On the role of shape features in metaphor generation for abstract images

IF 4.8 1区农林科学 Q1 AGRONOMY

Rice Pub Date : 2022-11-29 DOI:10.1109/SCISISIS55246.2022.10002047

Asuka Terai, Natsuki Yamamura, J. Chikazoe, Takaaki Yoshimoto, Norihiro Sadato, K. Jimura

{"title":"On the role of shape features in metaphor generation for abstract images","authors":"Asuka Terai, Natsuki Yamamura, J. Chikazoe, Takaaki Yoshimoto, Norihiro Sadato, K. Jimura","doi":"10.1109/SCISISIS55246.2022.10002047","DOIUrl":null,"url":null,"abstract":"In this study, we examined the role of shape features on metaphor generation for abstract images based on a simulation with a retrained convolutional neural network (CNN), which is in turn based on a pretrained CNN model (AlexNet). A computational experiment was conducted using three types of object recognition models, including a pretrained object recognition model (AlexNet) and recognition models that were retrained to recognize more or fewer shape features using edge-detected or blurred images from the ILSVRC-2012 dataset. A psychological experiment was conducted to collect metaphors that were used to explain the abstract images. The simulation results of the models for the abstract images were compared to examine how well they predicted the concepts used in the metaphors generated for the abstract images. The results of the computational experiment suggest that the model retrained to recognize fewer shape features performed best at predicting the generated metaphors. However, for some abstract images, the model retrained to recognize more shape features performed better. These results suggest that the role of shape features on metaphor generation differs depending on the types of abstract images.","PeriodicalId":21408,"journal":{"name":"Rice","volume":"3 1","pages":"1-5"},"PeriodicalIF":4.8000,"publicationDate":"2022-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Rice","FirstCategoryId":"97","ListUrlMain":"https://doi.org/10.1109/SCISISIS55246.2022.10002047","RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AGRONOMY","Score":null,"Total":0}

引用次数: 0

Abstract

In this study, we examined the role of shape features on metaphor generation for abstract images based on a simulation with a retrained convolutional neural network (CNN), which is in turn based on a pretrained CNN model (AlexNet). A computational experiment was conducted using three types of object recognition models, including a pretrained object recognition model (AlexNet) and recognition models that were retrained to recognize more or fewer shape features using edge-detected or blurred images from the ILSVRC-2012 dataset. A psychological experiment was conducted to collect metaphors that were used to explain the abstract images. The simulation results of the models for the abstract images were compared to examine how well they predicted the concepts used in the metaphors generated for the abstract images. The results of the computational experiment suggest that the model retrained to recognize fewer shape features performed best at predicting the generated metaphors. However, for some abstract images, the model retrained to recognize more shape features performed better. These results suggest that the role of shape features on metaphor generation differs depending on the types of abstract images.

查看原文本刊更多论文

论形状特征在抽象意象隐喻生成中的作用

在这项研究中，我们基于再训练卷积神经网络(CNN)的模拟，研究了形状特征在抽象图像隐喻生成中的作用，而CNN又基于预训练的CNN模型(AlexNet)。使用三种类型的物体识别模型进行了计算实验，包括预训练的物体识别模型(AlexNet)和使用ILSVRC-2012数据集中的边缘检测或模糊图像进行再训练以识别更多或更少形状特征的识别模型。进行了一项心理学实验，收集用来解释抽象图像的隐喻。对抽象图像模型的模拟结果进行比较，以检验它们对抽象图像生成的隐喻中使用的概念的预测程度。计算实验结果表明，经过再训练识别较少形状特征的模型在预测生成的隐喻方面表现最好。然而，对于一些抽象图像，重新训练的模型识别更多的形状特征表现更好。这些结果表明，形状特征在隐喻生成中的作用因抽象图像类型的不同而不同。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Rice AGRONOMY-

CiteScore

10.10

自引率

3.60%

发文量

审稿时长

>12 weeks

期刊介绍： Rice aims to fill a glaring void in basic and applied plant science journal publishing. This journal is the world''s only high-quality serial publication for reporting current advances in rice genetics, structural and functional genomics, comparative genomics, molecular biology and physiology, molecular breeding and comparative biology. Rice welcomes review articles and original papers in all of the aforementioned areas and serves as the primary source of newly published information for researchers and students in rice and related research.