We propose a novel MLaaS Dataset Generator (MDG) framework that creates configurable and reproducible datasets for evaluating Machine Learning as a Service (MLaaS) selection and composition. MDG simulates realistic MLaaS behaviour by training and evaluating diverse model families across multiple real-world datasets and data distribution settings. It records detailed functional attributes, quality of service metrics, and composition-specific indicators, enabling systematic analysis of service performance and cross-service behaviour. Using MDG, we generate more than ten thousand MLaaS service instances and construct a large-scale benchmark dataset suitable for downstream evaluation. We also implement a built-in composition mechanism that models how services interact under varied Internet of Things conditions. Experiments demonstrate that datasets generated by MDG enhance selection accuracy and composition quality compared to existing baselines. MDG provides a practical and extensible foundation for advancing data-driven research on MLaaS selection and composition.

The dynamic nature of Internet of Things (IoT) environments challenges the long-term effectiveness of Machine Learning as a Service (MLaaS) compositions. The uncertainty and variability of IoT environments lead to fluctuations in data distribution, e.g., concept drift and data heterogeneity, and evolving system requirements, e.g., scalability demands and resource limitations. This paper proposes an adaptive MLaaS composition framework to ensure a seamless, efficient, and scalable MLaaS composition. The framework integrates a service assessment model to identify underperforming MLaaS services and a candidate selection model to filter optimal replacements. An adaptive composition mechanism is developed that incrementally updates MLaaS compositions using a contextual multi-armed bandit optimization strategy. By continuously adapting to evolving IoT constraints, the approach maintains Quality of Service (QoS) while reducing the computational cost associated with recomposition from scratch. Experimental results on a real-world dataset demonstrate the efficiency of our proposed approach.

Artificial Intelligence Generated Content (AIGC) has rapidly emerged with the capability to generate different forms of content, including text, images, videos, and other modalities, which can achieve a quality similar to content created by humans. As a result, AIGC is now widely applied across various domains such as digital marketing, education, and public health, and has shown promising results by enhancing content creation efficiency and improving information delivery. However, there are few studies that explore the latest progress and emerging challenges of AIGC across different domains. To bridge this gap, this paper brings together 16 scholars from multiple disciplines to provide a cross-domain perspective on the trends and challenges of AIGC. Specifically, the contributions of this paper are threefold: (1) It first provides a broader overview of AIGC, spanning the training techniques of Generative AI, detection methods, and both the spread and use of AI-generated content across digital platforms. (2) It then introduces the societal impacts of AIGC across diverse domains, along with a review of existing methods employed in these contexts. (3) Finally, it discusses the key technical challenges and presents research propositions to guide future work. Through these contributions, this vision paper seeks to offer readers a cross-domain perspective on AIGC, providing insights into its current research trends, ongoing challenges, and future directions.

Empathy indicates an individual's ability to understand others. Over the past few years, empathy has drawn attention from various disciplines, including but not limited to Affective Computing, Cognitive Science, and Psychology. Detecting empathy has potential applications in society, healthcare and education. Despite being a broad and overlapping topic, the avenue of empathy detection leveraging Machine Learning remains underexplored from a systematic literature review perspective. We collected 849 papers from 10 well-known academic databases, systematically screened them and analysed the final 82 papers. Our analyses reveal several prominent task formulations – including empathy on localised utterances or overall expressions, unidirectional or parallel empathy, and emotional contagion – in monadic, dyadic and group interactions. Empathy detection methods are summarised based on four input modalities – text, audiovisual, audio and physiological signals – thereby presenting modality-specific network architecture design protocols. We discuss challenges, research gaps and potential applications in the Affective Computing-basedempathydomain, which can facilitate new avenues of exploration. We further enlist the public availability of datasets and codes. This paper, therefore, provides a structured overview of recent advancements and remaining challenges towards developing a robust empathy detection system that could meaningfully contribute to enhancing human well-being.

