Computing and Information Systems - Research Publications
Now showing items 1-12 of 1274
Profiling Mycobacterium tuberculosis transmission and the resulting disease burden in the five highest tuberculosis burden countries
BACKGROUND: Tuberculosis (TB) control efforts are hampered by an imperfect understanding of TB epidemiology. The true age distribution of disease is unknown because a large proportion of individuals with active TB remain undetected. Understanding of transmission is limited by the asymptomatic nature of latent infection and the pathogen's capacity for late reactivation. A better understanding of TB epidemiology is critically needed to ensure effective use of existing and future control tools. METHODS: We use an agent-based model to simulate TB epidemiology in the five highest TB burden countries-India, Indonesia, China, the Philippines and Pakistan-providing unique insights into patterns of transmission and disease. Our model replicates demographically realistic populations, explicitly capturing social contacts between individuals based on local estimates of age-specific contact in household, school and workplace settings. Time-varying programmatic parameters are incorporated to account for the local history of TB control. RESULTS: We estimate that the 15-19-year-old age group is involved in more than 20% of transmission events in India, Indonesia, the Philippines and Pakistan, despite representing only 5% of the local TB incidence. According to our model, childhood TB represents around one fifth of the incident TB cases in these four countries. In China, three quarters of incident TB were estimated to occur in the ≥ 45-year-old population. The calibrated per-contact transmission risk was found to be similar in each of the five countries despite their very different TB burdens. CONCLUSIONS: Adolescents and young adults are a major driver of TB in high-incidence settings. Relying only on the observed distribution of disease to understand the age profile of transmission is potentially misleading.
Socio-Technical Mitigation Effort to Combat Cyber Propaganda: A Systematic Literature Mapping
This systematic mapping literature aims to identify current research and directions for future studies in terms of combating cyber propaganda in the social media, which is used by both human effort and technological approaches (socio-technical) for mitigation. Out of 5176 retrieved articles, only 98 of them were selected for primary studies; classified based on research artifacts, mitigation effort, and the social media platforms involved in the research. The search was conducted using selected databases and applying selection criteria set for this research. Through the analysis, important research trends were identified based on human effort and technological approaches in mitigating and combating the cyber-propaganda issues. The authors also identified various mitigation socio-technical approaches such as identification, detection, image recognition, prediction, truth discovery and comprehension of rumours flow. The study also highlights areas for further improvements, to complement the performances of existing techniques. Besides, the study provides a brief review of cyber propaganda detection using classification techniques. Hence, it has set forth applicable research focus on the areas dealing with the mitigation of risk borne by cyber propaganda in the social media.
Designing for diversity in Aboriginal Australia: Insights from a national technology project
Aboriginal Australians have been colonized for over 230 years. As a result, many have been disconnected from their communities and identity. This paper reports on a national-scale HCI project that aims to design technology that allows Aboriginal Australians to reconnect with their communities and to reaffirm their Aboriginal identity. Our project faces significant challenges, some due to the effects of colonization and some due to the great (and underrecognized) diversity of Aboriginal Australia. In this paper, we report the design phase of our project, and discuss some of these challenges we faced. Through this, we offer insights for HCI designers and researchers undertaking similar work.
Automatic Repair of Same-Timestamp Errors in Business Process Event Logs
This paper contributes an approach for automatically correcting “same timestamp” errors in business process event logs. These errors consist in multiple events exhibiting the same timestamp within a given process instance. Such errors are common in practice and can be due to the logging granularity or the performance load of the logging system. Analyzing logs that have not been properly screened for such problems is likely to lead to wrong or misleading process insights. The proposed approach revolves around two techniques: one to reorder events with same-timestamp errors, the other to assign an estimated timestamp to each such event. The approach has been implemented in a software prototype and extensively evaluated in different settings, using both artificial and real-life logs. The experiments show that the approach significantly reduces the number of inaccurate timestamps, while the reordering of events scales well to large and complex datasets. The evaluation is complemented by a case study in the meat & livestock domain showing the usefulness of the approach in practice.
Does Smartphone Use Drive our Emotions or vice versa? A Causal Analysis
(Association for Computing Machinery, 2020-04-01)
In this paper, we demonstrate the existence of a bidirectional causal relationship between smartphone application use and user emotions. In a two-week long in-the-wild study with 30 participants we captured 502,851 instances of smartphone application use in tandem with corresponding emotional data from facial expressions. Our analysis shows that while in most cases application use drives user emotions, multiple application categories exist for which the causal effect is in the opposite direction. Our findings shed light on the relationship between smartphone use and emotional states. We furthermore discuss the opportunities for research and practice that arise from our findings and their potential to support emotional well-being.
