Shri Rai

The main contribution of this paper is to introduce a framework for integrating Machine Learning (ML), Human, and Virtual Reality (VR) into one platform to promote a collaborative visualisation environment that can assist in better analysis and improve the human-machine teaming capability. This platform was demonstrated using a case study in ana-lysing road traffic conditions. The ‘Ab-normal Machine Learning Road Traffic Detection in VR (AbnMLRTD-VR)’ prototype system was developed to assist the human analyst. The proposed system has two main integrative components: a data-driven ML model and a 3D real-time visualisation in a VR environment. An unsupervised ML model was built using real traffic data. The AbnMLRTD-VR system highlights the outliers in the road sections in actual road contexts of a road traffic network. This gives the human analyst a 3D real-time immersive visualisation in a VR environment to evaluate road conditions. The AbnMLRTD-VR system demonstrated that it could help minimise the need for human pre-labelling of the data. It enables the visualisation of the road traffic conditions more meaningfully and to understand the context of the road traffic conditions of road sections at any given time.

To improve users’ tendency towards shopping in Virtual Reality (VR), en-hancing the User Experience (UX) of the VR shopping environments is of primary importance. Product viewability, reachability, and personalisation are some of the primary UX factors in a shopping environment. This paper proposes and discusses three factors for a Personalised Adaptive Aisle (PAA) in a VR shopping environment to improve the shopping experience. They are 1) Shelf placement for viewability and reachability, 2) User view-point in VR, and 3) Personalised Product placement.

Engagement with upper limb rehabilitation post-stroke can improve rehabilitation outcomes. Virtual Reality can be used to make rehabilitation more engaging. In this paper, we propose a multiple case study to determine: (1) whether game design principles (identified in an earlier study as being likely to engage) actually do engage, in practice, a sample of stroke survivors with a Desktop Virtual Reality-based Serious Game designed for upper limb rehabilitation; and (2) what game design factors support the existence of these principles in the game. In this study, we considered 15 principles: awareness , feedback , interactivity , flow , challenge , attention , interest , involvement , psychological absorption , motivation , effort , clear instructions , usability , purpose , and a first-person view . Four stroke survivors used, for a period of 12 weeks, a Virtual Reality-based upper limb rehabilitation system called the Neuromender Rehabilitation System. The stroke survivors were then asked how well each of the 15 principles was supported by the Neuromender Rehabilitation System and how much they felt each principle supported their engagement with the system. All the 15 tested principles had good or reasonable support from the participants as being engaging. Use of feedback was emphasised as an important design factor for supporting the design principles, but there was otherwise little agreement in important design factors among the participants. This indicates that more personalised experiences may be necessary for optimised engagement. The insight gained can be used to inform the design of a larger scale statistical study into what engages stroke survivors with Desktop Virtual Reality-based upper limb rehabilitation.

The use of sensors for monitoring livestock has opened up new possibilities for the management of livestock in extensive grazing systems. The work presented in this paper aimed to develop a model for predicting the metabolisable energy intake (MEI) of sheep by using temperature, pitch angle, roll angle, distance, speed, and grazing time data obtained directly from wearable sensors on the sheep. A Deep Belief Network (DBN) algorithm was used to predict MEI, which to our knowledge, has not been attempted previously. The results demonstrated that the DBN method could predict the MEI for sheep using sensor data alone. The mean square error (MSE) values of 4.46 and 20.65 have been achieved using the DBN model for training and testing datasets, respectively. We also evaluated the influential sensor data variables, i.e., distance and pitch angle, for predicting the MEI. Our study demonstrates that the application of machine learning techniques directly to on-animal sensor data presents a substantial opportunity to interpret biological interactions in grazing systems directly from sensor data. We expect that further development and refinement of this technology will catalyse a step-change in extensive livestock management, as wearable sensors become widely used by livestock producers.

One of the fundamental tasks in the management of livestock is to understand their behaviour and use this information to increase livestock productivity and welfare. Developing new and improved methods to classify livestock behaviour based on their daily activities can greatly improve livestock management. In this paper, we propose the use of a hierarchical machine learning method to classify livestock behaviours. We first classify the livestock behaviours into two main behavioural categories. Each of the two categories is then broken down at the next level into more specific behavioural categories. We have tested the proposed methodology using two commonly used classifiers, Random Forest, Support Vector Machine and a newer approach involving Deep Belief Networks. Our results show that the proposed hierarchical classification technique works better than the conventional approach. The experimental studies also show that Deep Belief Networks perform better than the Random Forest and Support Vector Machine for most cases.

