These three datasets cover Western, Chinese and Japanese food used for food instance counting and segmentation evaluation. This dataset consists of 3 folders, storing images, instance segmentation masks and the splits of training, testing, and validation. This record includes the data and source code for the paper SibNet: Food instance counting and segmentation.

Food computing has recently attracted considerable research attention due to its significance for health risk analysis. In the literature, the majority of research efforts are dedicated to food recognition. Relatively few works are conducted for food counting and segmentation, which are essential for portion size estimation. This paper presents a deep neural network, named SibNet, for simultaneous counting and extraction of food instances from an image. The problem is challenging due to varying size and shape of food as well as arbitrary viewing angle of camera, not to mention that food instances often occlude each other. SibNet is novel for proposal of learning seed map to minimize the overlap between instances. The map facilitates counting and can be completed as an instance segmentation map that depicts the arbitrary shape and size of individual instance under occlusion. To this end, a novel sibling relation sub-network is proposed for pixel connectivity analysis. Along with this paper, three new datasets covering Western, Chinese and Japanese food are also constructed for performance evaluation.

The dataset supplements to the paper Koh, Yumi, Li, Jing, and Xu, Jianhuan, (2025) “Subway, Collaborative Matching, and Innovation.” Review of Economics and Statistics 107:2, 476–493. The full-text is available at https://ink.library.smu.edu.sg/soe_research/2541/.

Data and analysis code for paper PRANA, Gede Artha Azriadi; FORD, Denae; RASTOGI, Ayushi; LO, David; PURANDARE, Rahul; and NAGAPPAN, Nachiappan. Including everyone, everywhere: Understanding opportunities and challenges of geographic gender-inclusion in OSS. (2022). IEEE Transactions on Software Engineering. 48, (9), 3394-3409.https://doi.org/10.1109/TSE.2021.3092813Full Text is available at: https://ink.library.smu.edu.sg/sis_research/7628

Dataset for the research paper "Trait mindfulness is associated with enhanced daily affectivity and cognition independent of daily stressors exposure: Insights from large-scale daily diary studies in the US and Singapore".Trait mindfulness has been linked to various adaptive outcomes, including attenuated affective and cognitive responses to laboratory-induced stress. However, the role of trait mindfulness as a resilience factor against daily stressors exposures is less established. Across 2 studies, multilevel analysis was used to examine the relationships between trait mindfulness and daily affect and cognition, as well as affective and cognitive reactivity to and recovery from everyday stressor exposure. Trait mindfulness was significantly associated with higher daily positive affect in both studies, lower negative affect and cognitive failure, and lower cognitive reactivity to daily stressor exposure in Study 2. However, trait mindfulness did not attenuate cognitive reactivity in Study 1, nor affective reactivity to daily stressor exposure and affective and cognitive recovery from previous-day stressor exposure in both studies. Overall, results suggest that the mechanisms underlying the affective and cognitive buffering effect of trait mindfulness are not stress specific.

This record includes the replication data and code supporting results of a published book chapter Discoverability beyond the library: Search engine optimization (case study) [link to be added]. The case study compares the discoverability of two hosted institutional repository solutions, Digital Commons and Figshare using a randomized controlled experiment. Two randomly selected groups of journal articles were deposited and made open access in institutional repositories hosted on Digital Commons and Figshare respectively. Download count data were collected over 7 months to measure and compare the open access discoverability and search engine visibility of the two platforms. GENERAL INFORMATION This readme file was generated on 2022-07-04 by Dong Danping Author Contact Name: Dong Danping ORCID: 0000-0002-2229-6709 Institution: Singapore Management University Email: danpingzzz@gmail.com Name: Aaron Tay ORCID: 0000-0003-0159-013X Institution: Singapore Management University Email: aarontay@smu.edu.sg Date of data collection: 2021-04-01 to 2021-10-01 SHARING/ACCESS INFORMATION Licenses/restrictions placed on the data: CC-BY-4.0 License Links to publications that cite or use the data: [to be updated] Links to other publicly accessible locations of the data: https://github.com/dpdong19/IR-compare https://doi.org/10.25440/smu.19121768 Recommended citation for this dataset: Dong, D., & Tay, A. (2022). Data and code to compare the discoverability of Digital Commons and Figshare. https://doi.org/10.25440/smu.19121768 DATA & FILE OVERVIEW File List /analysis/2021-11_DataAnalysis_DCvsFig.ipynb This is the Jupyter Notebook containing notes and scripts of statistical analysis for the case study. /analysis/DCvsFigshare-downloads-combined-v1.csv This file contains clean data for analysis containing download stats from Apr to Oct 2021 for both InK(Digital Commons) and RDR(Figshare).

