A/Prof Shen gave a keynote talk at an international HCI meetup
9 Aug 2023
9 Aug 2023
A/Prof Shen was invited by Bangkok ACM Sigchi Chapter to give a keynote talk at 13:00 AEST (10:00 ICT), on Saturday, 26 August 2023 at its HCI Meetup#5. His talk on Human-AI Interaction through Annotated Attention Heatmaps highlights the challenges of human understanding in deep neural networks and its impact on trust in AI systems. He will present a human-AI interactive sensemaking framework using annotated attention heatmaps, allowing humans and AI algorithms to mutually understand and fine-tune reasoning. More information can be found from his Facebook page https://fb.me/e/4mfh52gJA. You can find some snapshoots of the talk from here.
A/Prof Shen was invited by University of Adelaide for one week research visit
30 Jul 2023
A/Prof Shen was invited by the Centre for Research on Engineering Software Technologies (CREST) in the School of Computer and Mathematical Sciences at University of Adelaide for a research visit between 26 June and 1 July. During the visit, they initiated joint projects on the applications of large language models and visualisation techniques in software engineering, which involve two postdoc researchers and a number of research students. This visit was following a fruitful visit between 4 - 18 Jan 2023 where a research partnership was established for A/Prof Shen to visit CREST twice a year until the end of 2024.
Third article on DevSecOps published by IET Software
27 Feb 2023
The goal of this study is to report the state-of-the-practice of DevSecOps, including the impact of DevOps on security, practitioners' understanding of DevSecOps, and the practices associated with DevSecOps as well as the challenges of implementing DevSecOps. We carried out a grey literature review on DevSecOps, and surveyed the practitioners of DevSecOps in industry. Three major software security risks are identified with DevOps, where the establishment of DevOps pipeline provides opportunities for security-related activities. Read the paper here.
The paper with the title of ``Fed-SC: One-Shot Federated Subspace Clustering over High-Dimensional Data'' proposes a one-shot federated subspace clustering scheme Fed-SC that can achieve remarkable clustering effectiveness on high-dimensional data while keeping communication cost low using only one round of communication for each local device. The paper further establishes theoretical guarantees on the clustering effectiveness of one-shot Fed-SC and exploits the benefits of statistical heterogeneity across distributed data. Extensive experiments on synthetic and real-world datasets demonstrate significant effectiveness gains of Fed-SC compared with both subspace clustering and one-shot federated clustering methods. Read the paper here.
Article accepted by the flagship IEEE Transactions on Software Engineering
1 Feb 2023
Service monitoring through anomaly detection and root cause analysis (RCA) is crucial for online microservice systems to provide stable and continued services. Microservice systems rely on telemetry data to perform service monitoring and RCA of service anomalies. In this paper, we propose a new RCA solution TrinityRCL that is able to localize the root causes of anomalies at multiple levels of granularity including application-level, service-level, host-level, and metric-level, with the unique capability of code-level localization by harnessing all three types of telemetry data to construct a causal graph representing the intricate, dynamic, and nondeterministic relationships among the various entities related to the anomalies. Read the paper here.
The paper with the title of ``How Do Developers' Profiles and Experiences Influence their Logging Practices? An Empirical Study of Industrial Practitioners'' presents an empirical study using mixed methods comprising questionnaire surveys, semi-structured interviews, and code analyses with practitioners from a wide range of companies across a variety of industrial domains to establish a comprehensive and deep understanding of the relationships between developers' profiles and experiences and their logging practices from a wider perspective. Read the paper here.
2022 ACU Provost Innovation Fund Project on making online learning better through machine learning
10 Oct 2022
Hilstlab is partnering with ACU's Provost Office, Centre for Education and Innovation, and the Institute for Positive Psychology and Education in the Faculty of Health Sciences on a multi-phase project funded by Provost Innovation Fund to improve the quality of video-based learning at ACU hosted by Echo360 through a novel data analytics approach. The project uses machine learning of lecture videos and transcripts to analyse teachers' motivational behaviours in their teaching.
ARC Funded PhD Stipend Scholarship: Machine Learning for IoT
26 September 2022
We are offering one (1) full-time, 3-year PhD scholarship to undertake research in the area of machine learning for Internet of Things (IoT). Specifically, the PhD research will focus on the intelligent modelling, classification, and annotation of multimodal IoT sensor data and will be based in the HilstLab. The research is part of and funded by an Australian Research Council (ARC) Discovery Project. The application closes on 11.59pm 23 October 2022 AEST. Interested candidates are encouraged to contact Dr Kewen Liao before submitting an application. You can find more details on this scholarship from here. In addition, you are also welcome to apply for ACU RTP Stipend Scholarships, the October Round, which is due by 11.59pm 16 October 2022 AEST.
Article published by International Journal of Human-Computer Studies
14 September 2022
ADHD is a neurobiological condition that often affects school children. A major symptom is short attention span, which may negatively influence their academic performance, specifically in those tasks that require concentration. In this work, we design and evaluate a novel gaze-based visual attentive interface for the purpose of extending ADHD children's attention span in performing reading tasks. The results show that these children performed better in the presence of any color scheme in comparison with using no color, with highlighting the most effective, followed by contrast and then sharpening/blurring. Read the paper here.
