Web-based intelligent system for predicting apricot yields using artificial neural networks

• Artificial neural networks and the PDCA method for predicting the apricot yield.
• Creation of a Web-based application that displays final research results.
• Future work relates to the successful application of association rule mining.

This paper shows the use of artificial neural networks and the PDCA (Plan, Do, Check, Act) method for predicting the apricot yield per hectare. The goal of the paper is to determine the possibilities for using artificial neural networks to predict the apricot yield per hectare if the following items are used as input parameters: amount of fertilizer, length of shoots, thickness of shoots, beginning of the harvest and fruit mass. The goal of the paper also includes creation of a web-based application that displays final research results, obtained through neural networks. The PDCA method was used in order to ensure the control and continual improvement of the process. The results point to the possibility of successful application of the above mentioned methods, highlighting the limitations, advantages and shortcomings. Future work relates to the successful application of association rule mining in order to detect the relationship between the apricot yield and other parameters.