Online Program Home
My Program

Abstract Details

Activity Number: 353
Type: Contributed
Date/Time: Tuesday, August 2, 2016 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistical Learning and Data Science
Abstract #319173
Title: Models for Understanding and Predicting Consumer Perception of Radiance
Author(s): Supriya A. K. Satwah* and Anthony Cece and Robert Velthuizen
Companies: Unilever and Unilever and Unilever
Keywords: Data modelling ; Regression -Partial Least Square ; Bayesian Nets ; Perceived radiance ; Consumer Perception ; Spots

Radiance is a hot topic throughout the cosmetics and personal care industries. We have focused on understanding Perceived Aging, Healthy Skin appearance for both Body and Face, and to a much lesser extent Radiance and or Glow. Research is to understand and quantify "natural radiance" obtained through photo evaluation via its relationship to quantitative parameters culled from image analysis and color measurements on groups of women ages 18-60 in Thailand and Shanghai. Perceived Radiance/Dullness in both countries appears to be related to color, spots, and texture however the balance of the contribution of each of these appears to be different between the two countries with increased focus on Spots and Texture in China. A variety of analysis techniques were employed to understand the relationships between consumer perception and the quantification of attributes. Bayesian nets were used to gain a sense of the overall structure and relationships in the data. Partitioning methods were used to build trees for understanding key factors related to the dependent variables. The primary model building tool used was Partial Least squares analysis to develop and refine the final models.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association