MODELING USING ARTIFICIAL NEURAL NETWORKS APPLIED TO INFERENCE OF THE TOMATO VARIETY MOST FAVORABLE TO DIFFERENT THERMAL PROCESSING

• Grant number: 23/06632-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2023
• End date: August 31, 2024
• Field of knowledge: Agronomical Sciences - Food Science and Technology - Food Technology
• Principal Investigator: Angela Vacaro de Souza
• Grantee: Vinícius Canato
• Host Institution: Faculdade de Ciências e Engenharia. Universidade Estadual Paulista (UNESP). Campus de Tupã. Tupã , SP, Brazil

Abstract

The objective of the study is to predict via artificial neural network the tomato variety(ies) best suited to the thermal processing studied from the evaluation of quality parameters compared to fresh fruit. Five tomato varieties produced in the city of Tupã-SP will be investigated regarding physical-chemical quality parameters - moisture, pH, titratable acidity (TA), soluble solids (SS), 'Ratio' (SS/TA ratio) - and biochemical parameters - phenolic compounds, flavonoids, carotenoids, lipid peroxidation, hydrogen peroxide, total sugars, lycopene, ²-carotene and chlorophylls a and b. The fruits will be harvested when they reach a bright red color and transported to the Biology Laboratory of the Science and Engineering College of Unesp in Tupã to be selected and evaluated in natura and after being submitted to cooking to obtain sauce, dehydrated and freeze-dried. At the end of the experiment, it is expected that the application of the artificial neural network will be able to predict the most appropriate varieties for the processes used, and that the results obtained will bring elucidations about the influence of thermal processing on the quantification of compounds mentioned and antioxidant activities of the varieties studied.