Predicting consumer ad preferences using physiological monitoring and AI

José Alexandre Lobo Marques; Andreia C. Neto; Susana C.  Silva; Enrique Bigne

doi:10.34632/b865755b-88a2-4aeb-a0fb-7ba0df58a1e9

Predicting consumer ad preferences using physiological monitoring and AI

José Alexandre Lobo Marques
, Andreia C. Neto
, Susana C. Silva
, Enrique Bigne

Research output: Other contribution

Abstract

This policy paper explores how combining neurophysiological tools—Electrodermal Activity (EDA) and Facial Expression Analysis (FEA)—with machine learning (ML) enhances the prediction of consumer preferences in advertising, addressing the biases of traditional self-report methods. Analyzing responses from 37 participants to various cosmetic ads revealed that emotions like joy and disgust significantly influenced ad preference, with the Random Forest ML model achieving high predictive accuracy. Explainable AI (XAI) identified key features such as attention and engagement, offering marketers actionable insights. The findings suggest that integrating neurophysiological data with AI can improve advertising strategies, targeting, and consumer engagement.

Original language	English
Type	Policy Briefs
Publisher	Universidade Católica Portuguesa
Number of pages	4
Place of Publication	Porto
DOIs	https://doi.org/10.34632/b865755b-88a2-4aeb-a0fb-7ba0df58a1e9
Publication status	Published - Nov 2024

Publication series

Name	Policy Briefs
Publisher	Research Center in Management and Economics

Access to Document

10.34632/b865755b-88a2-4aeb-a0fb-7ba0df58a1e9Licence: Unspecified

140022300Final published version, 771 KBLicence: Unspecified

http://hdl.handle.net/10400.14/56900Licence: Unspecified

CEGE: Research Center in Management and Economics
Silva, M. (Full Professor), Sousa, R. (Full Professor), Alves, P. (Assistant Professor), Madsen, A. (Invited Assistant Professor), Lourenço, A. (Associate Professor), Carvalho, A. S. (Associate Professor), Andrade, A. (Assistant Professor), Silva, P. D. (Associate Professor), Vlačić, B. (Invited Assistant Professor), Martins, C. (Invited Assistant Professor), Gomes, C. (Auxiliary Researcher), Gevrek, Z. E. (Invited Assistant Professor), Oliveira, F. G. D. (Associate Professor), Faria, G. (Invited Associate Professor), Marreiros, H. (Scholarship holder), Pinho, J. (Invited Assistant Professor), Machado, J. (Associate Professor), Araújo, J. (Auxiliary Researcher), Pinto, J. (Assistant Professor), Rego, A. (Full Professor), Gaspar, J. M. (Auxiliary Researcher), Corbo, L. (Invited Assistant Professor), Costa, L. (Associate Professor), Fernandes, L. (Assistant Professor), Sottomayor, M. (Assistant Professor), Martins, N. (Full Professor), Hernández-Marrero, P. (Auxiliary Researcher), Gonçalves, R. (Associate Professor), Ribeiro, R. (PI), Coelho, S. L. (Assistant Teacher), Pereira, S. M. (Auxiliary Researcher), Silva, S. (Associate Professor), Rodrigues, V. (Associate Professor), Sotiros, D. G. (Auxiliary Researcher), Valverde, C. (Assistant Professor), Leitão, A. (Assistant Professor), Julião, J. (Assistant Professor), Tavares, M. (Assistant Professor), Lages, C. R. (Auxiliary Researcher), Elmashhara, M. G. (Auxiliary Researcher) & Teymourifar, A. (Auxiliary Researcher)
1/01/20 → 31/12/25
Project: Research

Cite this

@misc{b865755b88a24aeba0fb7ba0df58a1e9,

title = "Predicting consumer ad preferences using physiological monitoring and AI",

abstract = "This policy paper explores how combining neurophysiological tools—Electrodermal Activity (EDA) and Facial Expression Analysis (FEA)—with machine learning (ML) enhances the prediction of consumer preferences in advertising, addressing the biases of traditional self-report methods. Analyzing responses from 37 participants to various cosmetic ads revealed that emotions like joy and disgust significantly influenced ad preference, with the Random Forest ML model achieving high predictive accuracy. Explainable AI (XAI) identified key features such as attention and engagement, offering marketers actionable insights. The findings suggest that integrating neurophysiological data with AI can improve advertising strategies, targeting, and consumer engagement.",

author = "Marques, \{Jos{\'e} Alexandre Lobo\} and Neto, \{Andreia C.\} and Silva, \{Susana C.\} and Enrique Bigne",

year = "2024",

month = nov,

doi = "10.34632/b865755b-88a2-4aeb-a0fb-7ba0df58a1e9",

language = "English",

series = "Policy Briefs",

publisher = "Universidade Cat{\'o}lica Portuguesa",

address = "Portugal",

type = "Other",

}

TY - GEN

T1 - Predicting consumer ad preferences using physiological monitoring and AI

AU - Marques, José Alexandre Lobo

AU - Neto, Andreia C.

AU - Silva, Susana C.

AU - Bigne, Enrique

PY - 2024/11

Y1 - 2024/11

N2 - This policy paper explores how combining neurophysiological tools—Electrodermal Activity (EDA) and Facial Expression Analysis (FEA)—with machine learning (ML) enhances the prediction of consumer preferences in advertising, addressing the biases of traditional self-report methods. Analyzing responses from 37 participants to various cosmetic ads revealed that emotions like joy and disgust significantly influenced ad preference, with the Random Forest ML model achieving high predictive accuracy. Explainable AI (XAI) identified key features such as attention and engagement, offering marketers actionable insights. The findings suggest that integrating neurophysiological data with AI can improve advertising strategies, targeting, and consumer engagement.

AB - This policy paper explores how combining neurophysiological tools—Electrodermal Activity (EDA) and Facial Expression Analysis (FEA)—with machine learning (ML) enhances the prediction of consumer preferences in advertising, addressing the biases of traditional self-report methods. Analyzing responses from 37 participants to various cosmetic ads revealed that emotions like joy and disgust significantly influenced ad preference, with the Random Forest ML model achieving high predictive accuracy. Explainable AI (XAI) identified key features such as attention and engagement, offering marketers actionable insights. The findings suggest that integrating neurophysiological data with AI can improve advertising strategies, targeting, and consumer engagement.

U2 - 10.34632/b865755b-88a2-4aeb-a0fb-7ba0df58a1e9

DO - 10.34632/b865755b-88a2-4aeb-a0fb-7ba0df58a1e9

M3 - Other contribution

T3 - Policy Briefs

PB - Universidade Católica Portuguesa

CY - Porto

ER -

Predicting consumer ad preferences using physiological monitoring and AI

Abstract

Publication series

Access to Document

Fingerprint

Projects

CEGE: Research Center in Management and Economics

Cite this