[Paper] Packaging Design Analysis by Predicting User Preference and Semantic Attribute

Abstract

Packaging design has a pronounced effect on consumer purchase behavior and can be a critical factor in marketing. Despite the importance, there are very few studies that have investigated optimal designs. In this work, in order to analyze packaging designs and support designing processes, we propose a deep learning based method with ensemble learning to predict user preference for packaging design. For qualitative analysis, we visualize the feature maps from the prediction model. Moreover, we predict the mentioned frequencies of semantic attributes in the questionnaires, which represent impressions that users have. The experimental results suggest that in terms of user preference prediction, the proposed model achieves a correlation coefficient of 0.652 between the ground truth user preference scores and the predicted values. As for the semantic attributes prediction, the highest correlation coefficient reaches 0.75. Also the visualization successfully indicates key elements in designs.

Robust robot image classification toward cyber-physical system-based closed-loop package design evaluation

“…Then, the estimated noisy transition matrix is used for neural network training. The estimation of the noisy transition matrix is based on the anchor points (Yu et al, 2018 ; Xia et al, 2022 ), which are defined as follows.…”

“…However, the above methods assume that the used data for training are clean. However, according to the results of Xia et al ( 2022 ), the classification based on esthetic assessment is different from the classification from the viewpoint of the consumers. The labels that reflect the viewpoint of the consumers can be obtained from the market survey.…”

“…Definitely, the performance will be good with deep learning models if the labels are correctly given. However, there are lots of unclear or incorrect labels in the data set of market surveys (Xia et al, 2022 ), which can be regarded as noisy labels. Noisy labels can cause severe over-fitting issues in deep learning models.…”