Visualization of the Relevance

Roßner, Daniel; Atzenbeck, Claus; Gross, Tom

doi:10.1145/3342220.3343659

Cited by 15 publications

(4 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Relevant related items are recommended to the users by presenting them as suggestion nodes (Roßner et al, 2019). Items are organised as nodes in a graph database, edges are weighted and undirected.…”

Section: Mother-combining Spatial Hypertext With Recommender Function...mentioning

confidence: 99%

Breaking the routine: spatial hypertext concepts for active decision making in recommender systems

Atzenbeck¹,

Herder

Roßner³

2023

New Review of Hypermedia and Multimedia

Self Cite

View full text Add to dashboard Cite

Recommender Systems are omnipresent in our digital life. Most notably, various media platforms guide us in selecting videos, but recommender systems are also used for more serious goals, such as news selection, political orientation and work decisions. As argued in this survey and position article, the paradigm of recommendation-based feeds has changed user behaviour from active decision making to rather passively following recommendations and accepting possibly suboptimal choices that are deemed "good enough". We provide a historic overview of media selection, discuss assumptions and goals of recommender systems and identify their shortcomings, based on existing literature. Then, the perspective changes to hypertext as a paradigm for structuring information and active decision making. To illustrate the relevance and importance of active decision making, we present a use case in the field of TV or media selection and (as a proof of concept) carried over to another application domain: maintenance in industry. In the discussion section, we focus on categorising these actions on a spectrum of "system-1" (fast and automated) tasks and "system-2" (critical thinking) tasks. Further, we argue how users can profit from tools that combine active (spatial) structuring and categorising with automatic recommendations, for professional tasks as well as private, leisure activities.

show abstract

Section: Mother-combining Spatial Hypertext With Recommender Function...mentioning

confidence: 99%

Breaking the routine: spatial hypertext concepts for active decision making in recommender systems

Atzenbeck¹,

Herder

Roßner³

2023

New Review of Hypermedia and Multimedia

Self Cite

View full text Add to dashboard Cite

show abstract

“…This is only used in the current client version for debugging purposes, as it allows the developers to "see" the system's parser results. (The visualisation used in DemoMedia is discussed in greater detail in Roßner et al, 2019. ) This interface closes the gap between distinct media for acquiring and consuming information and taking notes.…”

Section: Example 2: Project "Demomedia"mentioning

confidence: 99%

“…Now, man and machine make use of the same (visual) medium from which both create their own appropriate "internal" representation. The implementation is based on physics and a spring metaphor (Roßner et al, 2019), similar to how many graph visualisation tools work. However, they usually focus on the final, static result (e.g.…”

Section: Visualisation Of Suggestionsmentioning

confidence: 99%

Synthesising augmentation and automation

Atzenbeck¹,

Nürnberg²,

Roßner³

2021

New Review of Hypermedia and Multimedia

Self Cite

View full text Add to dashboard Cite

Historically, there has been a tendency to consider hypertext as a type of system, perhaps characterised by provision of links or other structure to users. In this article, we consider hypertext as a method of inquiry, a way of viewing arbitrary systems. In this view, what are traditionally called "navigational hypertext systems" might be considered as information retrieval systems. This opens the hypertext field to various other types of systems that traditionally would not be considered as part of the field. The change of view enables a deeper fusion of human and machine. In particular, today's AI-based, intelligent systems open the demand of synthesising automation (on the machine's side) and augmentation (on the user's side). This article is not about researching AI systems; it is about extending the view of hypertext systems to synthesise augmentation and automation. We specifically apply this view to intelligent systems, asking the question about how hypertext can act as a common medium between human and machine, particularly for knowledge intensive tasks. We propose spatial hypertext as a medium that enables users to create cognitive maps. Along these lines, we provide examples from multiple projects and examine how these applications can be productive.

show abstract

“…Since then, many other applications have been developed and enhanced according to the Spatial Hypertext principle, e.g. Wikis (Solís & Ali, 2008), visualization of relevant content (Roßner et al, 2019), use of a document store for data storage (Rubart, 2019) or interpretation of spatial ambiguity (Francisco-Revilla & Shipman, 2005).…”

Section: Related Workmentioning

confidence: 99%

Toward context-based text-to-3D scene generation

Henlein¹

View full text Add to dashboard Cite

People can describe spatial scenes with language and, vice versa, create images based on linguistic descriptions. However, current systems do not even come close to matching the complexity of humans when it comes to reconstructing a scene from a given text. Even the ever-advancing development of better and better Transformer-based models has not been able to achieve this so far. This task, the automatic generation of a 3D scene based on an input text, is called text-to-3D scene generation. The key challenge, and focus of this dissertation, now relate to the following topics: (a) Analyses of how well current language models understand spatial information, how static embeddings compare, and whether they can be improved by anaphora resolution. (b) Automated resource generation for context expansion and grounding that can help in the creation of realistic scenes. (c) Creation of a VR-based text-to-3D scene system that can be used as an annotation and active-learning environment, but can also be easily extended in a modular way with additional features to solve more contexts in the future. (d) Analyze existing practices and tools for digital and virtual teaching, learning, and collaboration, as well as the conditions and strategies in the context of VR. In the first part of this work, we could show that static word embeddings do not benefit significantly from pronoun substitution. We explain this result by the loss of contextual information, the reduction in the relative occurrence of rare words, and the absence of pronouns to be substituted. But we were able to we have shown that both static and contextualizing language models appear to encode object knowledge, but require a sophisticated apparatus to retrieve it. The models themselves in combination with the measures differ greatly in terms of the amount of knowledge they allow to extract. Classifier-based variants perform significantly better than the unsupervised methods from bias research, but this is also due to overfitting. The resources generated for this evaluation are later also an important component of point three. In the second part, we present AffordanceUPT, a modularization of UPT trained on the HICO-DET dataset, which we have extended with Gibsonien/telic annotations. We then show that AffordanceUPT can effectively make the Gibsonian/telic distinction and that the model learns other correlations in the data to make such distinctions (e.g., the presence of hands in the image) that have important implications for grounding images to language. The third part first presents a VR project to support spatial annotation respectively IsoSpace. The direct spatial visualization and the immediate interaction with the 3D objects should make the labeling more intuitive and thus easier. The project will later be incorporated as part of the Semantic Scene Builder (SeSB). The project itself in turn relies on the Text2SceneVR presented here for generating spatial hypertext, which in turn is based on the VAnnotatoR. Finally, we introduce Semantic Scene Builder (SeSB), a VR-based text-to-3D scene framework using Semantic Annotation Framework (SemAF) as a scheme for annotating semantic relations. It integrates a wide range of tools and resources by utilizing SemAF and UIMA as a unified data structure to generate 3D scenes from textual descriptions and also supports annotations. When evaluating SeSB against another state-of-the-art tool, it was found that our approach not only performed better, but also allowed us to model a wider variety of scenes. The final part reviews existing practices and tools for digital and virtual teaching, learning, and collaboration, as well as the conditions and strategies needed to make the most of technological opportunities in the future.

show abstract

Visualization of the Relevance

Cited by 15 publications

References 33 publications

Breaking the routine: spatial hypertext concepts for active decision making in recommender systems

Breaking the routine: spatial hypertext concepts for active decision making in recommender systems

Synthesising augmentation and automation

Toward context-based text-to-3D scene generation

Contact Info

Product

Resources

About