Split-fed learning (SFL) is a novel approach within distributed collaborative machine learning that combines federated learning (FL) and split learning (SL). While SFL benefits from FL's speed and SL's efficiency, it also inherits their disadvantages, including trust issues such as model poisoning and training-hijacking attacks, and additional reliability concerns such as a single point of failure and lack of motivation. Existing solutions have integrated blockchain technology to address all reliability issues and poisoning attacks, but are limited to FL. Moreover, there are limited solutions to address training-hijacking, such as the SplitGuard protocol. In this article, we propose two solutions, validated blockchained split-fed learning (VBSFL) and VBSL, focusing on VBSFL, which leverages blockchain technology by building on VBFL and incorporating the SplitGuard protocol to address these challenges. Experimental results with real-world datasets demonstrate the effectiveness, efficiency, and scalability of the proposed approach.

In recent years, the combination of unmanned aerial vehicles (UAVs) and wireless sensor networks (WSNs) has gained popularity in livestock management (LM) due to energy constraints and network instability. Limited energy storage of sensor nodes (SNs) and the possibility of packet loss contribute to fast energy consumption and unstable networks, respectively. UAVs serve as relay nodes and data sinks, addressing these issues by temporarily storing data to reduce SN workload and establishing mobile nodes for network stability. We propose two innovations based on previous work: 1) We introduce a multi-layer wireless network architecture, categorizing UAVs into two layers based on their functions including data collection and data processing. This enhances task parallelization, bridging performance gaps among multiple UAVs; 2) We overcome the mobility limitation of SNs, considering their real-time movement in the network. Through deep reinforcement learning, UAVs learn to cooperatively locate moving SNs. This accounts for the inevitable mobility of livestock in the industry. Additionally, we simulate the environment and compare our approach to traditional methods, evaluating metrics such as collected data per timestep (DCPS), energy consumed per timestep (ECPS), and network stability (NS). Experimental results demonstrate that our method outperforms traditional approaches, achieving a data collecting gain of 4.84% and 8.20% compared to the methods without considering SN mobility or the multi-layer characteristics of WSNs, respectively. Under energy consumption limits, our method yields energy savings of 3.00% and 1.35% respectively. Furthermore, we extensively study and validate our method against other path planning algorithms, including genetic particle swarm optimization (GPSO), modified central force optimization (MCFO), and rapidly-exploring random trees (RRT). Our approach surpasses these methods in terms of data collecting efficiency and network stability.

The popularity of Ethereum as a platform for Stablecoin transactions (for example, AUDN) continues to rise. It is therefore paramount that the integrity and security of transactions within these decentralized systems are guaranteed. The intricate network of interactions occurring during the exchange of cryptocurrencies made the task of identifying specific transactions as fraudulent difficult because fraudulent behaviour can be concealed within legitimate smart contract operations. Leveraging the inherent structure and interconnectedness of Ethereum transactions, this paper proposes a comprehensive framework to address issues such as Frontrunning within the cryptocurrency ecosystem. Constructing a knowledge graph representation of fraudulent Ethereum blockchain transactions, the proposed solution captures the relationships between addresses, transactions, and smart contracts and generates BotVictim recommendations based on Victim Receiver similarity scores exceeding 85%. These results are generated by excluding temporal transactions, a unique approach when examining the Ethereum network. Thus, our approach enables early detection and prevention of fraudulent activities, potentially safeguarding the interests of cryptocurrency users and mitigating potential financial losses. To evaluate the effectiveness of the proposed framework, its performance is compared against traditional fraud detection methods. The proposed solution demonstrates superiority in terms of accuracy and efficiency.

Visual scene understanding mainly depends on pixel-wise classification obtained from a deep convolutional neural network. However, existing semantic segmentation models often face difficulties in real-time applications due to their large network architecture. Although there are real-time semantic segmentation models available, their shallow backbone can degrade the performance considerably. This paper introduces SDBNetV2, a lightweight semantic segmentation model designed to improve real-time performance without increasing computational costs. A key contribution is a novel Short-term Dense Bottleneck (SDB) module in the encoder, which provides varied field-of-views to capture different geometrical objects in a complex scene. Additionally, we propose dense feature refinement and improved semantic aggregation modules at the decoder end to enhance contextualization and object localization. We evaluate the proposed model’s performance on several indoor and outdoor datasets in structured and unstructured environments. The results show that SDBNetV2 achieves superior segmentation performance over other real-time models with less than 2 million parameters.