Relationship between files: The Jupyter Notebook and data file should be placed in the same folder for the code to run. Data Dictionary DCvsFigshare-downloads-combined-v1.csv Number of variables: 15 Number of cases/rows: 92 Variable List Identifier: unique ID for each record. Also the URL to access the record. (Note: Figshare records have been unpublished after the study thus no longer accessible) IR: Name of the IR. InK is on Digital Commons and RDR is on Figshare. Title: title of the deposited journal article Column D-J: monthly download count excluding bots downloads from April to October 2021. Total: sum of column D-J, total download count during the study period AugToOct: sum of column H-J from Aug to Oct 2021 GS_avail: whether the record can be found in Google Scholar. uniq_PDF: whether the record provides the only PDF in Google Scholar primary: whether the record is displayed as the primary record in Google Scholar.

Missing data codes: blank METHODOLOGICAL INFORMATION Methods are described in 2021-11_DataAnalysis_DCvsFig.ipynb and published book chapter [link to be added]

Dataset supporting the paper "Motivations matter: moral and health-related motives indirectly relate to differential psychological health indicators among vegetarians".Due to rising popularity of vegetarianism in recent years, research interest has surged in examining the relationship between vegetarianism and psychological health. However, given inconsistent findings in prior research, the answer to whether practicing vegetarianism is associated with better or worse psychological health is still elusive. The present investigation aimed to demonstrate that vegetarians are not homogeneous in terms of psychological experiences, such that it is crucial to consider the motives behind vegetarians’ dietary choice when examining their psychological health. In a survey study with 266 vegetarians and 104 omnivores, it was shown that health vegetarians displayed higher levels of disordered eating as compared to moral vegetarians and omnivores. Mediation analyses further revealed that, among vegetarians, health motivation was positively correlated with disordered eating tendencies, indirectly linking it with poorer psychological health; moral motivation was positively correlated with prosocial behavior, which in turn predicted better psychological health. These findings have implications for understanding the psychological health of vegetarians with different dietary motives and for developing interventions to promote their psychological health.

The dataset supplements research paper HARTANTO, Andree, LEE, Kristine Y. X., CHUA, Yi Jing, QUEK, Frosch Y. X., & MAJEED, Nadyannam M..(2023). Smartphone use and daily cognitive failures: A critical examination using a daily diary approach with objective smartphone measures. British Journal of Psychology, 114(1), 70-85. Full test is available at: https://ink.library.smu.edu.sg/soss_research/3667ABSTRACTWhile smartphones have brought many benefits and convenience to users, there is continuing debate regarding their potential negative consequences on everyday cognition such as daily cognitive failures. A few cross‐sectional studies have found positive associations between smartphone use and cognitive failures. However, several research gaps remain, such as the use of a cross‐sectional design, confounds related to stable individual differences, the lack of validity in self‐report measures of smartphone use, memory biases in retrospective self‐reports, and the lack of differentiation between smartphone checking and smartphone screen time. To simultaneously address the aforementioned shortcomings, the current study examined the within‐person associations between various objective indicators of smartphone use and daily cognitive failures using a 7‐day daily diary study. Multilevel modelling revealed that smartphone checking, but not total smartphone screen time, predicted a greater occurrence of daily cognitive failures at the within‐person level. Surprisingly, we also found negative within‐person associations between smartphone screen time for social‐ and tools‐related applications and daily cognitive failures, suggesting that some types of smartphone use may temporarily benefit one's cognitive functioning. This finding demonstrates the importance of studying the specific functions of smartphone use and their differential cognitive consequences, as well as highlights the complex relations between smartphone use and cognition.

Dataset supporting the research paper "COVID-19 stress and cognitive failures in daily life : a multilevel examination of within- and between-persons patterns.The COVID-19 pandemic has claimed an extremely high number of lives worldwide, causing widespread panic and stress. The current research examined whether COVID-19 stress was associated with everyday cognitive failures, using data from a seven-day daily diary study of 253 young adults in Singapore. Multilevel modeling revealed that COVID-19 stress was significantly associated with cognitive failures even after adjusting for demographic factors, both at the within-person and between-persons levels. Specifically, individuals experienced more cognitive failures on days they experienced more COVID-19 stress (as compared to their own average levels of COVID-19 stress), and individuals who experienced more COVID-19 stress overall (as compared to individuals who experienced less COVID-19 stress overall) experienced more cognitive failures in general. While a large body of work has evidenced the detrimental effects of COVID-19 stress on individuals’ well-being, the current findings provide novel insights that these stressors may negatively impact individuals’ cognitive functioning as well.

