What and How long: Prediction of Mobile App Engagement

User engagement is crucial to the long-term success of a mobile app. Several metrics, such as dwell time, have been used for measuring user engagement. However, how to effectively predict user engagement in the context of mobile apps is still an open research question. For example, do the mobile usage contexts (e.g., time of day) in which users access mobile apps impact their dwell time? Answers to such questions could help mobile operating system and publishers to optimize advertising and service placement. In this article, we first conduct an empirical study for assessing how user characteristics, temporal features, and the short/long-term contexts contribute to gains in predicting users’ app dwell time on the population level. The comprehensive analysis is conducted on large app usage logs collected through a mobile advertising company. The dataset covers more than 12K anonymous users and 1.3 million log events. Based on the analysis, we further investigate a novel mobile app engagement prediction problem—can we predict simultaneously what app the user will use next and how long he/she will stay on that app? We propose several strategies for this joint prediction problem and demonstrate that our model can improve the performance significantly when compared with the state-of-the-art baselines. Our work can help mobile system developers in designing a better and more engagement-aware mobile app user experience.

Disparities in Oral Nutritional Supplement Usage and Dispensing Patterns across Primary Care in Ireland: ONSPres Project

When treating malnutrition, oral nutritional supplements (ONSs) are advised when optimising the diet is insufficient; however, ONS usage and user characteristics have not been previously analysed. A retrospective secondary analysis was performed on dispensed pharmacy claim data for 14,282 anonymised adult patients in primary care in Ireland in 2018. Patient sex, age, residential status, ONS volume (units) and ONS cost (EUR) were analysed. The categories of ‘Moderate’ (<75th centile), ‘High’ (75th–89th centile) and ‘Very High’ ONS users (≥90th centile) were created. The analyses among groups utilised t-tests, Mann–Whitney U tests and chi-squared tests. This cohort was 58.2% female, median age was 76 years, with 18.7% in residential care. The most frequently dispensed ONS type was very-high-energy sip feeds (45% of cohort). Younger males were dispensed more ONSs than females (<65 years: median units, 136 vs. 90; p < 0.01). Patients living independently were dispensed half the volume of those in residential care (112 vs. 240 units; p < 0.01). ‘Moderate’ ONS users were dispensed a yearly median of 84 ONS units (median cost, EUR 153), ‘High’ users were dispensed 420 units (EUR 806) and ‘Very High’ users 892 yearly units (EUR 2402; p < 0.01). Further analyses should focus on elucidating the reasons for high ONS usage in residential care patients and younger males.

Introducing the hormonal Intrauterine Device in Madagascar, Nigeria, and Zambia: results from a pilot study

Background The hormonal Intrauterine Device (IUD) is a highly effective contraceptive option growing in popularity and availability in many countries. The hormonal IUD has been shown to have high rates of satisfaction and continuation among users in high-income countries. The study aims to understand the profiles of clients who choose the hormonal IUD in low- and middle-income countries (LMICs) and describe their continuation and satisfaction with the method after 12 months of use. Methods A prospective longitudinal study of hormonal IUD acceptors was conducted across three countries—Madagascar, Nigeria, and Zambia—where the hormonal IUD had been introduced in a pilot setting within the of a broad mix of available methods. Women were interviewed at baseline immediately following their voluntary hormonal IUD insertion, and again 3 and 12 months following provision of the method. A descriptive analysis of user characteristics and satisfaction with the method was conducted on an analytic sample of women who completed baseline, 3-month, and 12-month follow-up questionnaires. Kaplan–Meier time-to-event models were used to estimate the cumulative probability of method continuation rates up to 12 months post-insertion. Results Each country had a unique demographic profile of hormonal IUD users with different method-use histories. Across all three countries, women reported high rates of satisfaction with the hormonal IUD (67–100%) and high rates of continuation at the 12-month mark (82–90%). Conclusions Rates of satisfaction and continuation among hormonal IUD users in the study suggest that expanding method choice with the hormonal IUD would provide a highly effective, long-acting method desirable to many different population segments, including those with high unmet need.

MUFFLE: Multi-Modal Fake News Influence Estimator on Twitter

To alleviate the impact of fake news on our society, predicting the popularity of fake news posts on social media is a crucial problem worthy of study. However, most related studies on fake news emphasize detection only. In this paper, we focus on the issue of fake news influence prediction, i.e., inferring how popular a fake news post might become on social platforms. To achieve our goal, we propose a comprehensive framework, MUFFLE, which captures multi-modal dynamics by encoding the representation of news-related social networks, user characteristics, and content in text. The attention mechanism developed in the model can provide explainability for social or psychological analysis. To examine the effectiveness of MUFFLE, we conducted extensive experiments on real-world datasets. The experimental results show that our proposed method outperforms both state-of-the-art methods of popularity prediction and machine-based baselines in top-k NDCG and hit rate. Through the experiments, we also analyze the feature importance for predicting fake news influence via the explainability provided by MUFFLE.