Recent studies show a significant growth in semantic segmentation . However, many semantic segmentation models still have a large number of parameters, making them unsuitable for resource-constrained embedded devices. To address this issue, we propose an efficient Shared Feature Reuse Segmentation (SFRSeg) model containing several novelties: a new yet effective shared-branch multiple sub-encoders design, a context mining module and a semantic aggregating module for better context granularity . In particular, our shared-branch approach improves the entire feature hierarchy by sharing the spatial and context knowledge in both shallow and deep branches. After every shared point in each sub-encoder, a proposed cascading context mining (CCM) module is deployed to filter out the noisy spatial details from the feature maps and provides a diverse size of receptive fields for capturing the latent context between multi-scale geometric shapes in the scene. To overcome the gradient vanishing issue at the early stage, we reduce the number of layers in the first sub-encoder and employ a unique multiple sub-encoders design which reprocesses the rich global feature maps through multiple sub-encoders for better feature refinement. Later, the rich semantic features generated by the efficient sub-encoders at different levels are fused by the proposed Hybrid Path Attention Semantic Aggregation (HPA-SA) module that effectively reduces the semantic gap between feature maps at different levels and alleviate the well-known boundary degeneration effect. To make it computationally efficient for resource-constrained embedded devices, a series of lightweight methods such as a lightweight encoder, a squeeze-and-excitation design, separable convolution filters , channel reduction (CR) are carefully exploited. With an exceptional performance on Cityscapes (70.6% test mIoU) and CamVid (74.7% test mIoU) data sets, the proposed model is shown to be superior over existing light real-time semantic segmentation models whilst having only 1.6 million parameters.

The Crowdsourced IoT Service (CIS) market is inherently different from other service markets, e.g., web services and cloud. The CIS market is dominated by transient services as both consumers and providers are dynamic in space and time. Consumer requests are usually long-term and demand continuity in service provision. We propose a novel egalitarian transient service composition framework from the CIS market perspective. We apply a Dynamic Bayesian Network to model the dynamic service provision behavior of the providers. The proposed framework transforms the composition of transient services into a multi-objective temporal optimization, i.e., providing continuous services to the maximum number of consumers, and minimizing the consumers’ cost of service usages over a long-term period. We incorporate a Pareto-based genetic algorithm to enable the fair distribution of services among the consumers. Experimental results prove the efficiency of the proposed approach in terms of continuous availability of service as well as fair distribution among consumers.

Sentiment Analysis in the Software Engineering community aims to make the development and maintenance of software a better experience by helping provide code and library suggestions, defect-related comments for source code, etc.The manual finding of sentiment-based comments may be an inaccurate prediction and a time-consuming process.Automating the sentiment analysis process by leveraging Machine Learning models can benefit software professionals by giving them insights into other developers and feelings about software products, libraries, development, and maintenance tasks at a glance.This study aims to develop software sentiment prediction models based on comments by (1) identifying the best embedding techniques to represent the word of the comments, not just as a number but as a vector in n-dimensional space (2) finding the best sets of vectors using different features selection techniques (3) finding the best methods to handle the class imbalance nature of the data, and (4) finding the best architecture of deep-learning for the training of models.The developed models are validated using 5fold cross-validation with four different performance parameters: accuracy, AUC, recall, and precision on three different datasets.The experimental finding shows that the models developed using the word embeddings with feature selection using Deep Learning classifiers on balanced data can significantly predict the underlying sentiments of textual comments.

Soccer is one of the most complex and dynamic games. It is challenging to figure out the game’s pattern in real-time. We propose a novel network metric and entropy-based live soccer analytic framework (NMELSA) that identifies the opponent team’s tactics in a live soccer match by observing all the events until the specified minute of the game. We design a live game replacement model which recommends substitute players based on the on-field players’ live game ratings. Experimental results on a real-world dataset demonstrate the efficiency of our proposed approach.

