time its assessment is based only on the experience of the observer. The assessment of the correlations between variables allows the estimation of the changes in a character based on the changes in other characters. This present study estimated the impact of agronomic characters related to the weight of the bunch measured in banana plants. The experiment was a test for uniformity, conducted in Guanambi, Bahia, by using the cultivar Tropical (YB42-21), an AAAB tetraploid hybrid. The vegetative characters evaluated during flowering included plant height, perimeter of the pseudostem, number of offshoots, and number of living leaves. The yield related characters were evaluated during the harvest and included, bunch's weight, number of hands and fruits, weight of the second hand, and length and diameter of the fruit in two production cycles. In the evaluations, each plant was considered as a basic unit (bu). This work described a protocol for prediction of banana yield by using the artificial neural networks (ANNs) method as modeling tool. The computational model ANN was used and the prediction of the weight of the bunch in banana plants cv. Tropical was estimated with precision and efficiency (R2=91%, MPE=1.40 and MSD=2.29).