Recommender systems aim to provide users with personalised recommendations of items based on their preferences. Such systems have during the last 15 years been applied in many domains and have enjoyed an increased popularity both in research communities and commerce. In this thesis our overlying aim is to work towards creating a recommender system for tourists visiting Trondheim. We begin this work by addressing the cold-start user problem, which is the problem of giving high-quality recommendations to new users who the system has little or no information about. The problem is severe in the tourist domain where the majority of users are cold-start users. To properly address the problem, we present a systematic literature review of the recommender system literature identifying nine types of solutions to the cold-start user problem. We evaluate the solution types in context of the tourist domain, and find that using demographic user data is the best solution in this domain. We include this solution as a part when we propose a design of a location-aware Bayesian recommender system for tourists visiting Trondheim.

Bruken av åpen kildekode i systemutviklingsprosjekter har blitt mer utbredt og utviklingstiden til slike prosjekter er ofte kortere enn prosjekter som ikke benytter seg av ferdigkomponenter. Det er stor uenighet i hvorvidt åpen kildekode er mer eller mindre sikkert enn lukket kildekode, eller om det er noen særlig forskjell i det hele tatt. I dette prosjektet har jeg sett på hvilke krav som stilles til sikkerhet i systemer som behandler helse- og personopplysninger. Jeg har forsøkt å realisere disse kravene med utstrakt bruk av åpen kildekode-komponenter i utviklingen av et egenomsorgssystem for overvektspasienter. Jeg har sett på hvilke sikkerhetsutfordringer dette fører med seg, og hvilke tiltak som kan brukes for å redusere dem.Gjennom mine erfaringer i forbindelse med utviklingsprosessen i denne studien kan jeg se fordelene ved å bruke fri programvare i motsetning til tradisjonell systemutvikling, og det hadde ikke vært mulig å realisere systemet i samme grad med en tradisjonell systemutviklingsprosess hvor alt blir utviklet fra bunnen av. De største utfordringene jeg møtte på var knyttet til integreringen av komponentene og utviklingen av en sømløs felles autentisering. For å redusere sikkerhetsutfordringene bør det gjøres en tidlig vurdering angående valg av komponenter og integrering av de valgte komponenter for å kartlegge muligheter og begrensninger som komponentene legger på systemet.

<p>QUality, Interoperability and Standards in e-Learning (QUIS) is an EU funded project, and is a part of the e-Learning program . The projects contract number within the e-Learning program is 2004-3538/001-001 ELE-ELEB14. The project has partners from Norway, Sweden, Italy and Hungary. This thesis is written to be a part of the projects’ Work Package 5 – “Standards for e-Learning”. A topic map is an information structure for describing and identifying resources. It consists of the major elements topics, occurrences and associations. A topic is the representation of a real world subject. An occurrence is a source of information about the topic. Internal occurrences are information sources within the topic map file. External occurrences are links to resources located elsewhere. An association is a relationship between topics. It describes how the topics are related to each other, by giving the different members of the association “roles”. All topic map elements can be scoped to give them extra semantic value. Two or more topic maps can be merged to form one large topic map. Topic maps usually follow the XML for Topic Maps (XTM) standard. In addition to being an information structure, the topic map standard also tries to overcome the so called “Web Identity Crisis”. The Web Identity Crisis is based on the statement that “there is no way of knowing what references on the World Wide Web identify today”. The Semantic Web is a similar technology to topic maps. It is based on the Resource Description Framework. QUIS topic map is a topic map made to represent subjects in the area of e-Learning standardization. It is made in a “topic map constructor system” based on the Topic Map 4 Java API. A “topic map application system” can be used to display the content of the topic map. It is shown that this approach offers easy and structural access to subjects within this area of interest. In comparison with regular web portals, the topic map application has several advantages: • The connection between related information is more intuitive. • The topic map is easier to maintain extend. • The topic map has better support for multiple languages. • The topic map can be merged with other topic maps. In order to exploit the full strengths of the topic map standard, more e-Learning communities have to take it into use. The creation of Published Subject Identifiers (PSI) will advance this process.</p>