Abstract Requirements engineering (RE) has been developed as a discipline to identify and then translate stakeholders’ needs into system requirements. Hence, RE is used to produce a set of specifications for developing a software system. The specifications can be applied to satisfy stakeholders and can be implemented, deployed and maintained by using their alternative design options. The past several years have seen significant improvements in RE, whereby the discipline supports the modelling and analysis of stakeholders’ goals (objectives) beyond merely incorporating these goals. Goals further help in deriving functional and non-functional requirements (NFRs) of a system. Goals play an important role in the RE process by helping elaborate the requirements. Goal-oriented requirements engineering (GORE) refers to the use of goals in RE for eliciting requirements. GORE is then used in the process of elaboration, organization, specification, analysis, negotiation, documentation and evolution of the elicited requirements. To model the software system requirements, GORE is implemented by using goals in view of goal models. Stakeholders’ goals are then represented through these goal models to assess their non-functional needs. We developed a technique for analysing conflicting goals of inter-dependent actors in a goal model. In this proposal, to ascertain stakeholders’ NFRs, we applied the cost-effectiveness analysis (CEA) to a multi-objective optimisation model in the i* goal model. This optimisation model can handle large, sophisticated systems. The requirements analyst can use information derived from the input data. The CEA further facilitates the requirements analyst by including the sensitivity of conflicting goals in the i* goal model. Based on the inter-dependency relationships, the proposed approach includes the optimisation of each objective function. This approach also uses sensitivity analysis based on the economic evaluation of derived optimal values to prioritize design options. The most cost-effective design option can hence be chosen and used to further the aim of achieving conflicting goals. This proposal uses a Telemedicine System case study, making evaluations through a simulation-based analysis.

Managing prosumers participation in energy trading becomes complicated, as the number of prosumers in smart grid is expected to rise. In this situation, it is safe to group active prosumers into groups or coalitions. By utilising analytical tools, like game theory, the method for grouping or coalition formation of prosumers can be carefully studied. After analysing prosumers behaviour in energy profile, we propose a novel dynamic, decentralised prosumer coalition formation method based on game theory that can satisfy utility grid demands. A conceptual framework is developed for coalition formation and energy management among prosumers in a dynamic environment. This approach has the potential to increase the size and number of coalitions without a central controller to better manage renewable energy usage and trading with energy buyers. We validate the performance of adopting a game theoretic approach, assessing the strategy using open and shared data sources.

Abstract Goal-oriented requirements engineering frameworks are used to model stakeholders’ objectives and requirements using goals. In a real-time environment, stakeholders’ requirements may have opposing objectives. Hence, a novel framework is needed that captures the real issues in order to achieve multi-objective optimization of inter-dependent actors. For obtaining an optimum strategy for inter-dependent actors in the i* goal model, a multi-objective two-person zero-sum game theory-based approach is applied in this paper, by balancing the opposing goals reciprocally. The proposed approach involves the generation of each objective function based on the inter-dependency relationships, the creation of decision pay-off matrices based on the objective function values and their variation to a final decision pay-off matrix. A Maxmin solution is formulated for the multi-objective game model, in which the optimization problem for each player is a linear programming problem. Finally, the most desirable strategies and their proportion values are found. By integrating Java with the IBM CPLEX optimization tool, a simulation model based on the proposed method was developed. A successful evaluation was conducted on various case studies from the existing literature. Evaluation results indicate that the developed simulation model helps users to choose an optimal alternative design option feasible in real-time competitive environments that have goals with opposing objectives.

In the new era of genetic epidemiology, there have been growing interest in studying genetic variants and their associations to complex diseases. Advances in modern computational approaches have led to the search for useful interacting genetic variants that are associated to the manifestation of a disease. However, these conventional strategies face number of challenges in predicting interesting interactions when data acquisition and dimensionality increases. Deep learning promises empirical success in number of applications including bioinformatics to drive insights of biological complexities. A deep neural network was previously proposed to identify true causative two-locus SNP interactions. The method was evaluated on various simulated and real datasets. In this study, the performance of the previously proposed deep learning method is maximized by improving network learning and avoiding overfitting. The method is further extended for performing sensitivity analysis. The performance of the method is evaluated on chronical dialysis patient's data for identifying higher-order interactions. It was observed that the highly ranked two-locus and three-locus SNP interactions in mitochondrial D-loop has the highest risk for the manifestation of disease.

