Analyzing user interactions with hypermedia systems

Adaptive hypermedia is an alternative to the traditional “one‐size‐fits‐all” approach in the development of hypermedia systems. Adaptive hypermedia (AH) systems build a model of the goals, preferences, and knowledge of each individual user, and use this model throughout the interaction with the user, in order to adapt to the needs of that user. Adaptive navigation support is a specific group of adaptive hypermedia techniques that become especially popular in educational hypermedia systems. This paper provides a brief overview of main adaptive navigation support techniques and analyzes the results of most representative empirical studies of these techniques. It demonstrates an evidence that different known techniques work most efficiently in different context. In particular, the studies summarized in the paper have provided evidence that users with different knowledge level of the subject may appreciate different adaptive navigation support technologies. The paper argues that more empirical studies are required to help the developers of adaptive hypermedia systems in selecting most relevant adaptation technologies. It also attempts to build a case for meta‐adaptive hypermedia systems, ie, systems that are able to adapt the very adaptation technology to the given user and context.

Users online tend to consume information adhering to their system of beliefs and ignore dissenting information. During the COVID-19 pandemic, users get exposed to a massive amount of information about a new topic having a high level of uncertainty. In this article, we analyze two social media that enforced opposite moderation methods, Twitter and Gab, to assess the interplay between news consumption and content regulation concerning COVID-19. We compare the two platforms on about three million pieces of content, analyzing user interaction with respect to news articles. We first describe users' consumption patterns on the two platforms focusing on the political leaning of news outlets. Finally, we characterize the echo chamber effect by modeling the dynamics of users' interaction networks. Our results show that the presence of moderation pursued by Twitter produces a significant reduction of questionable content, with a consequent affiliation toward reliable sources in terms of engagement and comments. Conversely, the lack of clear regulation on Gab results in the tendency of the user to engage with both types of content, showing a slight preference for the questionable ones which may account for a dissing/endorsement behavior. Twitter users show segregation toward reliable content with a uniform narrative. Gab, instead, offers a more heterogeneous structure where users, independent of their leaning, follow people who are slightly polarized toward questionable news.

A main feature of the next generation's computing software is supporting convenient user interactions. Interactive multimedia presentations that are executed in distributed environments raise some new computer communication behaviors, which we call “interactive multimedia communications”. In this paper, we (1) classify user interaction types, (2) point out the main problems in interactive multimedia communications, (3) distinguish two approaches for solving these main problems, (4) identify the possible control architectures for processing user interactions, and (5) propose some methods and control schemes accordingly.KeywordsVideo FrameUser InteractionFlow IndexControl ArchitectureMedia UnitThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Computer and multimedia technology plays an increasingly important role in education. However, current educational multimedia systems suffer from a major drawback that they do not support active engagement from users. In this paper we present a system that overcomes the hurdle. Our system offers rich user interactivity; it naturally supports and invites user involvement in a multimedia presentation. In the paper we first categorize the different levels of user interactivity that we deem vital for an interactive multimedia system, and then we describe a system design that enables active engagement.

Social Media platforms are increasing user engagements with their attractive User Interfaces (UI) design playing a crucial role in maintaining user involvement. Use of Emotional Intelligence (EI) into the UI transforming user interaction and engagement pattern significantly. With an emphasis on how design decisions affect user behaviour, this study explores the nuanced function of EIUIs (Emotionally Inteligence User Interfaces) in the context of social media. It specifically draws attention to the dangers of manipulative techniques like "dark patterns"-purposeful interface strategies intended to influence users to take actions they might not have otherwise taken-as well as the possibility of constructive involvement. By investigating these practices, the paper hopes to clarify the wider ethical consequences of interface design as well as the duty of platforms and designers to promote transparency, autonomy, and confidence in digital interactions. This paper aims to contribute to the discourse on user interface design by providing a comprehensive understanding of EIUIs' potential and pitfalls in the context of social media interactions. Additionally, this study focusses on finding information of the pattern of Emotionally Intelligent User Interfaces (EIUI) design to enhance user engagements emotionally and tend to take decisions and the impact of the design pattern in user experience.