Dataset supporting the research paper "Does trait self-esteem serve as a protective factor in maintaining daily affective well-being? Multilevel analyses of daily diary studies in the US and Singapore".Research suggests that self-esteem could be a protective factor in stressful or unfavorable situations. However, little research has been done on the buffering role of self-esteem in the context of daily stressors on affective reactivity. Three daily diary studies (of which two were conducted in Singapore and one in the United States) were carried out to examine this relationship. In all three studies, trait self-esteem was measured at baseline. Subsequently, daily assessment was conducted on exposure to daily stressors, and positive and negative affect for seven to eight days. Multilevel modeling showed that trait self-esteem did not moderate the relationship between daily stressor exposure and daily affect. An internal meta-analysis aggregating the findings of all three studies was also consistent with our findings. These findings are contrary to previous literature surrounding the stress-buffering role of self-esteem.

This dataset supplements the research paper CAMPANTE, Filipe R. and CHOR, Davin. Schooling, Political Participation, and the Economy. (2012). Review of Economics and Statistics. 94, (4), 841-859. The full-text is available at https://ink.library.smu.edu.sg/soe_research/1330.It supports a cross-national study investigating how the relationship between individual schooling and political participation varies according to country-level characteristics. Using individual-level survey data across multiple countries, the study finds that political participation is more responsive to schooling in land-abundant countries and less responsive in human capital-abundant countries, even after controlling for political institutions and cultural attitudes. The proposed theoretical framework centres on the opportunity cost of political engagement relative to productive economic activity. Countries with relative land abundance or scarce aggregate human capital are associated with higher levels of political participation and greater responsiveness to schooling, as the foregone production income is lower. Supporting evidence shows that political participation is less responsive to schooling in countries with higher skill premiums and among individuals in skilled occupations within countries. The dataset contributes to understanding the joint determinants of individual political participation and cross-country differences in public investment in education, with implications for political economy and human capital research.

Replication Data for "JUE Insight: Migration, Transportation Infrastructure, and the Spatial Transmission of COVID-19 in China"

This record contains the underlying research data for the publication "Acknowledging individual responsibility while emphasizing social determinants in narratives to promote obesity-reducing public policy: A randomized experiment" and the full-text is available from: https://ink.library.smu.edu.sg/lkcsb_research/4848This study tests whether policy narratives designed to increase support for obesity-reducing public policies should explicitly acknowledge individual responsibility while emphasizing social, physical, and economic (social) determinants of obesity. We use a web-based, randomized experiment with a nationally representative sample of American adults (n = 718) to test hypotheses derived from theory and research on narrative persuasion. Respondents exposed to narratives that acknowledged individual responsibility while emphasizing obesity’s social determinants were less likely to engage in counterargument and felt more empathy for the story’s main character than those exposed to a message that did not acknowledge individual responsibility. Counterarguing and affective empathy fully mediated the relationship between message condition and support for policies to reduce rates of obesity. Failure to acknowledge individual responsibility in narratives emphasizing social determinants of obesity may undermine the persuasiveness of policy narratives. Omitting information about individual responsibility, a strongly-held American value, invites the public to engage in counterargument about the narratives and reduces feelings of empathy for a character that experiences the challenges and benefits of social determinants of obesity.

This dataset comprises a set of Twitter accounts in Singapore that are used for social bot profiling research conducted by the Living Analytics Research Centre (LARC) at Singapore Management University (SMU). Here a bot is defined as a Twitter account that generates contents and/or interacts with other users automatically (at least according to human judgment). In this research, Twitter bots have been categorized into three major types:

Broadcast bot. This bot aims at disseminating information to general audience by providing, e.g., benign links to news, blogs or sites. Such bot is often managed by an organization or a group of people (e.g., bloggers). Consumption bot. The main purpose of this bot is to aggregate contents from various sources and/or provide update services (e.g., horoscope reading, weather update) for personal consumption or use. Spam bot. This type of bots posts malicious contents (e.g., to trick people by hijacking certain account or redirecting them to malicious sites), or promotes harmless but invalid/irrelevant contents aggressively.