In requirements engineering (RE), goal models have been employed to represent stakeholder objectives and to decide on suitable functional (goal) requirements, from among the system requirements. A large number of goal analysis procedure both qualitative and quantitative have been proposed for the selection of alternative requirements and goal achievement. All of these procedures perform goal analysis by considering the non-functional (softgoals) requirements with objective function of same nature, such as the maximising nature. However, there are real-time situations, where stakeholder’s requirements have opposing objectives/requirements (one to be maximized and other to be minimized). Hence, there is a need for a goal analysis procedure, which can select an alternative design option in situations where there is a goal model with goals of opposing objective functions. In this paper, the game theory (GT)-based approach has been proposed to perform the analysis of goals with opposing objective functions. A tool for the GT-based goal analysis has been developed in Java, integrated with the IBM Cplex optimization tool and evaluated with the adapted goal models from the existing RE literature. The results of this evaluation indicate that the proposed approach assists in the selection of alternatives in real-life situations where there are goals with opposing objective functions.

In this era of genome-wide association studies (GWAS), the quest for understanding the genetic architecture of complex diseases is rapidly increasing more than ever before. The development of high throughput genotyping and next generation sequencing technologies enables genetic epidemiological analysis of large scale data. These advances have led to the identification of a number of single nucleotide polymorphisms (SNPs) responsible for disease susceptibility. The interactions between SNPs associated with complex diseases are increasingly being explored in the current literature. These interaction studies are mathematically challenging and computationally complex. These challenges have been addressed by a number of data mining and machine learning approaches. This paper reviews the current methods and the related software packages to detect the SNP interactions that contribute to diseases. The issues that need to be considered when developing these models are addressed in this review. The paper also reviews the achievements in data simulation to evaluate the performance of these models. Further, it discusses the future of SNP interaction analysis.

Agent-oriented technology is arguably one of the most substantial advances in software development. Several platforms and architectures have been developed to deal with its conception and the advantages that it entails. Of these architectures, the belief-desire-intention architecture is amongst the most prominent. This architecture utilizes a series of beliefs known to the agent in order to fulfil its desires given its intentions. Regardless of the agent platform or architecture, however, they all share the common trait of complexity. In sight of this, an increasing amount of research has been done in relation to model-driven approaches to their development. One such model-driven approach is the dynamic non-functional requirements-based approach. This approach utilizes a series of non-functional requirements to control the agents decision-making process, using the extended non-functional requirements framework as its underlying model. This paper presents the inclusion of a unique change management system through out this model-driven approach. This system is capable of categorizing changes, assigning them priorities based on various dynamics and determining their optimal incorporation time. To our knowledge, this is the first time that a change management model has been used in relation to model-driven agent creation. This approach is verified using a randomized empirical-based evaluation.

A government organisation's capability is reliant on its public service offerings in parallel with acceptable satisfaction level of its service recipients. Government organisations utilise a Citizen's Charter in order to furnish the public with comprehensive details of their service offerings that defines how their overall organisational goals are achieved. Studies on the implementation of Citizen's Charter in developing countries indicated that important social factors were not considered. Social factors can accelerate service delivery with the use of i* in goal-oriented modelling as it represent conditions expected from social actors and their social dependencies. By modelling role dependencies among actors carrying out services in a service area and applying vulnerability and criticality metrics, problem service areas were identified using the Citizen's Charter as a source of these methodologies. Improvements in these problem service areas were based on vulnerability and criticality levels of their corresponding actors. Recommendations to address these service areas include monitoring of key performance indicators of actors and task delegation when necessary.

