Professor Yuan Miao received his PhD from Tsinghua University, Automation Department. He became a tenured associate professor at School of Computer Science and Mathematics, Victoria University 2004, full professor in 2011.

Professor Yuan Miao recent research interests are in the human knowledge modeling, intelligent software systems, and their application in big data analytics and smart systems. Two of his articles in fuzzy cognitive map modelling have been the top 10 most cited work from 2000 in the area (Google Scholar 2000-2016). He has applied the research in designation of innovative solutions in different domains such as health care and education.

Professor Yuan Miao is the Head of Information Technology, College of Engineering and Science, Victoria University.  For collaboration in both research and teaching programs, please contact Yuan at .

Fuzzy Cognitive Map for Causal Knowledge Modelling

Fuzzy Cognitive Map (FCM) is a visualized knowledge model representing human beings’ cognition of the external world. A fuzzy cognitive map represents factors as nodes/vertices and their causal relationships as links/arcs among nodes. The strength of the causal links is modeled with weights. Each factor can have either a binary/ternary set of states, or a multiple value set of states. It is a knowledge map of causally linked factors of a problem for facilitating human decision making.

As compared to many other knowledge models, it is much easier for domain experts to understand. However, domain experts with no computer science expertise still have difficulties in express their knowledge into FCM directly. Majority of the FCM applications in recent years are reported by computing experts in collaboration with domain experts. We provide a new FCM model to enable domain experts express their knowledge with FCM directly by removing the “specialized” tasks from the modeling process, including the definition of membership functions, the definition of decision functions and the normalization. This is an effort to reshape FCM as a tool for experts in a wide range of domains, aiming at a similar popularity or availability like that of concept maps.

For this new model, refer to Fuzzy Cognitive Map for Domain Experts with No Artificial Intelligence Expertise, ICARCV 2014. or its draft.

For dynamic causal relationships, refer to “Modelling Dynamic Causal Relationship in Fuzzy Cognitive Maps”,2014 IEEE International Conference on Fuzzy Systems; or more complex DCN, Dynamic Cognitive Network-an Extension of Fuzzy Cognitive Map, IEEE Transaction on Fuzzy Systems, Vo19, No5, pp.760-770, 2001 .

For an FCM analysis, refer to “On causal inference in Fuzzy Cognitive Map: Binary Concept States”, IEEE Transaction on Fuzzy Systems, Vol. 8, No.1, pp107-119, 2000.

Data Exchange and Service Integration with Applications in Heath Information Systems

Supported by ARC and Westgate GP Networks

Data exchange and service integration are highly desirable in most of the industry sectors including health care. After a few decades of development, most organisations now have information systems. The information systems act as the memory and nervous system of organisations. Most of the core business and services are carried out through the information systems. When a patient is treated by multiple organisations, data exchange and service integration of the different information systems are expected, e.g. when a patient is referred to a specialist or a hospital, or when a diabetes patient has a fracture, expecting two specialists to coordinate their services (treatment).

Data exchange and service integration faces a series of significant challenges. In many cases, it is not only technically difficult (heterogeneous platforms, programming languages, non access to source code, APIs), highly expensive (to modify the existing systems for integration), but also faces significant managerial hurdles (unwilling to integrate from different organisations for data or user protection). Although being highly desirable by patients, medical service providers and health managing departments, health data exchange and service integration are often not available due to these difficulties.

This project provides an intelligent data exchange and service integration model based on the knowledge of the individual systems and their users. It provides a smart data exchange of information allowing data flow out of one information system and into another information system; allowing synthesis of new services based on the services from individual information systems. The new models in this project enables an approach of data exchange and service integration without low level system integration. Therefore it is highly flexible and agile, can significantly lower the cost for service integration, at the same time address the managerial concerns on data privacy and confidence.

Privacy Preserving Data Sharing in Electronic Health Environment

Supported by ARC and Nexus Pte Ltd

Electronic Health Data (EHD) systems have significant advantages over traditional systems based on paper records. Over 90% of Australian medical service providers have adopted EHD systems. The value of EHDs can provide
vital information for medical resource allocation and medical research. However, privacy concerns are a major obstacle to obtaining benefits from analysing EHD systems. Previous research has addressed privacy concerns in different areas, but existing relevant technologies have three major weaknesses for the electronic health environment and so do not provide a privacy preserving solution for Australia:

•Existing technologies either are designed for one database or databases with an authentication centre. However,
Dr. Mukesh Haikerwal, lead clinician, National E-Health Transition Authority, suggested that national EHDs
will be in “a federation of repositories” rather than one centralised databank

•Current technologies cannot effectively protect records from insider attacks. In the past few years, Medicare
Australia investigated 1058 employees for possible privacy breaches; 54% had accessed records without
authorisation. Privacy concerns have blocked data value on EHDs and its significant benefits.

•The existing privacy preserving technologies were not designed for EHD federations and therefore cannot
obtain dynamic knowledge (which is essential in medical decision support), or assume the system has access to
unencrypted records, which are vulnerable to insider attacks.

This project will develop a new task-based framework for data sharing on EHD database federations while
protecting private data against both outsider and insider attacks, providing visualised, dynamic supports to medical
staff and government resource planners and policymakers.

Individualised Healthy Dietary Management

(call for sponsorship)

People have three meals a day, 365 days a year. Dietary intakes can have significant impact on human bodies. Some of the impacts are applicable to majority of people. For example, the over prevalence of diabetes in the modern society is mainly caused by unhealthy lifestyle and over intake of calories. Some of the impacts vary largely on individuals. A group of people, although being very slim or with perfect BMI, have fatty liver.

This project develops software agent which incorporate artificial intelligent technologies to perform individualised dietary education and management. Software agents as personal assistants are able to record and monitor people pre-clinical health condition, such as blood pressure trends, body temperature variations and etc. With body type classification and expert knowledge, software agents will perform data analysis of the record and help people understand their own body. Together with knowledge of nutrition characteristics of different food, recommend people on how they shall adjust their dietary intakes to best suit their body needs, prevent chronic disorder and improve general well being.

Generally, Australia public do not have these knowledge. Statistics has shown that many Australians have wrong impression of their health status. For example, about half of Australians who are overweighted do not realise that they are. Additionally, few Australian record, monitor or manage their health status, nor do they have tried to conduct data analysis on the records. In many cases, the disorder symptoms exist for a long time till it is developed into a disease. Personalised nutrition management is of significant benefit to individuals and the Medicare system.