No description available.

This thesis investigates the use of static, adaptive, adaptable and mixed-initiative approaches to the personalisation of content and graphical user interfaces (GUIs). This empirical study consisted of three experimental phases. The first examined the use of static, adaptive, adaptable and mixed-initiative approaches to web content. More specifically, it measured the usability (efficiency, frequency of error occurrence, effectiveness and satisfaction) of an e-commerce website. The experiment was conducted with 60 subjects and was tested empirically by four independent groups (15 subjects each). The second experiment examined the use of adaptive, adaptable and mixed-initiative approaches to GUIs. More specifically, it measured the usability (efficiency, frequency of error occurrence, effectiveness and satisfaction) in GUI control structures (menus). In addition, it investigated empirically the effects of content size on five different personalised menu types. In order to carry out this comparative investigation, two independent experiments were conducted, on small menus (17 items) and large ones (29 items) respectively. The experiment was conducted with 60 subjects and was tested empirically by four independent groups (15 subjects each). The third experiment was conducted with 40 subjects and was tested empirically by four dependent groups (5 subjects each). The aim of the third experiment was to mitigate the drawbacks of the adaptive, adaptable and mixedinitiative approaches, to improve their performance and to increase their usability by using multimodal auditory solutions (speech, earcons and auditory icons). The results indicate that the size of content affects the usability of personalised approaches. In other words, as the size of content increases, so does the need of the adaptive and mixed-initiative approaches, whereas that of the adaptable approach decreases. A set of empirically derived guidelines were also produced to assist designers with the use of adaptive, adaptable and mixed-initiative approaches to web content and GUI control structure.

Procedural content generation (PCG) can be a useful tool for aiding creativityand efficiency in the process of designing game levels. Mixed-initiative level genera-tion tools where a designer and an algorithm collaborate to iteratively generate gamelevels have been used for this purpose – taking advantage of the combination of com-putational efficiency and human intuition and creativity. However, it can be difficultfor designers to work with tools that do not respond to the common language of games:game design patterns.It has been demonstrated that game design patterns can be integrated into PCGalgorithms, but formally-defined and hierarchically-arranged game design patternshave not yet been used as a means of increasing gameplay-based control in mixed-initiative dungeon generators. We present a method for evolving dungeon rooms usingmulti-level game design patterns in the objective function of a genetic algorithm, aswell as an instantiation of this method in a mixed-initiative dungeon design tool. Ourresults show that we are able to control the frequency and type of design patterns ingenerated rooms using pattern-related input parameters, enabling us to create dungeonrooms containing a wide variety of patterns on different levels of abstraction.Results from a small-scale user study of professional game developers suggest thatthe use of game design patterns in mixed-initiative level design tools can be a promisingway of providing a good starting point when designing a level, as well as offeringmeaningful gameplay related feedback throughout the design process. We also identifychallenges that will need to be faced if game design pattern-based mixed-initiative leveldesign tools are to become a part of the game designer’s toolkit.

Writing is a notoriously time-consuming and challenging activity that is difficult to avoid during the development of a game, and the steady increase in complexity behind producing games is putting pressure on the industry to cut unnecessary costs and streamline processes. With recent breakthroughs in Neural Network research the capabilities of causal language models like the GPT models made by OpenAI have reached a level where they could be used to assist with creative assignments that previously only could be done to an acceptable level of quality by a human writer. This paper aims to combine the power of a language model with the versatility and control of the Mixed-Initiative Co-Creation approach. In order to limit the scope of the artifact to a manageable size the focus will be to generate a shorter biography with backstory for characters and items in a level made in the Evolutionary Dungeon Designer by Alvarez et al. The artifact was evaluated with a user study in which both quantitative ratings and qualitative feedback was collected. The results suggest that the artifact has potential as it has the ability to generate compelling narratives and users attested that it had a positive effect on their work.

Autonomous robots have proven to be useful for search and rescue (SAR) by being deployed in emergency situations and removing the need for direct human presence. However, there remains a need for effective communication between the robot and human operator to cooperate as a team. This thesis investigates the question of when to explain an autonomous agent’s behaviour in the setting of human-robot teaming. A game environment is developed to conduct a virtual SAR experiment to test the effect of explanation timings between two conditions: (1) always explaining - the robot provides explanations whenever possible, and (2) anticipatory explaining - the robot determines when to explain based on the context. When to explain is determined through the construction of argumentation frameworks modelling the context. The effect of anticipatory information pushing is tested on two metrics: team performance and explanation experience. Results indicate anticipatory explaining does not have a significant effect on team performance and participants’ explanation satisfaction. Additionally, participant feedback shows they prefer to be in control instead of cooperating as a team. These findings underline the importance of studying explanation presentation in high-demanding environments and indicate a need for interdisciplinary discussion on the design of human-robot teaming.