Using internet enabled mobile devices and social networking technologies to promote exercise as an intervention for young first episode psychosis patients
BACKGROUND: Young people with first episode psychosis are at an increased risk for a range of poor health outcomes. In contrast to the growing body of evidence that suggests that exercise therapy may benefit the physical and mental health of people diagnosed with schizophrenia, there are no studies to date that have sought to extend the use of exercise therapy among patients with first episode psychosis. The aim of the study is to test the feasibility and acceptability of an exercise program that will be delivered via internet enabled mobile devices and social networking technologies among young people with first episode psychosis. METHODS/DESIGN: This study is a qualitative pilot study being conducted at Orygen Youth Health Research Centre in Melbourne, Australia. Participants are young people aged 15-24 who are receiving clinical care at a specialist first episode psychosis treatment centre. Participants will also comprise young people from the general population. The exercise intervention is a 9-week running program, designed to gradually build a person's level of fitness to be able to run 5 kilometres (3 miles) towards the end of the program. The program will be delivered via an internet enabled mobile device. Participants will be asked to post messages about their running experiences on the social networking website, and will also be asked to attend three face-to-face interviews. DISCUSSION: This paper describes the development of a qualitative study to pilot a running program coupled with the use of internet enabled mobile devices among young people with first episode psychosis. If the program is found to be feasible and acceptable to patients, it is hoped that further rigorous evaluations will ultimately lead to the introduction of exercise therapy as part of an evidence-based, multidisciplinary approach in routine clinical care.
A learning based approach to predict shortest-path distances
(Open Proceedings, 2020-01-01)
Shortest-path distances on road networks have many applications such as finding nearest places of interest (POI) for travel recommendations. To compute a shortest-path distance, traditional approaches traverse the road network to find the shortest path and return the path length. When the distances are needed first (e.g., to rank POIs) while the shortest paths may be computed later (e.g., after a POI is chosen), one may precompute and store the distances, and answer distance queries by simple lookups. This approach, however, falls short in the worst-cast space cost – O(n2) for n vertices even with various optimizations. To address these limitations, we propose to learn an embedding for every vertex that preserves its distances to the other vertices. We then train a multi-layer perceptron (MLP) to predict the distance between two vertices given their embeddings. We thus achieve fast distance predictions without a high space cost. Experimental results on real road networks confirm these advantages. Meanwhile, our approach is up to 97% more accurate than the state-of-the-art approaches for distance predictions.
Detection and removal of infrequent behavior from event streams of business processes
(Elsevier Ltd, 2020-05-01)
Process mining aims at gaining insights into business processes by analyzing the event data that is generated and recorded during process execution. The vast majority of existing process mining techniques works offline, i.e. using static, historical data, stored in event logs. Recently, the notion of online process mining has emerged, in which techniques are applied on live event streams, i.e. as the process executions unfold. Analyzing event streams allows us to gain instant insights into business processes. However, most online process mining techniques assume the input stream to be completely free of noise and other anomalous behavior. Hence, applying these techniques to real data leads to results of inferior quality. In this paper, we propose an event processor that enables us to filter out infrequent behavior from live event streams. Our experiments show that we are able to effectively filter out events from the input stream and, as such, improve online process mining results.
Person-Generated Health Data in Simulated Rehabilitation Using Kinect for Stroke: Literature Review.