Requirement analysis involves elicitation of suitable functions or operations and relevant data to support software. A requirement analyst examines different alternative options to decide on an optimal alternative option that benefits the stakeholders of the system. The decision making of alternative design option is complicated by the unavailable or incomplete and imprecise input data. Optimisation, an operation research technique, can be used as a method to solve this problem. The goal-oriented framework, such as i* is used to present social models for the analysis of a software system during the early phase of the requirement's engineering process. This paper aims to develop an optimisation model for the i* goal models, using multi-objective optimisation. The optimisation model aims to fully automate the goal analysis and to handle large goal models. A simulation for the proposed approach was developed by integrating Visual C++ with Matlab and was evaluated with case studies from the existing literature. The evaluation results show that the proposed approach is feasible and offers guidance in the decision making of alternative options.

Approximate query processing with relatively small random samples is an effective way to deal with many queries on large databases. However, small random samples might miss relevant records for highly selective queries due to insufficient coverage. A multidimensional index tree called the k-MDI was proposed as an effective sampling scheme for highly selective decision support queries. It has been shown to support a fast response time and high accuracy, whereas implementation of the k-MDI on database tables was not discussed. This paper proposes the Multidimensional Cluster Sampling View based on the k-MDI. The view can be implemented with ease using common database tables and can be manipulated by SQL statements. Furthermore, it is able to provide trustable approximate answers quickly for any query condition. The response time and accuracy of approximation are validated on a large dataset based on TPC-DS specifications.

In requirement analysis, goal models play an important role in assessing alternative design options of a software system.Many qualitative and quantitative goal reasoning approaches have been proposed for goal models such as Knowledge Acquisition in Automated Space (KAOS), Non-Functional Requirements (NFR), and Goal Oriented Requirement Language (GRL).However, for i* goal model only qualitative reasoning has been proposed in Requirement Engineering literature.The aim of this paper is to present a quantitative goal reasoning for i* goal model.The proposed approach was validated with case studies from existing literature and offers a guide in the decision process.To support the validation a simulation was developed in Visual C++.

Non-functional or quality requirements such as, performance, timeliness and security are often crucial for the success of a software system. Several well-known techniques and frameworks have been developed to deal with the functional aspect of requirements engineering. Recent years have seen the emergence of frameworks that incorporate non-functional requirements (NFRs). The NFR Framework is a qualitative method that bridges the gap between the idea of NFRs, and a software design that encompasses these ideas. The framework functions by modelling NFRs and the associated implementation methods, eventually resulting in a fully documented decision regarding the implementation of said methods. This paper presents a formal linear programming optimization model for the NFR Framework with regard to operationalization selection. The optimization model has the capability of handling large, complicated graphs that were unwieldy in the original framework. The inclusion of a sensitivity analysis expands the functionality of the optimization model to provide useful data on even the smallest of problems. These additional data allow the optimization model to assist in conflict resolution with regard to the initial quantitative values. The approach is illustrated using two case studies from the literature and verified through a simulation-based analysis.

There have been many studies that depict genotype-phenotype relationships by identifying genetic variants associated with a specific disease. Researchers focus more attention on interactions between SNPs that are strongly associated with disease in the absence of main effect. In this context, a number of machine learning and data mining tools are applied to identify the combinations of multi-locus SNPs in higher order data. However, none of the current models can identify useful SNP-SNP interactions for high dimensional genome data. Detecting these interactions is challenging due to bio-molecular complexities and computational limitations. The goal of this research was to implement associative classification and study its effectiveness for detecting the epistasis in balanced and imbalanced datasets. The proposed approach was evaluated for two locus epistasis interactions using simulated data. The datasets were generated for 5 different penetrance functions by varying heritability, minor allele frequency and sample size. In total, 23,400 datasets were generated and several experiments are conducted to identify the disease causal SNP interactions. The accuracy of classification by the proposed approach was compared with the previous approaches. Though associative classification showed only relatively small improvement in accuracy for balanced datasets, it outperformed existing approaches in higher order multi-locus interactions in imbalanced datasets.

This paper presents a short case study on the implementation of SAP's cloud based solution SAP Business By Design (ByD) in one of the largest state department within Australia. Software as a Service (SaaS) has its own implementation challenges from data migration perspective. This paper provides brief detail of the data migration process followed including lessons learned during the migration activities.

