Modelling and predicting User Engagement in mobile applications

Data Science ◽  
2020 ◽  
Vol 3 (2) ◽  
pp. 61-77
Author(s):  
Eduardo Barbaro ◽  
Eoin Martino Grua ◽  
Ivano Malavolta ◽  
Mirjana Stercevic ◽  
Esther Weusthof ◽  
...  
Keyword(s):  
Random Forest ◽  
Mobile Applications ◽  
Negative Binomial ◽  
Numerical Models ◽  
Cox Proportional Hazards ◽  
Mobile App ◽  
User Engagement ◽  
Tree Model ◽  
Clustering Model ◽  
Boosted Tree

The mobile ecosystem is dramatically growing towards an unprecedented scale, with an extremely crowded market and fierce competition among app developers. Today, keeping users engaged with a mobile app is key for its success since users can remain active consumers of services and/or producers of new contents. However, users may abandon a mobile app at any time due to various reasons, e.g., the success of competing apps, decrease of interest in the provided services, etc. In this context, predicting when a user may get disengaged from an app is an invaluable resource for developers, creating the opportunity to apply intervention strategies aiming at recovering from disengagement (e.g., sending push notifications with new contents).In this study, we aim at providing evidence that predicting when mobile app users get disengaged is possible with a good level of accuracy. Specifically, we propose, apply, and evaluate a framework to model and predict User Engagement (UE) in mobile applications via different numerical models. The proposed framework is composed of an optimized agglomerative hierarchical clustering model coupled to (i) a Cox proportional hazards, (ii) a negative binomial, (iii) a random forest, and (iv) a boosted-tree model. The proposed framework is empirically validated by means of a year-long observational dataset collected from a real deployment of a waste recycling app. Our results show that in this context the optimized clustering model classifies users adequately and improves UE predictability for all numerical models. Also, the highest levels of prediction accuracy and robustness are obtained by applying either the random forest classifier or the boosted-tree algorithm.

scholarly journals A global association between Covid-19 cases and airborne particulate matter at regional level

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Angelo Solimini ◽  
F. Filipponi ◽  
D. Alunni Fegatelli ◽  
B. Caputo ◽  
C. M. De Marco ◽  
...  
Keyword(s):  
Air Pollution ◽  
Particulate Matter ◽  
Negative Binomial ◽  
Numerical Models ◽  
Regional Level ◽  
Pollution Level ◽  
Observation Data ◽  
Cumulative Number ◽  
Mixed Effect

AbstractEvidences of an association between air pollution and Covid-19 infections are mixed and inconclusive. We conducted an ecological analysis at regional scale of long-term exposure to air-borne particle matter and spread of Covid-19 cases during the first wave of epidemics. Global air pollution and climate data were calculated from satellite earth observation data assimilated into numerical models at 10 km resolution. Main outcome was defined as the cumulative number of cases of Covid-19 in the 14 days following the date when > 10 cumulative cases were reported. Negative binomial mixed effect models were applied to estimate the associations between the outcome and long-term exposure to air pollution at the regional level (PM10, PM2.5), after adjusting for relevant regional and country level covariates and spatial correlation. In total we collected 237,749 Covid-19 cases from 730 regions, 63 countries and 5 continents at May 30, 2020. A 10 μg/m3 increase of pollution level was associated with 8.1% (95% CI 5.4%, 10.5%) and 11.5% (95% CI 7.8%, 14.9%) increases in the number of cases in a 14 days window, for PM2.5 and PM10 respectively. We found an association between Covid-19 cases and air pollution suggestive of a possible causal link among particulate matter levels and incidence of COVID-19.

scholarly journals A Downscaling–Merging Scheme for Improving Daily Spatial Precipitation Estimates Based on Random Forest and Cokriging

2021 ◽  
Vol 13 (11) ◽  
pp. 2040
Author(s):  
Xin Yan ◽  
Hua Chen ◽  
Bingru Tian ◽  
Sheng Sheng ◽  
Jinxing Wang ◽  
...  
Keyword(s):  
High Resolution ◽  
Random Forest ◽  
Spatial Resolution ◽  
Daily Precipitation ◽  
Machine Learning Algorithms ◽  
Precipitation Data ◽  
Large Area ◽  
Tree Model ◽  
Daily Precipitation Data ◽  
Precipitation Estimates

