Mohd Hilmi Hasan

Monitoring web services' quality of service: a literature review

Monitoring Quality of Service (QoS) compliance is an important procedure in web service environment. It determines whether users’ expectations are met, and becomes the vital factor for them to decide whether to continue paying for the service or not. The monitoring is performed by checking the actual services performance against the QoS stated in Service Level Agreement (SLA). In relation to that, the need for monitoring vague QoS specifications in SLA has become more apparent nowadays. This paper reviews the published literature on web services QoS monitoring. A total of 60 selected articles were systematically analyzed. There were 23 of the articles selected through restrictive search criteria while the other 37 were selected based on unrestrictive search criteria. The review shows that little evidence exists on monitoring vague QoS specifications of web services. Providing ability for monitoring QoS that is specified vaguely in SLA could give new insights and implications to web services field. This paper concludes with some recommended future works to construct the theory and perform the empirical research.

Automated notification and document downloading in E-learning - development of an agent-based framework utilizing the push-pull technology interaction policy

This paper describes the processes of document downloading and announcement in the e-learning is automated via a proposed agent-based framework. In the framework, the interaction policy for the involved agents is based on combination technique which combined the push and pull mechanisms. The agentbased framework consists of four agents: search agent, priority agent, notify agent and download agents that are applied in the e-learning domain. To demonstrate the working of these agents within the e-learning domain, three scenarios with different level of complexity involving one course/subject are presented. In general, these agents assist the students to get prompt notifications every time new documents are available in the course registered. Thus, eliminating the students from not knowing that new documents have been uploaded by the lecturers. Moodle, the e-learning domain is chosen since Moodle is an open-source Content Management System that encourages any developers to contribute new enhancement of the existing feature or create the new features for the benefits of Moodle users.

Development of Web Services Fuzzy Quality Models using Data Clustering Approach

This paper presents the fuzzy clustering of web services’ quality of service (QoS) data using Fuzzy C-Means (FCM) algorithm. It was conducted based on actual QoS data gathered from the network. The work involved three data sets that represented three different QoS parameters. Each data set contained 1,500 data points. The clustering was validated using Xie-Beni index to ensure that it performed optimally. As a result, three fuzzy quality models were produced that represented the three QoS parameters. The work implies potential new findings on fuzzy-based web services’ applications, mainly in reducing computational complexity. The work also benefits the less technical-knowledgeable requestors as the fuzzy quality models can guide them to find services with realistic QoS performance. For future work, the fuzzy quality models will be employed in web services’ QoS monitoring application. They will also be equipped with an adaptive mechanism that supports the dynamic nature of web services.

Formal Verification for Interaction Protocol in Agent-Based E-Learning System Using Model Checking Toolkit - MCMAS

This paper presents a formal verification done for interaction protocol in agent-based e-learning system using a model checking toolkit – MCMAS (Model Checking Multi-Agent System). The goal of this interaction protocol is to automate the document downloading and notification in e-learning on behalf of the students. The specification of the interaction protocol for each agents are translated into Interpreted Systems Programming Language (ISPL) file as the programming language for MCMAS and a combination of temporal logic operators like Computation Tree Logic (CTL) and Linear Temporal Logic (LTL) are used to verify the formula for each tested reachable property. The purpose of executing this formal verification is to convince that this interaction protocol is sound and reachable for each state. Overall, this paper describes the idea of agent-based e-learning system, MCMAS toolkit - used as a model checker, CTL and LTL – logics in verification used, ISPL – programming language used under MCMAS platform and the results derived from the running verification.

Development of Multimedia Messaging Service (MMS)-based receipt system for mobile banking

Mobile devices are becoming more important in daily lives and they also have gained increasing acceptance in banking services, for example Short Messaging Service (SMS)-based m-banking system. In this particular system, an SMS receipt will be delivered to users for any transactions done. Though we could gain many benefits out of it, SMS is considered as informal receipt for banking services which could not be used as reference in nowadays formal businesses or transactions. Therefore, this paper proposes a Multimedia Messaging Services (MMS)-based receipt system for m-banking that produces transaction receipt in PDF document and sends it to users via MMS.