In this paper, we consider the problem of robust face recognition using color information. In this context, sparse representation-based algorithms are the state-of-the-art solutions for gray facial images. We will integrate the existing sparse representation-based algorithms with color information and this integration can improve the previous performances significantly. Furthermore, we propose a new performance metric, namely the discriminativeness (DIS) to describe the recognition effectiveness for sparse representation algorithms. We find out that the richer information in color space can be used to increase the DIS, i.e. enhancing the robustness in face recognition. Extensive experiments have been conducted under different conditions, including various feature extractors, random pixel corruptions and occlusions on AR and GT databases, to demonstrate the advantages of using color information in robust face recognition. Detailed analysis is also included for each experiment to explain why and how color improve the robustness of different sparse representation-based methods.

We present an algorithm that uses a low resolution 3D sensor for robust face recognition under challenging conditions. A preprocessing algorithm is proposed which exploits the facial symmetry at the 3D point cloud level to obtain a canonical frontal view, shape and texture, of the faces irrespective of their initial pose. This algorithm also fills holes and smooths the noisy depth data produced by the low resolution sensor. The canonical depth map and texture of a query face are then sparse approximated from separate dictionaries learned from training data. The texture is transformed from the RGB to Discriminant Color Space before sparse coding and the reconstruction errors from the two sparse coding steps are added for individual identities in the dictionary. The query face is assigned the identity with the smallest reconstruction error. Experiments are performed using a publicly available database containing over 5000 facial images (RGB-D) with varying poses, expressions, illumination and disguise, acquired using the Kinect sensor. Recognition rates are 96.7% for the RGB-D data and 88.7% for the noisy depth data alone. Our results justify the feasibility of low resolution 3D sensors for robust face recognition.

A long standing problem in software engineering is inadequate requirements elicitation, analysis, specification, validation and management. The lack of well defined requirements is one of the major causes of project failure. Several well-known techniques and frameworks have been developed to deal with the functional aspect of requirements engineering. Recent years have also seen the emergence of frameworks that incorporate non-functional requirements. The Non-Functional Requirements (NFR) Framework models non-functional requirements and associated implementation methods. This paper presents a process-orientated, lightweight, quantitative extension to the NFR Framework; focusing on providing quantitative support to the decision process and how decisions affect the system.

Finding a good color space is one of the main research goals for color face recognition. Existing research shows that RGB can improve over gray-scale, while some other color spaces (YQCr for instance) can improve over RGB. However, all developed color models consist of only three color components transformed linearly from RGB. Since three colors may not capture sufficient information for solving complex face recognition problems and non-linear transformed colors usually encode very different information, this paper investigates the feasibility and effectiveness of using more than three colors including some non-linear color spaces. We propose a novel color combination algorithm namely the Multiple Color Fusion (MCF) model to utilize multiple colors. Experiment 4 on FRGC2 is conducted to demonstrate the effectiveness of MCF. In particular, MCF outperforms any existing three-color based methods by at least 3% and improves over RGB by 8%.

Agent-oriented conceptual modeling notations such as i* represents an interesting approach for modeling early phase requirements which includes organizational contexts, stakeholder intentions and rationale. On the other hand, Unified Modeling Language (UML) is suitable for later phases of requirement capture which usually focus on completeness, consistency, and automated verification of functional requirements for the new system. In this paper, we propose a methodology to facilitate and support the combined use of notation for modeling requirement engineering process in a synergistic fashion. For organizational modeling/early phase requirements capturing we use the i* modeling framework that describes the organizational relationships among various actors and their rationales. For late (functional) requirements specification, we rely on UML activity diagram.

Business processes represent the operational capabilities of an organization. In order to ensure process continuity, the effective management of risk becomes an area of key concern. In this paper we propose an approach for supporting risk identification with the use of higher-level organizational models. We provide some intuitive metrics for extracting measures of actor criticality, and vulnerability from organizational models. This helps direct risk management to areas of critical importance within organization models. Additionally, the information can be used to assess alternative organizational structures in domains where risk mitigation is crucial. At the process level, these measures can be used to help direct improvements to the robustness and failsafe capabilities of critical or vulnerable processes. We believe our novel approach, will provide added benefits when used with other approaches to risk management during business process management, that do not reference the greater organizational context during risk assessment.