<p>In the last years digital home solutions have made their entry and are are now becoming mainstream. From a service provider's point of view, this creates an interesting opportunity. Today, if a service provider wishes to change the services offered, the flash memory of the customers' Set-Top Boxes must be updated or the Set-Top Boxes must perhaps be replaced for larger updates. Replacing the Set-Top Box is both costly and time consuming. As digital home solutions become more normal, it is possible to use such a system to offer services. This way adding, removing or updating a service can be done in software only. Our proposed architecture does this, and with minimal involvment of the customers. The architecture offers a generic interface, providing for third party development. As the world gets more and more digitalized, the expectation of everything to be availabe from everywere gets more common. Our architecture lets a service provider offer their customers remote access to, and control of, their digital home. We have based our system on Microsoft Media Center Edition. This was chosen after first studying the concept of an ideal digital home and then researching which existing digital home solution would bring us closest to this ideal situation.</p>

<p>This master thesis presents a new and redesigned version of Peer2Me, a framework for developing mobile collaborative applications on mobile phones. The first version of Peer2Me was designed and created by Carl-Henrik Wolf Lund and Michael Sars Norum in 2005, which was presented in their master thesis, The Peer2Me Framework, [31]. We evaluated their framework in our depthstudy [5], fall of 2005. The evaluation showed that the framework lacked some desired and necessary functionality, had some bugs and was a bit hard to use. This thesis also describes the history of Peer2Me along with cental concepts regarding peer-to-peer networking in an mobile ad hoc environment. There are a lot of on going and finished projects that can be related to Peer2Me. We have chosen to investigate the most interesting and relevant projects, which are presented in Chapter 11, State of the Art. Since a redesign of the Peer2Me framework was necessary, we have performed a research in the most recognized architectural tactics, design patterns and architectural patters. Before embarking on the task of designing the framework, a research in the latest technology was necessary. In our depthstudy [5], we had already performed such a research, so we only had to obtain the latest development in the related areas. Special attention was given to the Bluetooth wireless network technology. All created packages, classes and interfaces are thoroughly described along with their roles in the framework. We felt that a mere description of the modules was not enough, so we wrote Chapter 16, Design Decisions, which discusses the different crossroads we faced with during development, and the path we chose. To give the reader an impression of how the framework can be used, we also developed some applications that utilizes the new framework. Lastly we evaluated our work, compared the old and new framework, discussed the problems we encountered, answered our research questions and summarized the thesis. All source code, javadoc and a functional, new version of Peer2Me are attached along with this report.</p>

<p>Software configuration management deals with control of the evolution of complex computer systems. The ability to handle changes, corrections and extensions is decisive for the outcome of a software project. Automated processes for handling these elements are therefore a crucial part of software development. This thesis focuses on build and release management, in the context of developing a control system for the world’s biggest particle accelerator. Build and release cover topics such as build support, versioning, dependency management and release management. The main part of the work has consisted of extending an in-house solution supporting the development process of accelerator control software at CERN. The main focus of this report is on the practical work done in this context. Based on a literature survey and examining of available tools, this thesis presents the state of the art concerning build and release management before elaborating on the practical work. Based on the experience gained from the work of this thesis, I conclude with a discussion of whether or not it is beneficiary to stick with in-house solution, or if switching to an external tool could prove better for the development process implemented.</p>

<p>Security is an important aspect for all kinds of software development, but it is especially important for web applications since they usually are exposed to the Internet. Web Services offer application to application connectivity using the SOAP protocol. Web Services are quite often built as an extension to already existing applications to provide business to business communication. Since it is often necessary to expose critical business functions through the Web Services, e.g., ordering an item or sending an invoice, security in Web Services are vital for a company’s daily operations. In this project we have created an input validation framework for Web Services, to aid developers in creating more secure Web Services in an easier and more reusable manner. We have focused on creating a lightweight policy configuration based on XML, and a set of highly configurable and extendable validators. The framework is implemented in Java and is not dependent on a specific SOAP framework. To keep the framework general and compatible with multiple SOAP frameworks, we have developed a set of interceptors to support the two most common open source SOAP frameworks, Codehaus XFire and Apache Axis2. This report first presents theory and rationale behind the need for a new way of performing input validation. Further the implementation of the framework is documented together with an example application, which demonstrates an example use of the framework.</p>

