Claim Missing Document
Check
Articles

Found 1 Documents
Search

Deep Learning-Based Consumer Preference Analysis for Batik Packaging Design Using Convolutional Neural Networks Edi Wahyudin; Agus Bahtiar; Irfan Ali; Muhammad Nurhidayat
JISA(Jurnal Informatika dan Sains) Vol 8, No 2 (2025): JISA(Jurnal Informatika dan Sains)
Publisher : Universitas Trilogi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31326/jisa.v8i2.2494

Abstract

Packaging design plays an essential role in shaping consumers’ first impressions of a product, particularly in the batik industry, where cultural meaning and visual identity are deeply intertwined. This study aims to explore how a Convolutional Neural Network (CNN) can help identify consumer preferences toward various batik packaging designs. The dataset consists of real packaging from local SMEs as well as prototype designs created specifically for this research, incorporating variations in motifs, colors, and structural formats. All images were standardized and normalized to ensure consistency before being processed by the CNN model. The architecture consists of several convolutional layers, pooling layers, and fully connected layers, with dropout applied to reduce overfitting. Model training was conducted using the Adam optimizer and the sparse categorical cross-entropy loss function. The results demonstrate that the model achieved a testing accuracy of 92.51%. Stable performance across precision, recall, and F1-score indicates that the CNN effectively captures visual patterns associated with consumer appeal. These findings highlight the potential for batik SMEs to utilize deep learning as a decision-support tool, enabling them to design packaging that is more appealing, relevant, and aligned with contemporary consumer preferences.