This study critically examines the design of multimedia in education. The study begins by reviewing existing media in education and then uses a series of empirical studies to uncover, and then examine the key issues for the design of educational multimedia.<br><br>The research involves two studies. The first, a preliminary study, that identifies specific areas of interest for the research. This study looks at the existing use of an interactive video disc program for training in Price Waterhouse. The literature search, combined with the outcomes of the preliminary study, identified the areas for further research as: the use of video, forms of interaction, and the role of narrative. The main study examines these areas using two phases. The first phase analyses three treatments of the same educational text, on linear video, multimedia, and structured multimedia, each treatment maintaining the same content and narrative structure, but differing in presentation and control. The second phases builds on the results of the first with the creation and analysis of an interactive multimedia program that takes advantage of the identified strengths of multimedia, and specifically tackles problems found in the first phase. Qualitative data collection techniques are used in both phases, and form the basis of the findings.<br><br>The findings are presented as implications for multimedia design, and discuss the use and development of narrative and grammar in multimedia, as well as the importance of carefully designed user interaction and goal definition.

Building hypermedia for learning: a framework based on the design of user interface

Characterizing user navigation and interactions in online social networks

Intelligent user interfaces are becoming essential components of modern data-driven systems, where users interact with complex applications and large volumes of information. Traditional user interfaces rely on static layouts and predefined interaction patterns that often fail to adapt to individual user behavior and changing usage contexts. This study presents a reinforcement learning approach for developing intelligent user interfaces that dynamically adapt to user interaction patterns and optimize interface configurations over time. The proposed framework models interface adaptation as a sequential decision-making process in which a learning agent continuously observes user behavior and adjusts interface elements such as layout structure, component ordering, and navigation paths. The proposed approach is evaluated through simulated user interaction scenarios representing typical data-driven applications. Experimental observations indicate that reinforcement learning–based adaptation improves interface efficiency by reducing selection time and increasing accessibility of frequently used components. The results demonstrate that reinforcement learning can serve as an effective foundation for engineering intelligent user interfaces that continuously evolve with user behavior in modern data-driven environments.

The Nintendo Switch is a popular hybrid portable gaming console that employs a novel modular design, enabling it to be played in multiple controller configurations: Handheld mode, left and right JoyCon Sticks, and the JoyCon Grip. However, there are notes of user complaints about the complexity of performing basic tasks as a result of the unintuitiveness of its user interface. This can result to poor gaming experience and may potentially affect customer buying decisions. To assess the affordance of the user interface design of the Nintendo Switch, a total of 16 Filipino participants, classified as either novice or expert users, based on familiarity with the Switch or other gaming consoles and their functionalities, were made to perform a predetermined set of tasks on navigating through the Nintendo Switch’s basic menus, and playing a racing game across different controller orientations. Affordance, overall usability, and gaming experience satisfaction were measured through 5-point Likert scales (5 being the highest) for affordance surveys, system usability scale (SUS) surveys, and in-task and post-task surveys, respectively. Separate sets of surveys were done for general navigation and gaming tasks. Additionally, quantitative measures of effectiveness (task completion success rates) and efficiency (time to successfully complete tasks) were taken for general navigation tasks, and compared between novice and expert users. Task efficiency rates were also compared to baseline values obtained from experienced Nintendo Switch users. Results show a significant difference between expert and novice users’ effectiveness rates, with expert users all successfully performing 73% of navigation tasks, compared to novice rates of 18%. Efficiency rates for both expert and novice participant groups consistently fell significantly behind baseline task completion times. Affordance survey results showed consistent ratings across different controller configurations for expert users, and high preference for JoyCon Grip and Handheld mode configurations for novice participants. Of all SUS scores, only Game Navigation for expert users scored above the 68-point threshold. Recommendations to improve the Nintendo Switch user interface affordance and overall user satisfaction were made.

This paper presents the design and evaluation of a hypermedia system for blind users, making use of a non-visual interface, non-speech sounds, three input devices, and a 37-node hypermedia module. The important components of an effective auditory interface are discussed, together with the design of the auditory interface to hypermedia material. The evaluation is described, which was conducted over several weeks, and used a range of complementary objective and subjective measures to assess usability, performance and user preferences. The findings from evaluations with 9 visually impaired student participants are presented. The results from this research can be applied to the design and evaluation of other non-visual hypermedia systems, such as auditory WWW browsers and digital talking books.