High-spatial-resolution precipitation data are of great significance in many applications, such as ecology, hydrology, and meteorology. Acquiring high-precision and high-resolution precipitation data in a large area is still a great challenge. In this study, a downscaling–merging scheme based on random forest and cokriging is presented to solve this problem. First, the enhanced decision tree model, which is based on random forest from machine learning algorithms, is used to reduce the spatial resolution of satellite daily precipitation data to 0.01°. The downscaled satellite-based daily precipitation is then merged with gauge observations using the cokriging method. The scheme is applied to downscale the Global Precipitation Measurement Mission (GPM) daily precipitation product over the upstream part of the Hanjiang Basin. The experimental results indicate that (1) the downscaling model based on random forest can correctly spatially downscale the GPM daily precipitation data, which retains the accuracy of the original GPM data and greatly improves their spatial details; (2) the GPM precipitation data can be downscaled on the seasonal scale; and (3) the merging method based on cokriging greatly improves the accuracy of the downscaled GPM daily precipitation data. This study provides an efficient scheme for generating high-resolution and high-quality daily precipitation data in a large area.

Initiating Collaboration among Organ Transplant Professionals through Web Portals and Mobile Applications

2013 ◽  
Vol 18 (2) ◽  
Author(s):  
Susan Alexander ◽  
Haley Hoy ◽  
Manil Maskey ◽  
Helen Conover ◽  
John Gamble ◽  
...  
Keyword(s):  
Mobile Applications ◽  
Healthcare Providers ◽  
Organ Transplant ◽  
Pilot Project ◽  
Mobile App ◽  
Lessons Learned ◽  
Ease Of Use ◽  
Future Directions ◽  
Efficient Access ◽  
Health Related

The knowledge base for healthcare providers working in the field of organ transplantation has grown exponentially. However, the field has no centralized ‘space’ dedicated to efficient access and sharing of information.The ease of use and portability of mobile applications (apps) make them ideal for subspecialists working in complex healthcare environments. In this article, the authors review the literature related to healthcare technology; describe the development of health-related technology; present their mobile app pilot project assessing the effects of a collaborative, mobile app based on a freely available content manage framework; and report their findings. They conclude by sharing both lessons learned while completing this project and future directions.

scholarly journals Seeing Things Differently: The use of mobile app interpretation and its effect on visitor experience at two heritage sites in Aotearoa New Zealand

2021 ◽  
Author(s):  
◽  
Jessica Aitken
Keyword(s):  
New Zealand ◽  
Case Studies ◽  
Mobile Applications ◽  
Mobile Apps ◽  
Qualitative Interviews ◽  
Mobile App ◽  
Research Approach ◽  
Visitor Experience ◽  
Heritage Sites ◽  
High Street

<p>The practice of contemporary heritage interpretation has seen increased investment in digital technologies and more recently in mobile applications. However, few empirical studies assess how effective mobile apps are to the visitor experience of heritage sites. What kind of visitor experience do mobile apps provide? How do mobile apps deliver on the aims of interpretation for heritage sites? What types of apps work best? What are the challenges for developers and heritage professionals?  A qualitative research approach is used to examine two case studies; High Street Stories: the life and times of Christchurch’s High Street Precinct and IPENZ Engineering Tours: Wellington Heritage Walking Tour. These case studies ask what kind of experience mobile apps offer as an interpretation tool at these heritage sites. To investigate the topic, email interviews were carried out with heritage professionals and digital developers; together with qualitative interviews with visitors recruited to visit the case study sites using the mobile applications.   This study explores two current examples of mobile app technology in the heritage sector in a New Zealand context. The results of this study aim to augment current literature on the topic of digital interpretation. This study seeks to offer heritage managers and interpreters some key factors to consider when making decisions regarding the methods used to present and interpret heritage sites to visitors and in developing new interpretation and digital strategies that include mobile applications. Although each scenario presents its particular set of considerations and all heritage sites are different, it is hoped these recommendations can be applied and offer working models and strategies.</p>

scholarly journals Adults’ Preferences for Behavior Change Techniques and Engagement Features in a Mobile App to Promote 24-Hour Movement Behaviors: Cross-Sectional Survey Study (Preprint)

