Publications

Automated Traceability for Domain Modelling Decisions Empowered by Artificial Intelligence
Rijul Saini
Gunter Mussbacher
Jin L.C. Guo
Jörg Kienzle
Domain modelling abstracts real-world entities and their relationships in the form of class diagrams for a given domain problem space. Model… (voir plus)lers often perform domain modelling to reduce the gap between understanding the problem description which expresses requirements in natural language and the concise interpretation of these requirements. However, the manual practice of domain modelling is both time-consuming and error-prone. These issues are further aggravated when problem descriptions are long, which makes it hard to trace modelling decisions from domain models to problem descriptions or vice-versa leading to completeness and conciseness issues. Automated support for tracing domain modelling decisions in both directions is thus advantageous. In this paper, we propose an automated approach that uses artificial intelligence techniques to extract domain models along with their trace links. We present a traceability information model to enable traceability of modelling decisions in both directions and provide its proof-of-concept in the form of a tool. The evaluation on a set of unseen problem descriptions shows that our approach is promising with an overall median F2 score of 82.04%. We conduct an exploratory user study to assess the benefits and limitations of our approach and present the lessons learned from this study.
2021-08-31
IEEE International Requirements Engineering Conference (publié)
doi.org
DoMoBOT: A Modelling Bot for Automated and Traceable Domain Modelling
Rijul Saini
Gunter Mussbacher
Jin L.C. Guo
Jörg Kienzle
In the initial phases of the software development cycle, domain modelling is typically performed to transform informal requirements expresse… (voir plus)d in natural language into concise and analyzable domain models. These models capture the key concepts of an application domain and their relationships in the form of class diagrams. Building domain models manually is often a time-consuming and labor-intensive task. The current approaches which aim to extract domain models automatically, are inadequate in providing insights into the modelling decisions taken by extractor systems. This inhibits modellers to quickly confirm the completeness and conciseness of extracted domain models. To address these challenges, we present DoMoBOT, a domain modelling bot that uses a traceability knowledge graph to enable traceability of modelling decisions from extracted domain model elements to requirements and vice-versa. In this tool demo paper, we showcase how the implementation and architecture of DoMoBOT facilitate modellers to extract domain models and gain insights into the modelling decisions taken by our bot.
2021-08-31
IEEE International Requirements Engineering Conference (publié)
doi.org
Routing policy choice prediction in a stochastic network: Recursive model and solution algorithm
Tien Mai
Xinlian Yu
Song Gao
Emma Frejinger
2021-08-31
Transportation Research Part B: Methodological (publié)
doi.org
A Survey of Exploration Methods in Reinforcement Learning
Susan Amin
Maziar Gomrokchi
Harsh Satija
Herke van Hoof
Doina Precup
2021-08-31
ArXiv (prépublication)
arxiv.org
arxiv.org
Abg-CoQA: Clarifying Ambiguity in Conversational Question Answering
Meiqi Guo
Mingda Zhang
Siva Reddy
Malihe Alikhani
Effective communication is about the dissemination of properly worded meaningful ideas/messages that are comprehensible to both sen… (voir plus)der and receiver and which ultimately can attract the desired response or feedback. For machines to engage in a conversation, it is therefore essential to enable them to clarify ambiguity and achieve a common ground. We introduce Abg-CoQA, a novel dataset for clarifying ambiguity in Conversational Question Answering systems. Our dataset contains 9k questions with answers where 1k questions are ambiguous, obtained from 4k text passages from five diverse domains. For ambiguous questions, a clarification conversational turn is collected. We evaluate strong language generation models and conversational question answering models on Abg-CoQA. The best-performing system achieves a BLEU-1 score of 12.9% on generating clarification question, which is 27.9 points behind human performance (40.8%); and a F1 score of 40.1% on question answering after clarification, which is 35.1 points behind human performance (75.2%), indicating there is ample room for improvement.