The increasing ubiquity of networked mobile devices such as cell phones and PDAs has created new opportunities for the transmission and display of multimedia content. However, any mobile device has inherent resource constraints: low network bandwidth, small screen sizes, limited input methods, and low commitment viewing. Mobile systems that provide information display and access thus need to mitigate these various constraints. Despite progress in information retrieval and content recommendation, there has been less focus on issues arising from a network-oriented and mobile perspective. This dissertation investigates a coordinated design approach to networked multimedia on mobile devices, and considers the abovementioned system perspectives. Within the context of accessing news video on mobile devices, the goal is to provide a cognitively palatable stream of videos and a seamless, low-latency user experience. Mixed-initiative---a method whereby intelligent services and users collaborate efficiently to achieve the user's goals, is the cornerstone of the system design and integrates user relevance feedback with a content recommendation engine and a content- and network-aware video buffer prefetching technique. These various components have otherwise been considered independently in other prior system designs. To overcome limited interactivity, a mixed-initiative user interface was used to present a sequence of news video clips to the user, along with operations to vote-up or vote-down a video to indicate its relevance. On-screen gesture equivalents of these operations were also implemented to reduce user interface elements occupying the screen. Semantic relevancy was then improved by extracting and indexing the content of each video clip as text features, and using a Na"ive Bayesian content recommendation strategy that harnessed the user relevance feedback to tailor the subsequent video recommendations. With the system's knowledge of relevant videos, a content-aware video buffer prefetching scheme was then integrated, using the abovementioned feedback to lower the user perceived latency on the client-end. As an information retrieval system consists of many interacting components, a client-server video streaming model is first developed for clarity and simplicity. Using a CNN news video clip database, experiments were then conducted using this model to simulate user scenarios. As the aim of improving semantic relevancy sometimes opposes user interface tools for interactivity and user perceived latency, a quantitative evaluation was done to observe the tradeoffs between bandwidth, semantic relevance, and user perceived latency. Performance tradeoffs involving semantic relevancy and user perceived latency were then predicted. In addition, complementary human user subjective tests are conducted with actual mobile phone hardware running on the Google Android platform. These experiments suggest that a mixed-initiative approach is helpful for recommending news video content on a mobile device for overcoming the mobile limitations of user interface tools for interactivity and client-end perceived latency. Users desired interactivity and responsiveness while viewing videos, and were willing to sacrifice some content relevancy in order gain lower perceived latency. Recommended future work includes expanding the content recommendation to incorporate viewing data from a large population, and the creation of a global hybrid content-based and collaborative filtering algorithm for better results. Also, based on existing user behaviour, users were reluctant to provide more input than necessary. Additional user experiments can be designed to quantify user attention and interest during video watching on a mobile device, and for better definition and incorporation of implicit user feedback.

Searching for lost people in a Wilderness Search and Rescue (WiSAR) scenario is a task that can benefit from large numbers of agents, some of whom may be robotic. These agents may have differing levels of autonomy, determined by the set of tasks they are performing. In addition, the level of autonomy that results in the best performance may change due to varying workload or other factors. Allowing a supervisor and a searcher to jointly decide the correct level of autonomy for a given situation (“mixed initiative”) results in better overall performance than giving an agent absolute control over their level of autonomy (“adaptive autonomy”) or giving a supervisor absolute control over the agent's level of autonomy (“adjustable autonomy”).

This paper describes a modification to the mixed-initiative co-creative tool, the Evolutionary Dungeon Designer, and a study performed to evaluate and analyze the co-creative relationship between human designer and AI. The relationship between AI and human designer in creative processes is delicate, as adjusting the autonomy of the AI can negatively affect the user experience. To perfect this type of Human-AI collaboration, further research is needed. In this thesis, different degrees of initiative of the AI are explored, to gain further understanding of mixed-initiative co-creative tools. A user study was performed on the Evolutionary Dungeon Designer, with three varying degrees of AI-initiative. The study highlighted elements of frustration that the human designer experiences when using the tool, and the behaviour in the AI that led to possible strains on the relationship. The paper concludes with the identified issues and possible solutions, as well as suggested further research.

I denna uppsats presenteras ett verktyg för att assistera skapandet av spelbanor i dungeon-miljö. Skapandet av sådana banor är ofta resurskrävande under produktionen och har fokus på design. För att behålla designaspekten men underlätta arbetet är verktyget framtaget för att med tillgängliga parametrar producera grottgångar för sådana banor. Vi undersöker med användartester hur verktyget kan användas för att effektivisera en känd metod för att skapa spelbanor, men samtidigt skapa den kvalité som eftertraktas. Med detta avser vi inte bara att ta fram ett effektivt verktyg, utan även att demonstrera en metod för att använda procedurell generering av spelinnehåll för ett nytt ändamål inom speldesign. / In this article a tool for assisting the creation of game levels in a dungeon environment is presented. Creating such game levels often requires a large amount of resources during a game production and has focus on design. To keep the aspect of design but ease work, the tool is created to produce caverns for such game levels with accessible parameters. We examine with user tests how the tool can be used to make an existing method for creation of game levels more effective. But at the same time producing the quality that is coveted. With this we do not only hope to produce an effective tool, but also to demonstrate a method for using procedural generation for a new purpose in game design.

<p dir="ltr">Recognition of human behavior and object status plays a significant role in context-aware applications. Although researchers have explored methods to detect users’ activities, there still exists a research gap where end-users are not able to build personalized applications that accurately recognize their activities in the physical environment. Therefore, in this thesis, to fill the gap, we explore different in-situ context-aware Augmented Reality (AR) applications that leverage hand-object recognition and support users to rapidly author context-aware applications by referring to their activities. We first explore the possibility that end-users develop AR applications with customized freehand interactions and introduce GesturAR, an end-to-end authoring tool that supports users to create freehand AR applications through embodied demonstration and visual programming. Next, we explore hand interactions with physical objects and propose ARnnotate, an AR system enabling end-users to create custom data. To further study context-aware applications with haptic feedback, we present Ubi Edge, an AR authoring tool that allows end-users to customize edges on daily objects as tangible user interface (TUI) inputs to control varied digital functions. We develop an integrated AR device and a vision-based detection workflow to track 3D edges and detect the touch interaction between fingers and edges. Finally, to enable end-users to apply AR applications in different physical environments, we propose AdapTUI, a system that can automatically adapt the geometric-based TUIs in various environments. In summary, our contribution lies in the development of several context-aware AR applications that effectively sense users’ activities and facilitate users’ authoring process intuitively and conveniently.</p>