Utilizing Electronic Institution for protocol specification in agent-based e-learning system

This paper describes the specification of interaction protocol for automating the document downloading and notification in the e-learning by adopting the Electronic Institution (El) infrastructure. This interaction protocol is designed using specific tool called ISLANDER editor which is one of the Electronic Institution Integrated Development Environment (EIDE) components. The idea of designing this interaction protocol is believed can eliminate the tendency of students from missing out the latest updates available in their e-learning. Furthermore, with the adoption of agent-based environment for e-learning will add a new feature to the existing e-learning in a way of assisting a busy lifestyle of students to hand over the task of downloading lecture materials and checking the latest announcements available in the e-learning to the software agent straight to their desktop. Hence, the students will not miss any important announcements or latest learning materials uploaded by their lecturers. In this paper, we are only focusing and describing the process of specifying interaction protocol using ISLANDER editor.

A Rough-Fuzzy Inference System for Selecting Team Leader for Software Development Teams

Inappropriate team composition is one of the important factors that can impact the overall process of software development. Numerous models for team composition have also been suggested, yet they have been disapproved by the researchers and organisations for having ineffectiveness in yielding positive results. Therefore, this study proposes a rough-fuzzy model for selecting team leader for software development teams to avoid the limitations of individual techniques (i.e., Rough Set Theory (RST) or Fuzzy Set Theory (FST)). Moreover, the model development was divided into two portions: Decision Rules Development and Fuzzy Inference System (FIS) development. Johnson Algorithm (JA) was applied using ROSETTA toolkit under rough set theory principles for decision rule construction. Decision rules were then used under Mamdani’s fuzzy inference method. At the end, the developed model was validated based on the results of prediction accuracy and F1-measures.

Interaction protocol for agent-based e-learning system

This paper presents a development of interaction protocol specification for automating the document downloading and notifying in e-learning system. From agent-roles diagram, it is now being transformed into its illocution schema as the real conversation message that three main agents consisting of e-learning agent, lecturer agent and student agent will convey when performing their goals. This interaction protocol is defined using ISLANDER editor as the graphical verification tool that will check the correctness of the interaction protocol defined using the Electronic Institution (EI) approach. The process of this interaction protocol specification will be explained in detail. Preliminarily, this interaction protocol specification is concentrating on the one-to-one agent interaction first since it is assumed to be a generic template for complex scenario soon. Once this interaction protocol is specified correctly, then it is ready to be monitored using its monitoring tool to check its execution. This paper describes the interaction protocol specification using Electronic Institutions Integrated Development Environment (EIDE) methodology.

A Fuzz Logic-Based Cloud Computing Provider’s Evaluation and Recommendation Model

The growing number of cloud computing providers has made the selection process difficult and time consuming. This motivates many research to propose cloud computing recommendation system. However, generating an overall rating of a provider based on accumulation of several performance parameters involves uncertainties especially when the computation comprises network-related performance parameters. Hence, this paper proposes a fuzzy logic-based model that evaluates providers based on downlink, uplink and latency parameters, and determines the best provider. The model was constructed based on historical data of the real cloud computing providers implementation. The model was compared against non-fuzzy model, and it showed that the proposed fuzzy logic-based model is better in terms of accuracy. This work implies faster and accurate selection of cloud computing provider.

Algorithms for frequent itemset mining: a literature review

Data Analytics plays an important role in the decision making process. Insights from such pattern analysis offer vast benefits, including increased revenue, cost cutting, and improved competitive advantage. However, the hidden patterns of the frequent itemsets become more time consuming to be mined when the amount of data increases over the time. Moreover, significant memory consumption is needed in mining the hidden patterns of the frequent itemsets due to a heavy computation by the algorithm. Therefore, an efficient algorithm is required to mine the hidden patterns of the frequent itemsets within a shorter run time and with less memory consumption while the volume of data increases over the time period. This paper reviews and presents a comparison of different algorithms for Frequent Pattern Mining (FPM) so that a more efficient FPM algorithm can be developed.