Tracking sentiments toward fat acceptance over a decade on Twitter

The fat acceptance (FA) movement aims to counteract weight stigma and discrimination against individuals who are overweight/obese. We developed a supervised neural network model to classify sentiment toward the FA movement in tweets and identify links between FA sentiment and various Twitter user characteristics. We collected any tweet containing either “fat acceptance” or “#fatacceptance” from 2010–2019 and obtained 48,974 unique tweets. We independently labeled 2000 of them and implemented/trained an Average stochastic gradient descent Weight-Dropped Long Short-Term Memory (AWD-LSTM) neural network that incorporates transfer learning from language modeling to automatically identify each tweet’s stance toward the FA movement. Our model achieved nearly 80% average precision and recall in classifying “supporting” and “opposing” tweets. Applying this model to the complete dataset, we observed that the majority of tweets at the beginning of the last decade supported FA, but sentiment trended downward until 2016, when support was at its lowest. Overall, public sentiment is negative across Twitter. Users who tweet more about FA or use FA-related hashtags are more supportive than general users. Our findings reveal both challenges to and strengths of the modern FA movement, with implications for those who wish to reduce societal weight stigma.

Investigation of the current usage of BIM capabilities by large-sized building contractors in Kenya based on theory of innovation diffusion

Purpose This study aims to focus on the four user characteristics of innovation diffusion (availability, observability and trialability [AOT], simplicity, relative advantage [RA] and interoperability) to observe their influence on building information modelling (BIM) usage. This study focuses on the Kenyan construction industry, specifically the building contractors. Design/methodology/approach This study uses purposive sampling and specifically focusses on active construction sites that met requirements needed for BIM usage to thrive. Data was collected manually using questionnaires (N = 62). Findings This paper contributes to the analysis of the current state of BIM usage by the Kenyan construction industry specifically among building contractors and confirms that Kenya is at the early majority adopters’ stage of diffusion characterised by low BIM usage. In terms of correlation, this study found out that AOT had a strong positive correlation with usage, RA had a moderate positive correlation with usage, simplicity had a weak positive correlation with usage and interoperability had no correlation with usage. Practical implications This study gives a clear trend on BIM usage among building contractors to assist potential BIM users make informed decision. The recommendations in this study can be adopted by any late adopter jurisdiction whose structure of the construction industry is similar to Kenya’s. Originality/value This paper highlights variables that enable or subdue BIM usage. It goes further to localise and contextualise the barriers for deeper understanding of what makes these barriers be major hindrances towards BIM usage and giving practical solutions to these barriers.

Anbar: Collection and analysis of a large scale Urdu language Twitter corpus

The confluence of high performance computing algorithms and large scale high-quality data has led to the availability of cutting edge tools in computational linguistics. However, these state-of-the-art tools are available only for the major languages of the world. The preparation of large scale high-quality corpora for low resource language such as Urdu is a challenging task as it requires huge computational and human resources. In this paper, we build and analyze a large scale Urdu language Twitter corpus Anbar. For this purpose, we collect 106.9 million Urdu tweets posted by 1.69 million users during one year (September 2018-August 2019). Our corpus consists of tweets with a rich vocabulary of 3.8 million unique tokens along with 58K hashtags and 62K URLs. Moreover, it contains 75.9 million (71.0%) retweets and 847K geotagged tweets. Furthermore, we examine Anbar using a variety of metrics like temporal frequency of tweets, vocabulary size, geo-location, user characteristics, and entities distribution. To the best of our knowledge, this is the largest repository of Urdu language tweets for the NLP research community which can be used for Natural Language Understanding (NLU), social analytics, and fake news detection.

A new spreading model in the environment of epidemic-related online rumors

This paper takes COVID-19-related online rumors as the research object, and explores the law of spreading public opinion in social networks. The paper also conducts empirical research on the relationship between rumor spreading, user characteristics and subject interest differences, and analyzes the common influence of individual factors and social environment. In the process of public opinion dissemination, measures that can effectively regulate the dissemination of public opinion are proposed. Based on the susceptible-exposed-infected-recovered (SEIR) model, this paper analyzes the influence of individual differentiation characteristics, friend factors, and time-dependent decline on user status changes. The study found that the user’s environment can affect the spread and popularity of public opinion information, and prolong the survival time of public Controlling the propagation threshold and exit threshold of the platform helps to control the large-scale dissemination of online public opinion. The extinction of public opinion is affected by the decline of time and heat rather than certain probability.

Understanding the Heterogeneity of Human Mobility Patterns: User Characteristics and Modal Preferences

Characterizing individual mobility is critical to understand urban dynamics and to develop high-resolution mobility models. Previously, large-scale trajectory datasets have been used to characterize universal mobility patterns. However, due to the limitations of the underlying datasets, these studies could not investigate how mobility patterns differ over user characteristics among demographic groups. In this study, we analyzed a large-scale Automatic Fare Collection (AFC) dataset of the transit system of Seoul, South Korea and investigated how mobility patterns vary over user characteristics and modal preferences. We identified users’ commuting locations and estimated the statistical distributions required to characterize their spatio-temporal mobility patterns. Our findings show the heterogeneity of mobility patterns across demographic user groups. This result will significantly impact future mobility models based on trajectory datasets.