In an interactive multimedia information system, users should have the flexibility to browse and choose various scenarios they want to see. This means that two-way communications should be captured by the conceptual model. Digital video has gained increasing popularity in many multimedia applications. Instead of sequential access to the video contents, the structuring and modeling of video data so that users can quickly and easily browse and retrieve interesting materials has become an important issue in designing multimedia information systems. An abstract semantic model called the augmented transition network (ATN), which can model video data and user interactions, is proposed in this paper. An ATN and its subnetworks can model video data based on different granularities, such as scenes, shots and key frames. Multimedia input strings are used as inputs for ATNs. The details of how to use multimedia input strings to model video data are also discussed. Key frame selection is based on the temporal and spatial relations of semantic objects in each shot. These relations are captured from our proposed unsupervised video segmentation method, which considers the problem of partitioning each frame as a joint estimation of the partition and class parameter variables. Unlike existing semantic models, which only model multimedia presentation, multimedia database searching or browsing, ATNs together with multimedia input strings can model these three in one framework.

With the diversity of digital platforms available nowadays, captur- ing, integrating, and analyzing user interactions across multiple platforms in a uniform way is critical for determining the over- all system's success. We present recommendations for a cross- platform system for logging interaction data. The ideal logging system should be: expressive, customizable and future-ready, struc- tured and consistent, developer-friendly, unobtrusive, and easy- to-interpret and convey to others. Based on these recommenda- tions we propose Service-Action-Objects, an approach that cap- tures interaction data on each platform as events in the form of a Service, a distinct feature such as a \blog," an Action, an action that a user has performed within the service such as \wrote an article," and Objects, an object or set of objects being acted on by the user such as the identi er of the blog article written. We discuss the strengths and limitations of this approach and present examples.

Nowadays, the possibilities offered by state-of-the-art deep neural networks allow the creation of systems capable of recognizing and indexing visual content with very high accuracy. Performance of these systems relies on the availability of high quality training sets, containing a large number of examples (e.g. million), in addition to the the machine learning tools themselves. For several applications, very good training sets can be obtained, for example, crawling (noisily) annotated images from the internet, or by analyzing user interaction (e.g.: on social networks). However, there are several applications for which high quality training sets are not easy to be obtained/created. Consider, as an example, a security scenario where one wants to automatically detect rarely occurring threatening events. In this respect, recently, researchers investigated the possibility of using a visual virtual environment, capable of artificially generating controllable and photo-realistic contents, to create training sets for applications with little available training images. We explored this idea to generate synthetic photo-realistic training sets to train classifiers to recognize the proper use of individual safety equipment (e.g.: worker protection helmets, high-visibility vests, ear protection devices) during risky human activities. Then, we performed domain adaptation to real images by using a very small image data set of real-world photographs. We show that training with the generated synthetic training set and using the domain adaptation step is an effective solution to address applications for which no training sets exist.

Push notifications play a critical role in the engagement strategies of e-commerce apps, but excessive or irrelevant notifications often lead to user frustration, app uninstalls, or opt-outs. Traditional notification strategies, which rely on time-based schedules or trigger-based systems, fail to account for the nuances of user behavior and intent. This project proposes an AI-driven solution to optimize the push notification system by minimizing the number of notifications sent while maximizing their relevance and timeliness. The approach leverages user behavior analysis, predictive modeling, and intent recognition through machine learning algorithms. By monitoring and analyzing user interactions, purchase patterns, and engagement metrics, the AI system identifies moments of genuine purchase intent. Notifications are then strategically sent only when they are most likely to align with the user's current needs and interests. KEY WORDS: Push notifications, e-commerce apps, notification fatigue, AI-driven solution, user behavior analysis, predictive modeling, machine learning algorithms, personalized notifications, engagement metrics, purchase intent prediction, notification management, real-time data analysis, intent recognition, behavioral analytics, notification optimization.