Investigating food pairing hypothesis based on deep learning: Case of Algerian cuisine

IF 3.2 2区农林科学 Q2 FOOD SCIENCE & TECHNOLOGY

International Journal of Gastronomy and Food Science Pub Date : 2025-01-04 DOI:10.1016/j.ijgfs.2024.101098

Siham Bacha , Mohand Saïd Allili , Tarek Kerbedj , Racha Chahboub

{"title":"Investigating food pairing hypothesis based on deep learning: Case of Algerian cuisine","authors":"Siham Bacha , Mohand Saïd Allili , Tarek Kerbedj , Racha Chahboub","doi":"10.1016/j.ijgfs.2024.101098","DOIUrl":null,"url":null,"abstract":"<div><div>Traditional cuisine is considered a core element of cultural identity. The choice of food can often be influenced by identity, culture, and geography. This work investigates the traditional Algerian cuisine by exploring the food pairing hypothesis, which stipulates that combined ingredients with common flavor compounds taste better than their counterpart. To gain insight into the ingredients compounds found in this cuisine, we analyze their characteristics using spectral clustering. Then, we propose a model based on LSTMs to test the food pairing hypothesis in the Algerian cuisine on a collected corpus. Our research shows that the Algerian cuisine has a negative food pairing tendency, which is consistent with the South European cuisine, suggesting broader regional culinary patterns. To the best of our knowledge, this is the first study to investigate the FPH in Algerian cuisine, contributing to a deeper understanding of the food pairing tendencies specific to this region and offering a comparative perspective with neighboring Mediterranean cuisines.</div></div>","PeriodicalId":48594,"journal":{"name":"International Journal of Gastronomy and Food Science","volume":"39 ","pages":"Article 101098"},"PeriodicalIF":3.2000,"publicationDate":"2025-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Gastronomy and Food Science","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1878450X24002312","RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"FOOD SCIENCE & TECHNOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Traditional cuisine is considered a core element of cultural identity. The choice of food can often be influenced by identity, culture, and geography. This work investigates the traditional Algerian cuisine by exploring the food pairing hypothesis, which stipulates that combined ingredients with common flavor compounds taste better than their counterpart. To gain insight into the ingredients compounds found in this cuisine, we analyze their characteristics using spectral clustering. Then, we propose a model based on LSTMs to test the food pairing hypothesis in the Algerian cuisine on a collected corpus. Our research shows that the Algerian cuisine has a negative food pairing tendency, which is consistent with the South European cuisine, suggesting broader regional culinary patterns. To the best of our knowledge, this is the first study to investigate the FPH in Algerian cuisine, contributing to a deeper understanding of the food pairing tendencies specific to this region and offering a comparative perspective with neighboring Mediterranean cuisines.

查看原文本刊更多论文

基于深度学习的食物配对假说研究：以阿尔及利亚烹饪为例

传统烹饪被认为是文化认同的核心要素。食物的选择通常会受到身份、文化和地理的影响。这项工作通过探索食物配对假说来研究传统的阿尔及利亚烹饪，该假说规定混合了共同风味化合物的食材比它们的对应物味道更好。为了深入了解这种烹饪中发现的成分化合物，我们使用光谱聚类分析它们的特征。然后，我们提出了一个基于lstm的模型，在收集的语料库上测试阿尔及利亚烹饪中的食物配对假设。我们的研究表明，阿尔及利亚菜有消极的食物搭配倾向，这与南欧菜一致，表明更广泛的区域烹饪模式。据我们所知，这是第一个调查阿尔及利亚美食中FPH的研究，有助于更深入地了解该地区特有的食物配对趋势，并提供与邻近地中海美食的比较视角。

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

International Journal of Gastronomy and Food Science Social Sciences-Cultural Studies

CiteScore

5.30

自引率

10.50%

发文量

170

审稿时长

45 days

期刊介绍： International Journal of Gastronomy and Food Science is a peer-reviewed journal that explicitly focuses on the interface of food science and gastronomy. Articles focusing only on food science will not be considered. This journal equally encourages both scientists and chefs to publish original scientific papers, review articles and original culinary works. We seek articles with clear evidence of this interaction. From a scientific perspective, this publication aims to become the home for research from the whole community of food science and gastronomy. IJGFS explores all aspects related to the growing field of the interaction of gastronomy and food science, in areas such as food chemistry, food technology and culinary techniques, food microbiology, genetics, sensory science, neuroscience, psychology, culinary concepts, culinary trends, and gastronomic experience (all the elements that contribute to the appreciation and enjoyment of the meal. Also relevant is research on science-based educational programs in gastronomy, anthropology, gastronomic history and food sociology. All these areas of knowledge are crucial to gastronomy, as they contribute to a better understanding of this broad term and its practical implications for science and society.