Web Information Systems and Technologies

In the last few years, in spite of concerns about the hype, cloud computing has expanded steadily both horizontally – across industries, and vertically – in organizations’ technology stacks. Most technologies that enable cloud services existed prior to cloud computing’s existence, although these days they rejuvenate, evolve and stimulate the computational ecosystem transformations. Actually the radical change for organizations is in rethinking and reengineering their traditional IT resources advancing them with cloud architectures and implementing services based on cloud computing delivery models. The change is underway on a large scale: from vendors and developers to providers and customers, and the key issues of “cloudiness” are not only in economics and management, but in provisioning, interoperability and security of the integrated services.

The Cloud Computing phenomenon likewise creates exciting challenges and opportunities for the entire educational system. For faculty, students, administration, and IT professionals it is a thrilling journey driven by many agendas–cost cutting, delivering dynamic mobile and interactive computational services, utilizing and leveraging integrated IT infrastructures and systems. This talk will explore the impact of cloud computing on the educational socio-technical system and will provide the author’s experience in strategizing and utilizing cloud-based applications and services.

High performing firms are working in business networks with advanced decision making capabilities. Decision making in business networks is a new research area that provides knowledge and insight about how decision rights are allocated and how decision processes are designed and implemented in evolving business networks [22]. In this article we focus on a particular type of support: software agents. Software agents are software programs that act on behalf of users or other programs. Software agents can be autonomous (capable of modifying the way in which they achieve their objectives), intelligent (capable of learning and reasoning), and distributed (capable to being executed on physically distinct computers). Software agents can act in multi-agent systems (e.g. distributed agents that do not have the capabilities to achieve an objective alone and thus must be able to communicate) and as mobile agents (e.g. these relocate their execution onto different processors). Recent research shows that software agents are able to act as a decision support tool or a training tool for negotiations with people. For example, [16] Lin and Kraus (2010) identified several types of agents in several variations of negotiation settings. These agents differ in the number of negotiators, encounters, and attributes they can handle. The identified agents are: Diplomat, AutONA, Cliff-Edge, Colored-Trails, Guessing Heuristic, QOAgent, and Virtual Human. Although software agents are popular in scientific research programs, the use of software agents in real life business situations is limited. We will explore the use of software agents in the flower industry with its complex logistics, commercial, and financial processes on a global scale.

Several approaches have been proposed to deal with the web service matchmaking problem. Unfortunately, most of these solutions are purely syntactic measures based on the input/output interface specifications of web services and consequently lake accuracy. This is a serious drawback in a fast growing Internet that is facing the challenge to deal with an increasing number of published services. The proposed solutions to cope with this limitation consider the process part of a service description as a graph in the similarity measure. This kind of solutions has better accuracy but suffer from high computational complexity because they rely on time consuming graph matching tools. To avoid this heavy time computing overhead, we propose in this paper a solution that decomposes the process graph into smaller subgraphs and construct similarity of web services based on the similarity of their subgraphs. Simulation results show that our solution is both accurate and fast to compute.

This study characterizes distributed XML processing on networking nodes. XML documents are sent from a client node to a server node through relay nodes, which process the documents before arriving at the server. When the relay nodes are connected in tandem, the XML documents are processed in a pipelining manner. When the relay nodes are connected in parallel, the XML documents are processed in a parallel fashion. We evaluate distributed XML processing with synthetic and realistic XML documents. Well-formedness and grammar validation pipelining and parallel processing characterization reveals inherent advantages of the parallel processing model.

In this paper, we present an approach to enable collaborative endpoint network for mobile phones and devices. In particular, we expose functions on mobile phones as REST web services and make mobile phones as web service providers for rapid integration with communication and collaboration applications. To accommodate the distinct features and constraints in mobile environment, this paper describes a lightweight and efficient protocol, Compact HTTP, which consists of a small subset of HTTP 1.1 to reduce the footprint of REST services. We expand bindings of HTTP to multiple messaging protocols, including XMPP, and make the REST services invariant to network and protocol changes. These expanded bindings introduce asynchrony into REST, a desired property for communication and collaboration services. Furthermore, HTTP over XMPP described in our approach introduces the concept of hyperlink presence in collaboration, and it is used to mitigate the broken link issue which is acute in mobile environments as devices and resources can come and go in an unpredictable pattern. To provide end-to-end message security, a symmetric key based security scheme is described for service authentication and authorization. A prototype system based on the proposed approach is developed that allows both local operators and remote directors to control and monitor resources in a secure manner on Android phones and devices, e.g. camera, camcorder, audio, location, telephony, motion, power, etc. Experimental results indicate that the proposed approach is feasible, lightweight, and has satisfactory performance.

