Using social media to explore regional cuisine preferences in China
| Date | 11 November 2019 |
| Pages | 1098-1114 |
| DOI | https://doi.org/10.1108/OIR-08-2018-0244 |
| Published date | 11 November 2019 |
| Author | Chengzhi Zhang,Zijing Yue,Qingqing Zhou,Shutian Ma,Zi-Ke Zhang |
| Subject Matter | Library & information science,Information behaviour & retrieval,Collection building & management,Bibliometrics,Databases,Information & knowledge management,Information & communications technology,Internet,Records management & preservation,Document management |
Using social media to explore
regional cuisine preferences
in China
Chengzhi Zhang, Zijing Yue, Qingqing Zhou and Shutian Ma
Department of Information Management,
Nanjing University of Science and Technology, Nanjing, China, and
Zi-Ke Zhang
College of Media and International Culture,
Zhejiang University, Hangzhou, China and
Alibaba Research Center for Complexity Sciences,
Hangzhou Normal University, Hanghzou, China
Abstract
Purpose –Food plays an important role in every culture around the world. Recently, cuisine preference
analysis has become a popular research topic. However, most of these studies are conducted through
questionnairesand interviews, whichare highly limited by the time, costand scope of data collection, especially
when facinglarge-scale survey studies.Some researchers have, therefore,attempted to mine cuisine preferences
based on online recipes, while this approach cannot reveal food preference from people’s perspective. Today,
people are sharing what they eat on social media platforms by posting reviews about the meal, reciting the
names of appetizersor entrees, and photographingas well. Such large amountof user-generated contents(UGC)
has potentialto indicate people’s preferencesover different cuisines.Accordingly, the purpose of thispaper is to
explore Chinese cuisine preferences amongonline users of social media.
Design/methodology/approach –Based on both UGC and online recipes, the authors first investigated the
cuisine preference distribution in different regions. Then, dish preference similarity between regions was
calculated and few geographic factors were identified, which might lead to such regional similarity appeared
in our study. By applying hierarchical clustering, the authors clustered regions based on dish preference and
ingredient usage separately.
Findings –Experimental results show that, among 20 types of traditional Chinese cuisines, Sichuan cuisine
is most favored across all regions in China. Geographical proximity is the more closely related to differences
of regional dish preference than climate proximity.
Originality/value –Different from traditional definitions of regions to which cuisine belong, the authors
found new association between region and cuisine based on dish preference from social media and ingredient
usage of dishes. Using social media may overcome problems with using traditional questionnaires, such as
high costs and long cycle for questionnaire design and answering.
Keywords Food culture, Cuisine preferences, Regional similarity, Social media mining
Paper type Research paper
1. Introduction
A cuisine is a style of cooking characterized by distinctive ingredients, techniques and dishes,
and usually associated with a specific culture or geographic region[1]. Currently, researchers
from different disciplines have presented interesting findings based on different kinds of food
data, such as ingredients, recipe texts and so on (Allhoff and Monroe, 2007; Civitello, 2011).
As one of the popular research topics, cuisine preferences can reveal humans’food choices and
preferences which can provide useful suggestions for restaurants in different places. As cuisine
denotes the use of food, such preferences can also reflect the regional food culture. Thus, the
Online Information Review
Vol. 43 No. 7, 2019
pp. 1098-1114
© Emerald PublishingLimited
1468-4527
DOI 10.1108/OIR-08-2018-0244
Received 17 August 2018
Revised 6 December 2018
Accepted 12 October 2019
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
This work is supported by the National Social Science Fund (No. 14BTQ033), Zhejiang Provincial
Natural Science Foundation of China (Grant No. LR18A050001) and Natural Science Foundation of
China (Grant No. 61673151).
1098
OIR
43,7
cuisine preference analysis can help us to understand the regional context of diet and its
relevant culture from the cuisine perspective. Mor eover, there are also some researchers who
investigate about factors that have aninfluence on a region’scuisine. For example, Wahlqvist
and Lee (2007) showed that geographical conditions play a leading role in local food culture and
food diversity. Zhu et al. (2013) found that geographical proximity is the key factor in
determining the similarity of regional cuisines.
So far, most existing studies of cuisine preferences mainly rely on questionnaires and
interviews ( Fildes et al., 2015;Park et al., 2015), which are mature andfeasible. However, these
obtained results willbe affected by the data sampling approaches and sample size. With the
increment of onlinerecipes, researchers have recentlyattempted to mining such data set (Zhu
et al., 2013), which can help to avoid the aforementioned limitations. Since online recipes
cannot reflectreal and dynamic cuisine preferencesof humans, corpus that expressedpeople’s
choices on differentdishes or cuisines is more effective than static recipe contents. Therapid
growth of socialmedia makes it possible to get such effective corpus, namely, user-generated
contents(UGC). As one of the most popularsocial media platforms in China,Sina Weibo[2] has
nearly 313 million active users monthly by the end of 2016[3]. Users can exchange
information, makecomments, and express their opinions about daily life, with popular topics
including food, clothing, housing, and transportation. Hereby, integrating these UGC and
recipe contents can mine cuisine differences effectively. In this paper, we integrated two
corpuses: UGC gathered from Sina Weibo and online Chinese recipes crawled from
MeiShijie[4]. Specifically, online recipes were utilized to gain information about Chinese
cuisines (e.g.Beijing cuisine, Sichuan cuisine,Shanghai cuisine), and each cuisinecontains lots
of specific dishes. Then, we used Sina Weibo streaming application programming interface
(API) to crawl microblogs which containthose dish. Based on these two data sets, this paper
aims to explore the cuisine preferences of users in social media to see if there are any
differences between regions in China, so as to extract new phenomenon or pattern about
regional cuisines and ingredient usages, which may overcome shortcomings of traditional
questionnaire-based methods, and give more real-time and multi-level results.
2. Related work
In this section, related works are divided into two categories: online user surveys via social
media and cuisine preference analysis.
2.1 Online user surveys via social media
Recently, users in social networking sites have rapidly increased in number, which provides
valuable opportunities for researchersto conduct user surveys. Goyal et al. (2015) proposed a
model to analyze and mine the influence of “Like”in social media “Likes.”Experiment on
Facebook data showed that pages or photos suggested to users are more likely to be
recommended basedon the distributions of user topic interest and page topic. Liu et al. (2015)
further examinedthe relationship between emotionalexpressions in Facebook status updates
and subjective well-being. Their findings support understanding of how UGC reflect users’
psychological states. Chung and Zeng (2016) developed a framework for public policy
informatics, and then built a real-time online system for related messages collected, user
sentiment discovery and network analysis from Twitter.Their work has influenced on public
policy decision making, online social community study, and US immigration and border
security trends. Li and Liu (2016) investigated information-based, temporal patterns in Sina
Weibo. They found that the pattern formation of retweeting is related to social contacts of
users, content, and publication time. Hu et al. (2017) proposed a novel text summarization
technique to identify the top-k mostinformative sentences from onlinehotel reviews, so as to
aid travelers in selecting hotels. Wu et al. (2017) conducted a study on understanding and
modeling users’temporal behaviors on SNS platforms, validatingthat users’preferences and
1099
Regional
cuisine
preferences in
China
To continue reading
Request your trial