<p>Mobile communication markets are known for frequent innovations with potentially high network effects. The conceptual work in economics and innovation studies show how the growth pattern for innovations in such markets could vary depending on the competition and market characteristics. However, the empirical research within this field is limited. This thesis introduces a computer simulation model for analyzing the development, adoption and diffusion of innovations in a mobile market. The model is based on Agent-Based Computational Economics (ACE), and makes use of behavioral theory of firms, economics, and sociology, to leverage the theoretical understating of the diffusion of innovations. The results of the simulation runs on the developed simulation software show that the topologies of social networks have strong effects on diffusion. However, it is also found that in situations where several competing companies launch their innovations sequentially, a winner-takes-all outcome is the most likely when the actors are completely rational. Further, when the information in the market is imperfect, the topology of social networks can create equilibria where the market is shared between several providers. Finally, the variance in consumer characteristics affects both the rate and the outcome of innovation diffusion. The thesis reaffirms that computer simulation is an effective way to combine the sociological and economic theories of innovation diffusion. The results show that there is still a need for more research on the field to better understand why some innovations fail, while other succeed and becomes accepted in the market. The outcome of an innovation launch is shown to be affected by several factors, including timing, network structure, market noise, and consumer characteristics. By using the simulation model to study the influence of such factors in a specific market, service providers may improve their competitive power.</p>

<p>Software development is a highly dynamic process, primarily caused by its foundation in the dynamic human world. Requirements traceability alleviates the detrimental effects of this dynamism by providing increased control over the artifacts of the software development processes and their interrelationships. This thesis investigates how an RT tool should be designed and implemented in order to assist with the tasks of requirements traceability, and outlines a tool that primarily focuses on reducing the work overhead associated with the tasks of implementing requirements traceability in software development projects. Preparatory to the development of the RT tool, the applicability of the traceability models presented in the in-depth study has been confirmed through empirical work. A detailed representation of the models has been compiled, elaborating on the internal representation of artifacts and traces. The models were extended to be able to represent organisational hierarchies, enabling trace information analysis to deduce the context of important decisions throughout the software development processes, an important tool in understanding how requirements are determined. The thesis presents a requirements specification and architecture with a firm foundation in the findings of the in-depth study, outlining an RT tool that addresses important issues concerning the implementation of requirements traceability, in particular focusing on reducing the associated work overhead. Based on the requirements specification and architecture, a evolutionary prototype is developed, giving its users an impression of the functionality of the outlined RT tool. The prototype addresses the issues pointed out by the requirements specification and architectural description, and, throughout development, attention is given the evolvability of the prototype. Consequently, the prototype provides a good foundation for the future development of a complete RT tool.</p>

<p>This master thesis presents a new open source framework for functional testing of Flex applications, FunFX. FunFX is a framework that encourages test-driven development. The project was initiated by BEKK due to Flex’s lack of a proper tool for functional testing. This report will focus on testing, and will describe central concepts within the area of software testing. Similar testing frameworks for other technologies will also be described and compared to FunFX. It will try to document the usefulness of such an open source tool, and try to discover what entry level this framework imposes on the user. During development, test cases using the framework will be created to be able to document both positive and negative aspects of the framework. A usage test was also arranged, to be able to document the framework better. This test session resulted in valuable information about the usability and the reliability of the created tests. The design and implementation are thoroughly described together with each class created, along with their roles in the framework. The issue of synchronization is handled as its own part. To make the implementation decisions easier to understand, the different design patterns used are elaborated. The final result is a framework that has the ability to interact with a Flex application programmatically. When used together with any test unit framework for Ruby, it is a fully functional testing tool for test-driven development. Source code, a deployable library file of the FunFX Flex adapter, and a gem of the FunFX framework, together with a Flash movie showing the framework in use can be found on the attached CD. Keywords: Framework, Flex, Ruby, Open Source, Software testing, Functional testing, Flash, Test-driven development. Keywords: Framework, Flex, Ruby, Open Source, Software testing, Functional testing, Flash, Test-driven development.</p>