Today, we observe a continuous expansion in the use of Internet technologies, mainly Web services, for electronic B2B transactions. This has triggered an increase in the number of companies’ Web services registries. In this context, Web service discovery can be a cumbersome task for a service requester and a costly one for a discovery system. To deal with this issue, one obvious solution is to group Web service registries into communities. Due to the dynamic nature of services oriented environments, such an organization should be managed to maintain its organizational consistency. In this paper, we specify the needed management operations to ensure the communities consistency during a registry/community life-cycle. To test the feasibility of our management approach, we simulate a network of registry communities and develop a community manager.

Whenever a restricted bandwidth is the bottleneck of an application that transfers data from a server hosting a huge XML document to a client with limited resources, there exist mainly two approaches to overcome this problem: caching previously downloaded data on the one hand, and transferring compressed data on the other hand. We present a combined approach that combines both approaches, i.e., that proposes a caching framework working with compressed XML data. Our performance evaluation demonstrates that this combination not only yields a benefit in terms of the data to be transferred, but it also yields a shorter response time for queries to be evaluated on the client. The benefit reached by the combination of caching and compression is even stronger than the sum of the benefits reached by each of these techniques alone.

Usability is an important factor influencing users’ interaction and adoption of e-government, which needs to be considered in e-government development. However, current research has not paid much attention to assessing usability. This empirical study conducts a systematic approach to evaluate the usability of current e-government websites, identifying existing usability problems and providing proposed design solutions for further e-government usability development. Such research results can help designers and developers to understand the importance of usability and guide their particular attention to develop more usable e-government.

Open Source Software (OSS) communities do not often invest in marketing strategies to promote their products in a competitive way. Even the home pages of the web portals of well-known OSS products show technicalities and details that are not relevant for a fast and effective evaluation of the product’s qualities. So, final users and even developers who are interested in evaluating and potentially adopting an OSS product are often negatively impressed by the quality perception they have from the web portal of the product and turn to proprietary software solutions or fail to adopt OSS that may be useful in their activities. In this paper, we define OP2A, an evaluation model and we derive a checklist that OSS developers and web masters can use to design (or improve) their web portals with all the contents that are expected to be of interest for OSS final users. We exemplify the use of the model by applying it to the Apache Tomcat web portal and we apply the model to 47 web sites of well-known OSS products to highlight the current deficiencies that characterize these web portals.

Nowadays, SME have to take into account new business and management approaches, like CRM, as well as the support provided by ICT. In this scenario, SME would take advantage from a Web-based service like ARNEIS, supporting an intelligent matching between supply and demand of CRM-related tools. Such a service needs a detailed knowledge base, and a friendly user interface (UI) enabling users to interact with the formal knowledge base. In this chapter we claim that both the knowledge representation and the user interaction mechanisms have to be based on a domain analysis (how users talk about CRM). The chapter presents three UI, based on on-line forms, business processes, and natural language. All three exploit semantic templates, i.e., formal representations of dialog topics (key concepts in the descriptions of CRM activities). The proposed approach suggests a general solution to enable users to enter formal representations in systems based on semantic technologies.

Working with data can be very abstract without a proper visualization. Yet, once the data is visualized, it presents a dead end, so the user has to return to the data level to make enrichments. With Vizgr (vizgr.org), we offer an elegant simplification to this workflow by giving the opportunity to enrich the data in the visualization itself. Data, e.g. statistical data, data entered by the user, from DBpedia or other data sources, can be visualized by graphs, tag clouds, on maps and in timelines. The data points can be connected with each other, with data in other visualizations and any web address, regardless of the source. It allows users to make data presentations without changing to the data level, once the data is in the system. In an evaluation, we found that over 85% of the participants were able to use and understand this technology without any training or explicit instructions.

With the growing amount of information available on the Web, the task of retrieving documents of interest has become increasingly difficult. Personalized search has got significant attention because it considers the user’s preference into the search process to improve the results. However, studies have shown that users are reluctant to provide any explicit input on their personal preference. In this study, we investigate how a search engine can elicit users’ preferences by exploring the user’s tagging activity from various sources. We propose a simple yet flexible model to succinctly represent user preferences based on multiple factors. Our experiments show that users’ preferences can be elicited from a multi-factor tagging data and personalized search based on user preference yields significant precision improvements over the existing ranking mechanisms in the literature.

Most educational centers are currently using e-learning tools to provide the pedagogical resources for the students, especially in higher education. Nevertheless, some students are not able to access to this information because these authoring tools are not as accessible as they should be. The main aim of this paper is to evaluate if one of the most widely e-learning tool used around the world, Moodle, is accessible for visually impaired people. The evaluation shows that the accessibility guidelines provided by the World Wide Web Consortium (W3C) are not accomplished by the tool. Moreover, it shows that people using screen readers are not able to access to the majority of the functionality of Moodle.