Paper accepted by the 31st ACM International Conference on Information and Knowledge Management (CIKM)
5 August 2022
The paper "Do Simpler Statistical Methods Perform Better in Multivariate Long Sequence Time-Series Forecasting?" investigates two simple statistical methods SNaive and linear regression for MLSTF and provides analysis on them to indicate that linear regression owns a lower upper bound of error comparing to deep learning methods and SNaive can act as an effective nonparametric method for MLSTF with unpredictable trends. The paper can be found here.
HilstLab welcomes Dr MingJian Tang as a visiting fellow
4 August 2022
Dr. Tang is currently a Principal Data Scientist at Westpac with over 12 years post-PhD industry experience in Healthcare (Australian Government), Transport, Banking (Commonwealth Bank of Australia and Westpac Banking Corp.), Telecommunication (Optus, SingTel), and ICT. His research expertise includes cyber security, data science, machine learning, deep learning, and adversarial learning.
A/Prof Shen holds dual Senior Memberships of both the ACM and the IEEE
27 July 2022
A/Prof Shen has been elevated to Senior Member of the IEEE in April and then Senior Member of the ACM in July 2022. Senior member is the highest grade ACM and IEEE members can apply and one of the society's most prestigious member grades. Only 10% of IEEE's more than 400,000 members hold this grade worldwide, while less than 3% of the 100,000 current ACM members hold this grade.
Article accepted by the flagship IEEE Transactions on Software Engineering (TSE)
15 April 2022
Logs are valuable assets in software development. While various logging best practices are suggested to developers, have the critical issues been well addressed by current research endeavor? Check out our TSE paper on logging practices: why, where, what, and how from here. Through a systematic mapping study on logging practices with 56 primary studies, our work provides a holistic report of the existing research on logging practices by systematically synthesizing and analyzing the focus and inter-relationship of the existing research in terms of issues, research topics and solution approaches.
Article accepted by IEEE Journal of Biomedical and Health Informatics
22 February 2022
Laid on the foundation of the Human-AI Interactive and Continuous Sensemaking framework published at ACM CHI 2021, this paper explores the effectiveness of regularizing CNN network with human-provided attention guidance on where in the image the network should look for answering clues. On two orthopedics radiographic fracture classification datasets, through extensive experiments we demonstrate that explicit human-guided attention indeed can direct correct network attention and consequently significantly improve classification performance. You can find the paper
here.
HilstLab is now partnering with ACU's Sports Performance, Recovery, Injury and New Technologies (SPRINT) Research Centre.
28 January, 2022
SPRINT research focuses on the interactions between athlete performance, fatigue, sleep and recovery interventions and the prevention, risk identification and rehabilitation of injury with a strong emphasis on findings that directly translate into practice by developing new, and using established, research techniques. HilstLab contributes their technical expertise in data analytics, machine learning, computer vision, and deep learning.
Dr Kewen Liao is on a team that has recently been awarded a Discovery Project 2022 by ARC.
17 January, 2022
This project, led by Swinburne University of Technology, aims to devise a novel Internet of Things (IoT) sensor sharing marketplace that permits IoT applications to discover, integrate, and pay for any IoT sensor data that is made available by other parties. It will devise highly-scalable sensor classification, query processing, and transactions solutions and incorporate them in a pair of novel blockchains that work in tandem to securely manage all the information and contracts needed by IoT applications. This grant is the third one in a row from the HilstLab, following one in 2019-2020 and another in 2020-2021, both from Department of Defence.
The third paper in a row published by Information and Software Technology in 2021 and the second paper on DevSecOps.
15 September, 2021
This paper systemizes the knowledge about the challenges faced by practitioners when adopting DevSecOps and the proposed solutions reported in the literature. We also aim to identify the areas that need further research in the future. Through a systematic literature review, we identified 21 challenges related to adopting DevSecOps, 31 specific solutions, and the mapping between these findings. You can find the paper
here.
A/Prof Haifeng Shen becomes a visiting fellow at EPICentre.
3 September, 2021
A/Prof Haifeng Shen is now affiliated with UNSW's EPICentre (The Expanded Perception and Interaction Centre) as a visiting fellow. He was also invited to give a keynote talk at Visualisation Matters 2021.
Article published by Theoretical Computer Science
12 August, 2021
This paper investigates a novel theoretical problem of finding maximum disjoint paths with different colors (MDPDC). In MDPDC, transmission frequencies in a network are modeled as different colors on network nodes. The aim is to find a maximum number of color-constrained node-disjoint paths where nodes must share the same color within any disjoint path, and differ in color among different disjoint paths. The outcome of this research will greatly reduce network burden on big data transmission in order to achieve better network bandwidth and load balancing. You can find the paper
here.