2019 ◽  
Author(s):  
Ann DeSmet ◽  
Ilse De Bourdeaudhuij ◽  
Sebastien Chastin ◽  
Geert Crombez ◽  
Ralph Maddison ◽  
...  
Keyword(s):  
Behavior Change ◽  
Repeated Measures ◽  
Self Regulation ◽  
Mobile App ◽  
User Preferences ◽  
Movement Behavior ◽  
Survey Study ◽  
User Engagement ◽  
Behavior Change Techniques ◽  
Self Monitoring

BACKGROUND There is a limited understanding of components that should be included in digital interventions for 24-hour movement behaviors (physical activity [PA], sleep, and sedentary behavior [SB]). For intervention effectiveness, user engagement is important. This can be enhanced by a user-centered design to, for example, explore and integrate user preferences for intervention techniques and features. OBJECTIVE This study aimed to examine adult users’ preferences for techniques and features in mobile apps for 24-hour movement behaviors. METHODS A total of 86 participants (mean age 37.4 years [SD 9.2]; 49/86, 57% female) completed a Web-based survey. Behavior change techniques (BCTs) were based on a validated taxonomy v2 by Abraham and Michie, and engagement features were based on a list extracted from the literature. Behavioral data were collected using Fitbit trackers. Correlations, (repeated measures) analysis of variance, and independent sample <italic>t</italic> tests were used to examine associations and differences between and within users by the type of health domain and users’ behavioral intention and adoption. RESULTS Preferences were generally the highest for information on the health consequences of movement behavior self-monitoring, behavioral feedback, insight into healthy lifestyles, and tips and instructions. Although the same ranking was found for techniques across behaviors, preferences were stronger for all but one BCT for PA in comparison to the other two health behaviors. Although techniques fit user preferences for addressing PA well, supplemental techniques may be able to address preferences for sleep and SB in a better manner. In addition to what is commonly included in apps, sleep apps should consider providing tips for sleep. SB apps may wish to include more self-regulation and goal-setting techniques. Few differences were found by users’ intentions or adoption to change a particular behavior. Apps should provide more self-monitoring (<italic>P</italic>=.03), information on behavior health outcome (<italic>P</italic>=.048), and feedback (<italic>P</italic>=.04) and incorporate social support (<italic>P</italic>=.048) to help those who are further removed from healthy sleep. A virtual coach (<italic>P</italic><.001) and video modeling (<italic>P</italic>=.004) may provide appreciated support to those who are physically less active. PA self-monitoring appealed more to those with an intention to change PA (<italic>P</italic>=.03). Social comparison and support features are not high on users’ agenda and may not be needed from an engagement point of view. Engagement features may not be very relevant for user engagement but should be examined in future research with a less reflective method. CONCLUSIONS The findings of this study provide guidance for the design of digital 24-hour movement behavior interventions. As 24-hour movement guidelines are increasingly being adopted in several countries, our study findings are timely to support the design of interventions to meet these guidelines.

scholarly journals Mobile Apps as Audience-Centered Health Communication Platforms (Preprint)

2020 ◽  
Author(s):  
Michael Mackert ◽  
Dorothy Mandell ◽  
Erin Donovan ◽  
Lorraine Walker ◽  
Mike Garcia ◽  
...  
Keyword(s):  
Health Communication ◽  
Mobile Phones ◽  
Mobile Apps ◽  
Mobile App ◽  
Health Interventions ◽  
User Engagement ◽  
Health Campaigns ◽  
Efficient Tool ◽  
Limited Budget ◽  
Expectant Fathers