It is challenging to ensure security and to minimize economic impacts due to cyber-attacks because of the heavy reliance on ICT in different organizations and this paper presents an approach to estimate the cost of cyber security in public and private sector organizations. The paper also describes an approach for selecting the type of cyber security improvements to ensure that organizational goals are achieved. Different types of cyber-attacks and the subsequent impacts of these attacks are considered. A Value Analysis method is proposed to support the decision-making process by determining the priorities of deployment of various cyber security technologies. The proposed method is based on security costs related to and the losses due to attacks. Examples are provided in the paper to illustrate the proposed approach.

In traditional research, data fusion is referred to as multi-sensor data fusion. The theory is that data from multiple sources can be combined to provide more accurate, reliable and meaningful information than that provided by a single data source. Applications in this field of study were originally in the military domain; more recently, investigations for its application in various civilian domains (eg: computer security) have been undertaken. Multi-sensor data fusion as applied to biometric authentication is termed multi-modal biometrics. The objective of this study was to apply feature level fusion of fingerprint feature and keystroke dynamics data for authentication purposes, utilizing Artificial Neural Networks (ANNs) as a classifier. Data fusion was performed adopting the cooperative paradigm, a less researched approach. This approach necessitates feature subset selection to utilize the most discriminatory data from each source. Experimental results returned a false acceptance rate (FAR) of 0.0 and a worst case false rejection rate (FRR) of 0.0006, which were comparable to—and in some cases, slightly better than—other research using the cooperative paradigm.

The advanced computational capabilities in modern personal computers have made it possible for consumers to experience simulations with a high degree of verisimilitude through simulation games (a.k.a. Sims). In recent years, the cross-boundary technology exchange between game and simulation technology, along with other factors, has contributed to the confusion as to what makes a simulation game and what makes a simulator. This article provides a user's and designer's perspective on a definitive comparison of the similarities and differences between games in general, simulation games, and simulators. It also introduces a method that can be easily used to distinguish games and simulation games from simulators by using observable design characteristics. On the other hand, the convergence of functionality and technology in simulation games and simulators has created new applications of simulation. One such application is in serious games. Serious games and simulation games are confusingly similar in many ways. However, they greatly differ in functionality. This article also provides a method to distinguish serious games from simulation games, to clarify the strict categorization between these two applications of simulation.

Recent developments in learning technology such as hypermedia is becoming widespread and offer significant contribution to improve the delivery of learning and teaching materials. A key factor in the development of hypermedia learning system is cognitive style (CS) as it relates to users' information processing habits, representing individual user's typical modes of perceiving, thinking, remembering and problem solving. The sample comprised of 217 students from Murdoch University who were enrolled in a first-year undergraduate unit. A survey was carried out every second semester over a period of 3 years (1999-2001). Both generalized linear model and tree-based regression were used to analyse the interaction among the learning dimensions and the effect on students' CS. When comparing both models, tree-based regression outperformed generalized linear model in this study. The research findings indicated that non-linear learning is the primary dimension that determines students' CS. This is subsequently, followed by multiple tools (MT) and learner control (LC) dimensions. The results also confirm that background information has effects on students' CS. The overall findings suggest that students' preference of learning dimensions such as linear vs. non-linear, level of LC and the range of MT must be taken into consideration in order to enrich students' quality of education by means of motivating students' acquisition of subject matter through individualize instruction when designing, developing, and delivering educational resources.

Lifelong learning is quickly becoming an integrated part of todays working life because of the demand for keeping up to date with latest developments due to rapid change in technology and business. To maintain flexibility and quality, online technology is often used as a medium of educational service delivery. In this chapter, we examine a case where online technology has been used to coordinate virtual project teams (in an educational setting) around the world. The experience from the case study is that although online technology promises to offer an independent learning environment anytime and anywhere, only some aspects of the technology are useful depending on the nature of the task. In addition, there are behavioural and cultural issues, which can be exacerbated by underdeveloped personal relationships due to constraints imposed by online technology. The case study also shows that synchronous communication is not necessarily a better means to coordinate than its asynchronous counterpart. A challenge faced by online education providers is to find the right mix and how to come up with a framework that will provide optimal results.