A/Prof Haifeng Shen was invited to give a talk on "Human-Centred Technologies for the Common Good" at the second ACU Inter-Faculty Research Meetup on the Strathfield campus.
7 June, 2021
This meetup provided an opportunity for all ACU academics across all faculties to come together to share teaching and research, and engage in intellectual cross-disciplinary dialogue. You can find the presentation slides
here.
The second article published by Information and Software Technology in 2021
3 May, 2021
The article on "Processes, challenges and recommendations of Gray Literature Review: An experience report" aims to shed some light on Grey Literature Review (GLR) through a case study that uses Systematic Literature Review (SLR) and Multivocal Literature Review (MLR) guidelines to conduct a GLR on an emerging topic in SE - DevSecOps - to specifically answer the questions related to the reasons of using GL, the processes of conducting GL, and the impacts of GL on review results. Read the paper
here.
EAI ROSENET 2021
25 March, 2021
The 5th EAI International Conference on Robotics and Networks will take place in Shenzhen, China, during November 6-8. There have been more and more notable developments of robotics in many applications, such as minimally invasive surgery, manufacturing, unmanned vehicles. On the other hand, computer networks have been covering every corner of our daily life. However, existing robotic and networking techniques are still facing grand challenges, from intelligence to security. Hence, we are looking for high-quality, original research articles that advance the fields of robotics and network towards more intelligent, capable, autonomous systems. Authors of selected papers will be invited to submit an extended version to Mobile Networks and Applications (MONET) Journal.
MODSIM 2021, 5 - 10 December 2021
18 March, 2021
The 24th International Congress on Modelling and Simulation (MODSIM2021) will be held at The University of Sydney and International Convention Centre Sydney. HilstLab and UNSW's EPICentre are jointly organising Session C4: Immersive simulations for AI and hybrid analytics . This session will focus on Immersive Simulations for Explainable Artificial Intelligence (xAI) and Hybrid Analytics (HA) which is a science of analytical reasoning facilitated by interactive visual interfaces, deployed through effective use of multi-modal Augmented Reality / Virtual Reality / Mixed Reality systems. You are welcome to submit your papers to this session.
Paper accepted by the Late Breaking Work track of 2021 ACM SIGCHI Conference on Human Factors in Computing Systems
19 February, 2021
The article on "Human-AI Interactive and Continuous Sensemaking: A Case Study of Image Classification using Scribble Attention Maps" presents a conceptual framework for supporting effective human-AI interaction. The framework, referred to as HAICS, has been validated by an image classification application using deep CNNs as a browser-based tool that displays network attention maps to the human for explainability and collects human's feedback in the form of scribble annotations overlaid onto the maps. Experimental results using a real-world fracture dataset has shown significant improvement of AI’s classification accuracy after fine-tuning itself to incorporate human’s feedback. Read the paper
here.
Article published by Information and Software Technology
11 December, 2020
The article on "quality assessment in systematic literature reviews: a software engineering perspective" presents a tertiary study that systematically reviews quality assessment (QA) used by systematic literature reviews (SLRs) in software engineering during 2004 and 2018 with the aim to depict a clear picture by aggregating and distilling good practices, including the commonly used QA instruments as well as the major roles and aspects of QA in research.
Read the paper
here.
Dr Maoying Qiao won ACU's Research Award for Women Academic Staff 2021
24 November, 2020
Dr Qiao's project "diverse hard negative mining for imbalanced learning and its application for object detection and pneumonia detection" aims to develop a diversity integrated solution to the imbalanced learning problem. Expected outcomes are with strong potential to benefit a wide variety of real-world applications in AI-enabled healthcare, smart agriculture and autonomous vehicles.
Article published by International Journal of Human-Computer Studies
1 November,2020
The article on "Understanding the effects of real-time sentiment analysis and morale visualisation in backchannel systems: A case study" presents the design and evaluation of a backchannel system for connecting presenters with audience using a morale graph based interface based on real-time sentiment analysis and morale visualisation.
Read the paper
here.
Our research lab, together with collaborators from Western Sydney University, Defence Science and Technology Group, and Total Technology Partners Pty Ltd won a $95,000 grant from Next Generation Technologies Fund to develop a collaborative human-centred AI system for submarine control room console operators and study its impact on the decision making performance in a high risk, time critical environment.
Paper accepted by EASE 2020
3 February 2020
Our submission to the International Conference on Evaluation and Assessment in Software Engineering (EASE) 2020 was accepted as a full paper. The full paper track acceptance rate was 23%. EASE is a top conference in Software Engineering. The paper systematically studies the characteristics of imbalanced and time-series data and further provides a comprehensive evaluation for predictive CI/CD with the data from real projects.
Research project funded by Defence Science and Technology Group
4 March 2019
Partnering with UNSW Sydney and UNSW Canberra, this project is part of a larger project on simulation and visualisation using data farming for joint operating concept analysis. It aims to develop a framework and a concept demonstrator showing the feasibility of running high-end graphics-intensive visualisations across multiple platforms through visualisation as a service.