Online Program Home
My Program

Abstract Details

Activity Number: 230 - Innovative STEAMS Methdology Over STEM
Type: Topic Contributed
Date/Time: Monday, July 29, 2019 : 2:00 PM to 3:50 PM
Sponsor: Quality and Productivity Section
Abstract #301748
Title: STEAMS Applications on Foods Science and Analytics
Author(s): Kaitlyn Zhang* and Mason Chen
Companies: Stanford OHS and Mission San Jose High School, Stanford OHS
Keywords: Chocolate; Antioxidant; Artificial Intelligence; DOE; STEAMS; Overfit

This paper will apply “STEAMS” methodology on Chocolate Science. The science will mainly address how the antioxidants in chocolate help reduce free radical formation. Free radicals, atoms with an odd number of electrons, damage blood vessels when oxidized by LDL which consequently reducing the risk of heart disease (Technology and Engineering). Data was collected on 20+ chocolate ingredient nutrition contents from 60+ different types of chocolate but were missing the Cocoa%. AI Neural Network algorithm was utilized to impute the missing Cocoa%. The hyperbolic tangent activation function was used to create the hidden layer. In order to overcome the Neural over-fit issue, definitive screening design (DSD) DOE technique was designed to optimize the AI Neural algorithm. The optimal Neural setting can improve validation fitness R-Square by more than 20%. Based on the optimized neural model, Chocolate Type and Vitamin C are the highest predictors of estimating Cocoa%. Because fruit is high in Vitamin C, there could be further health benefits as potential 4th Chocolate Type-Fruit Chocolate. However, commercial fruit chocolate adds a lot of sugar and Vitamin C is destroyed after process.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program