2021-08-30
AKBC.ws/2021/Conference (accepté)
doi.org
openreview.net
Problems in the deployment of machine-learned models in health care
Joseph Paul Cohen
Tianshi Cao
Joseph Viviano
Chinwei Huang
Michael Fralick
Marzyeh Ghassemi
M. Mamdani
R. Greiner
Yoshua Bengio
2021-08-29
Canadian Medical Association Journal (publié)
doi.org
A Data Mining Analysis of Cross-Regional Study of Apparel Consumption
Osmud Rahman
Benjamin C. M. Fung
2021-08-27
Comprehensible Science (publié)
doi.org
Approximate Bayesian Optimisation for Neural Networks
Nadhir Hassen
Irina Rish
2021-08-26
ArXiv (prépublication)
arxiv.org
arxiv.org
Extracting Weighted Automata for Approximate Minimization in Language Modelling
Clara Lacroce
Prakash Panangaden
Guillaume Rabusseau
In this paper we study the approximate minimization problem for language modelling. We assume we are given some language model as a black bo… (voir plus)x. The objective is to obtain a weighted finite automaton (WFA) that fits within a given size constraint and which mimics the behaviour of the original model while minimizing some notion of distance between the black box and the extracted WFA. We provide an algorithm for the approximate minimization of black boxes trained for language modelling of sequential data over a one-letter alphabet. By reformulating the problem in terms of Hankel matrices, we leverage classical results on the approximation of Hankel operators, namely the celebrated Adamyan-Arov-Krein (AAK) theory. This allows us to use the spectral norm to measure the distance between the black box and the WFA. We provide theoretical guarantees to study the potentially infinite-rank Hankel matrix of the black box, without accessing the training data, and we prove that our method returns an asymptotically-optimal approximation.
2021-08-24
Proceedings of the Fifteenth International Conference on Grammatical Inference (publié)
doi.org
proceedings.mlr.press
HAD-Net: A Hierarchical Adversarial Knowledge Distillation Network for Improved Enhanced Tumour Segmentation Without Post-Contrast Images
Saverio Vadacchino
Raghav Mehta
Nazanin Mohammadi Sepahvand
Brennan Nichyporuk
James J Clark
Tal Arbel
Segmentation of enhancing tumours or lesions from MRI is important for detecting new disease activity in many clinical contexts. However, ac… (voir plus)curate segmentation requires the inclusion of medical images (e.g., T1 post contrast MRI) acquired after injecting patients with a contrast agent (e.g., Gadolinium), a process no longer thought to be safe. Although a number of modality-agnostic segmentation networks have been developed over the past few years, they have been met with limited success in the context of enhancing pathology segmentation. In this work, we present HAD-Net, a novel offline adversarial knowledge distillation (KD) technique, whereby a pre-trained teacher segmentation network, with access to all MRI sequences, teaches a student network, via hierarchical adversarial training, to better overcome the large domain shift presented when crucial images are absent during inference. In particular, we apply HAD-Net to the challenging task of enhancing tumour segmentation when access to post-contrast imaging is not available. The proposed network is trained and tested on the BraTS 2019 brain tumour segmentation challenge dataset, where it achieves performance improvements in the ranges of 16% - 26% over (a) recent modality-agnostic segmentation methods (U-HeMIS, U-HVED), (b) KD-Net adapted to this problem, (c) the pre-trained student network and (d) a non-hierarchical version of the network (AD-Net), in terms of Dice scores for enhancing tumour (ET). The network also shows improvements in tumour core (TC) Dice scores. Finally, the network outperforms both the baseline student network and AD-Net in terms of uncertainty quantification for enhancing tumour segmentation based on the BraTs 2019 uncertainty challenge metrics. Our code is publicly available at: https://github.com/SaverioVad/HAD_Net
2021-08-24
Proceedings of the Fourth Conference on Medical Imaging with Deep Learning (publié)
doi.org
proceedings.mlr.press
Monitoring non-pharmaceutical public health interventions during the COVID-19 pandemic
Yannan Shen
Guido Powell
Iris Ganser
Qulu Zheng
Chris Grundy
Anya Okhmatovskaia
David L. Buckeridge
Measuring and monitoring non-pharmaceutical interventions is important yet challenging due to the need to clearly define and encode non-phar… (voir plus)maceutical interventions, to collect geographically and socially representative data, and to accurately document the timing at which interventions are initiated and changed. These challenges highlight the importance of integrating and triangulating across multiple databases and the need to expand and fund the mandate for public health organizations to track interventions systematically.
2021-08-23
Scientific Data (publié)
doi.org
Generating community measures of food purchasing activities using store-level electronic grocery transaction records: an ecological study in Montreal, Canada
Hiroshi Mamiya
Alexandra M. Schmidt
Erica E.M. Moodie
Yu Ma
David L Buckeridge
2021-08-22
Public Health Nutrition (publié)
doi.org