Prof. Ghose will present a tutorial on "data-driven acquisition of conceptual models" at the premier conference on conceptual modeling (ER-2016, Nagoya, Japan).
A team led by Dr. Hoa Dam has won a $100,000 grant from Samsung to build deep learning models for predicting where safety hazards are likely to exist in the code base of software systems. This addresses a significant problem in this era of Internet of Things (IoT) where safety-critical systems (e.g. autonomous vehicles) are connected with consumer-oriented software applications (e.g. those running on mobile platforms or IoT devices). The co-investigators on this project are Prof Aditya Ghose (UOW), Dr Truyen Tran (Deakin) and Prof John Grundy (Deakin). Past winners of this award were from well-known groups at Stanford, Harvard, MIT, UC Berkeley, CMU, Oxford and Cambridge.
A team led by Dr. Hoa Dam (and including Dr. Truyen Tran and Prof. John Grundy from Deakin University as well as Prof. Aditya Ghose from the DSL) has developed a novel deep learning framework - called DeepSoft - for software analytics. The first vision paper outlining this project has been accepted at the ACM SIGSOFT International Symposium on Foundations of Software Engineering (FSE-2016).
DSL's latest research outcomes on mining social norms from open source communities have been presented at the 24th European Conference on Information Systems and 13th International Conference on Mining Software Repositories. The project is led by PhD student Daniel Avery, Dr. Hoa Dam and Prof. Aditya Ghose, and involves collaboration with University of Otago's researchers.
Prof. Aditya Ghose has been invited to deliver a 2-week course on "Computational Service Science" as part of the Prime Minister of India's key educational project, GIAN, under which under which foreign scholars teach at India's leading technology institutes to build knowledge and transform higher education across the country.
DSL authors Renuka Sindhgatta (also of IBM Research India), Prof. Aditya Ghose and Dr. Hoa Dam will report on a novel framework that uses data analytics for optimal business process provisioning at the 2016 International Conference on Advanced Information Systems Engineering (CAISE), the premier conference in this area (acceptance rate: 16.5%)
Prof. Aditya Ghose, Prof. Andrew Miller and Dr. David Stirling were among the authors of a submission (with Prof. David Thwaites of the University of Sydney as lead author) entitled "Mining routine radiation oncology clinical datasets within a distributed rapid learning framework: the potential for supporting optimised clinical decisions" that won the Best Abstract Award at the Cancer Institute of NSW Innovations in Cancer Treatment and Care Conference 2014.
Prof. John Mylopoulos, eminent software engineering and conceptual modelling researcher, is currently visiting the Decision Systems Lab for a period of 3 weeks. His visit was preceded by a visit by Prof. Akhil Kumar from Penn State University in early January.
The Decision Systems Lab/Centre for Oncology Informatics has initiated a new collaboration with the Liverpool Hospital on radiation oncology decision support, funded by a $114,000 research contract.
SPEAKER: Metta Santiputri UOW
DATE: Friday, October 28, 2016
TIME: 4pm onwards
TITLE: Mining task post-conditions: Automating the acquisition of process semantics
ABSTRACT: Semantic annotation of business process model in the business process designs has been addressed in a large and growing body of work, but these annotations can be difficult and expensive to acquire. This paper presents a data-driven approach to mining and validating these annotations (and specifically context-independent semantic annotations). We leverage event objects in process execution histories which describe both activity execution events (typically represented as process events) and state update events (represented as object state transition events). We present an empirical evaluation, which suggests that the approach provides generally reliable results.