This categorization is general enough to cater for new, emerging types of bot (e.g., chatbots can be viewed as a special type of broadcast bots). The dataset was collected from 1 January to 30 April 2014 via the Twitter REST and streaming APIs. Starting from popular seed users (i.e., users having many followers), their follow, retweet, and user mention links were crawled. The data collection proceeds by adding those followers/followees, retweet sources, and mentioned users who state Singapore in their profile location. Using this procedure, a total of 159,724 accounts have been collected. To identify bots, the first step is to check active accounts who tweeted at least 15 times within the month of April 2014. These accounts were then manually checked and labelled, of which 589 bots were found. As many more human users are expected in the Twitter population, the remaining accounts were randomly sampled and manually checked. With this, 1,024 human accounts were identified. In total, this results in 1,613 labelled accounts. Related Publication: R. J. Oentaryo, A. Murdopo, P. K. Prasetyo, and E.-P. Lim. (2016). On profiling bots in social media. Proceedings of the International Conference on Social Informatics (SocInfo’16), 92-109. Bellevue, WA. https://doi.org/10.1007/978-3-319-47880-7_6

Previous research has found that young adults exhibit patterns of poor sleep and that poor sleep is associated with a host of negative psychological consequences. One potential intervention to improve sleep quality is listening to music at bedtime. Although there exist previous works investigating the efficacy of listening to music as a form of sleep aid, these works have been hindered by statistically weak designs, a lack of systematic investigation of critical characteristics of music that may affect its efficacy, and limited generalizability. In light of the limitations in the existing literature, a 15-day randomized cross-over trial was carried out with 62 young adults. Participants completed 5 nights of bedtime listening in each condition (happy music vs. sad music vs. pink noise, which acted as an active control condition) over 3 weeks. Upon awakening each morning, participants rated their subjective sleep quality, current stress, positive and negative affective states, and current life satisfaction. Frequentist and Bayesian multilevel modeling revealed that happy and sad music were both beneficial for subjective sleep quality and next-morning well-being, compared with the pink noise condition; potential nuances are discussed. The current study bears potential practical applications for health-care professionals and lay individuals.

This is the artifact for the ICSE 2021 paper "Smart Contract Security: a Practitioners' Perspective".

What the artifact does: 1) an interview guide that we used for our semi-structured interviews (PDF); Note that the transcripts cannot be part of the artifacts according to the confidentiality and privacy clauses as specified in the IRB. 2) a questionnaire that we used for our online survey (PDF); 3) the valid responses of our online survey (CSV).

For the codebook of interviews and scripts to reproduce the figures in the paper, please follow the link below: https://github.com/zhiyuan-wan/ICSE_2021_smart_contract_security

Fieldwork conducted in August 2023 in Shandong Province, China, investigating forms of agricultural production in several sectors.

Fieldwork sites: 1. Rongcheng City, Weihai 2. Qixia, Yantai 3. Changyi, Weifang 4. Shouguang, Weifang

This is the replication data for the journal article "Subjective social class and life satisfaction: Role of class consistency and identity uncertainty" published in Asian Journal of Social Psychology.

Drawing on a recent perspective that inconsistent class identities can negatively impact psychological outcomes, the current research explored if the relative benefit of higher subjective social class for life satisfaction would differ depending on whether it is consistent with one’s objective social class. In Study 1, across two independent samples from Singapore (N = 1045) and the US (N = 492), higher subjective social class predicted higher life satisfaction more strongly among those high in objective social class, but less strongly among those low in objective social class. In Study 2, these patterns were replicated in another large US sample (N = 1030), and appeared to be driven by lower status-based identity uncertainty (SBIU) linked to higher subjective social class perceptions among high objective social class participants. The role of class-identity perceptions in explaining social class disparities in subjective well-being is discussed.

This is the replication data for the paper "The Intergenerational Mortality Tradeoff of COVID-19 Lockdown Policies" by Lin Ma, Gil Shapira, Damien de Walque, Quy-Toan Do, Jed Friedman, and Andrei Levchenko.

For steps to replicate the results, please refer to the readme file included alongside the data files.

This is the online appendix to the working paper "Labor market implications of Taiwan’s accession to the WTO: A dynamic quantitative analysis", available at https://ink.library.smu.edu.sg/soe_research/2613/