UNSTRUCTURED Health communication campaigns often suffer from the shortcomings of a limited budget and limited reach, resulting in a limited impact. This paper suggests a shift to audience-centered communication platforms – particularly apps on mobile phones. By using a common platform, multiple interventions and campaigns can combine resources and increase user engagement, resulting in a larger impact on health behavior. Given the widespread use of mobile phones, mobile apps can be an effective and efficient tool to provide health interventions. One such platform is Father’s Playbook, a mobile app designed for men to be more involved during their partner’s pregnancy. Health campaigns and interventions looking to reach expectant fathers can use Father’s Playbook as a vehicle for their messages.

scholarly journals Incorporating Contextual Information into Personalized Mobile Applications Recommendation

2021 ◽  
Author(s):  
Ke Zhu ◽  
Yingyuan Xiao ◽  
Wenguang Zheng ◽  
Xu Jiao ◽  
Chenchen Sun ◽  
...  
Keyword(s):  
Mobile Applications ◽  
Large Scale ◽  
Contextual Information ◽  
Mobile App ◽  
Mobile Internet ◽  
Current Time ◽  
Spatiotemporal Information ◽  
Data Overload ◽  
Effect Experiment ◽  
Contextual Characteristic

Abstract With the rise of the mobile internet, the number of mobile applications (apps) has shown explosive growth, which directly leads to the apps data overload. Currently, the recommender system has become the most effective method to solve the app data overload. App has the functional exclusiveness feature, which means the target users will not reuse apps with the same function in a certain spatiotemporal information. Most of the existing recommended methods for apps ignore the functional exclusiveness feature which makes it difficult to further improve the recommendation performance of the app recommendation. To solve this problem, we aim to improve the app recommendation performance, and propose a Personalized Context-aware Mobile App Recommendation Approach, called PCMARA. PCMARA comprehensively considers the user and app contextual information, which can mine the users app usage preference effectively. Specifically, (1) PCMARA explores the contextual characteristic of app, and constructs the app contextual factors for app which represent the function of app. (2) For the app functional exclusiveness problem, PCMARA leverages the app contextual factor to design a novel app similarity model, which enable to effectively eliminate this problem. (3) PCMARA considers the contextual information of users and apps to generates a recommendation list for target users based on the target users' current time and location. We applied the PCMARA to a real-world dataset and conducted a large-scale recommendation effect experiment. The experimental results show that the recommendation effect of PCMARA is satisfactory.

scholarly journals Mobile Application for Electric Power Monitoring on Energy Consumptions at a Campus University

2018 ◽  
Vol 11 (2) ◽  
pp. 637
Author(s):  
Murizah Kassim ◽  
Maisarah Abdul Rahman ◽  
Cik Ku Haroswati Che Ku Yahya ◽  
Azlina Idris
Keyword(s):  
Energy Consumption ◽  
Electric Power ◽  
Mobile Applications ◽  
Working Hours ◽  
Mobile App ◽  
Smart Meter ◽  
University Campus ◽  
Smart Meters ◽  
Power Monitoring ◽  
Energy Consumptions

This paper presents a research on electric power monitoring prototype mobile applications development on energy consumptions in a university campus. Electric power energy consumptions always are the issue of monitoring usage especially in a broad environment. University campus faces high used of electric power, thus crucial analysis on cause of the usage is needed. This research aims to analyses electric power usage in a university campus where implemented of few smart meters is installed to monitor five main buildings in a campus university. A Monitoring system is established in collecting electric power usage from the smart meters. Data from the smart meter then is analyzed based on energy consume on 5 buildings. Results presents graph on the power energy consume and presented on mobile applications using Live Code coding. The methodology involved the setup of the smart meters, monitoring and data collected from main smart meters, analyzed electrical consumptions for 5 buildings and mobile system development to monitor. A Live Code mobile app is designed then data collected from smart meter using ION software is published in graphs. Results presents the energy consumed for 5 building during day and night. Details on maximum and minimum energy consumption presented that show load of energy used in the campus. Result present Tower 1 saved most eenergy at night which is 65% compared to block 3 which is 8% saved energy although block 3 presents the lowest energy consumption in the working hours and non-working hours. This project is significant that can help campus facility to monitor electric power used thus able to control possible results in future implementations.

Sign in / Sign up

Export Citation Format

Share Document