(JMIR Publications Inc., 2018-05-08)
BACKGROUND: Person- or patient-generated health data (PGHD) are health, wellness, and clinical data that people generate, record, and analyze for themselves. There is potential for PGHD to improve the efficiency and effectiveness of simulated rehabilitation technologies for stroke. Simulated rehabilitation is a type of telerehabilitation that uses computer technologies and interfaces to allow the real-time simulation of rehabilitation activities or a rehabilitation environment. A leading technology for simulated rehabilitation is Microsoft's Kinect, a video-based technology that uses infrared to track a user's body movements. OBJECTIVE: This review attempts to understand to what extent Kinect-based stroke rehabilitation systems (K-SRS) have used PGHD and to what benefit. METHODS: The review is conducted in two parts. In part 1, aspects of relevance for PGHD were searched for in existing systematic reviews on K-SRS. The following databases were searched: IEEE Xplore, Association of Computing Machinery Digital Library, PubMed, Biomed Central, Cochrane Library, and Campbell Collaboration. In part 2, original research papers that presented or used K-SRS were reviewed in terms of (1) types of PGHD, (2) patient access to PGHD, (3) PGHD use, and (4) effects of PGHD use. The search was conducted in the same databases as part 1 except Cochrane and Campbell Collaboration. Reference lists on K-SRS of the reviews found in part 1 were also included in the search for part 2. There was no date restriction. The search was closed in June 2017. The quality of the papers was not assessed, as it was not deemed critical to understanding PGHD access and use in studies that used K-SRS. RESULTS: In part 1, 192 papers were identified, and after assessment only 3 papers were included. Part 1 showed that previous reviews focused on technical effectiveness of K-SRS with some attention on clinical effectiveness. None of those reviews reported on home-based implementation or PGHD use. In part 2, 163 papers were identified and after assessment, 41 papers were included. Part 2 showed that there is a gap in understanding how PGHD use may affect patients using K-SRS and a lack of patient participation in the design of such systems. CONCLUSIONS: This paper calls specifically for further studies of K-SRS-and for studies of technologies that allow patients to generate their own health data in general-to pay more attention to how patients' own use of their data may influence their care processes and outcomes. Future studies that trial the effectiveness of K-SRS outside the clinic should also explore how patients and carers use PGHD in home rehabilitation programs.
IoT-Enabled Flood Severity Prediction via Ensemble Machine Learning Models
(Institute of Electrical and Electronics Engineers (IEEE), 2020-04-06)
River flooding is a natural phenomenon that can have a devastating effect on human life and economic losses. There have been various approaches in studying river flooding; however, insufficient understanding and limited knowledge about flooding conditions hinder the development of prevention and control measures for this natural phenomenon. This paper entails a new approach for the prediction of water level in association with flood severity using the ensemble model. Our approach leverages the latest developments in the Internet of Things (IoT) and machine learning for the automated analysis of flood data that might be useful to prevent natural disasters. Research outcomes indicate that ensemble learning provides a more reliable tool to predict flood severity levels. The experimental results indicate that the ensemble learning using the Long-Short Term memory model and random forest outperformed individual models with a sensitivity, specificity and accuracy of 71.4%, 85.9%, 81.13%, respectively.
Using Multimodal Sensing to Improve Awareness in Human-AI Interaction
(Association for Computing Machinery (ACM), 2020)
In recent years, we have leveraged a broad range of AI techniques to improve our understanding and use of implicit human inputs for enhancing the capabilities of future AI-infused systems. At the same time, these new capabilities have given rise to novel interactions with AI, which require HCI techniques for improving its use through design and evaluation. In this paper, we promote the use of AI-supported unobtrusive multimodal sensing by presenting two ongoing projects that together explore intention, attention and activity recognition for developing and enabling three facets of awareness respectively—situation-, cognition- and context-awareness. Our collective efforts show a snapshot of how AI and HCI techniques can be combined to inform the design of interactive explainable AI systems and how we can better design their interactions.
Automated discovery of declarative process models with correlated data conditions
Automated process discovery techniques enable users to generate business process models from event logs extracted from enterprise information systems. Traditional techniques in this field generate procedural process models (e.g., in the BPMN notation). When dealing with highly variable processes, the resulting procedural models are often too complex to be practically usable. An alternative approach is to discover declarative process models, which represent the behavior of the process as a set of constraints. Declarative process discovery techniques have been shown to produce simpler models than procedural ones, particularly for processes with high variability. However, the bulk of approaches for automated discovery of declarative process models focus on the control-flow perspective, ignoring the data perspective. This paper addresses the problem of discovering declarative process models with data conditions. Specifically, the paper tackles the problem of discovering constraints that involve two activities of the process such that each of these two activities is associated with a condition that must hold when the activity occurs. The paper presents and compares two approaches to the problem of discovering such conditions. The first approach uses clustering techniques in conjunction with a rule mining technique, while the second approach relies on redescription mining techniques. The two approaches (and their variants) are empirically compared using a combination of synthetic and real-life event logs. The experimental results show that the former approach outperforms the latter when it comes to re-discovering constraints artificially injected in a log. Also, the former approach is in most of the cases more computationally efficient. On the other hand, redescription mining discovers rules with higher confidence (and lower support) suggesting that it may be used to discover constraints that hold for smaller subsets of cases of a process.