Sven Schwarz's research while affiliated with Deutsches Forschungszentrum für Künstliche Intelligenz and other places

Publications (40)

Article
Knowledge workers face an ever increasing flood of informa- tion in their daily lives. To counter this and provide better support for information management and knowledge work in general, we have been investigating solutions inspired by human forgetting since 2013. These solutions are based on Semantic Desktop (SD) and Managed Forgetting(MF) techno...
Preprint
Full-text available
A growing number of applications users daily interact with have to operate in (near) real-time: chatbots, digital companions, knowledge work support systems -- just to name a few. To perform the services desired by the user, these systems have to analyze user activity logs or explicit user input extremely fast. In particular, text content (e.g. in...
Preprint
Full-text available
Knowledge workers face an ever increasing flood of information in their daily lives. To counter this and provide better support for information management and knowledge work in general, we have been investigating solutions inspired by human forgetting since 2013. These solutions are based on Semantic Desktop (SD) and Managed Forgetting (MF) technol...
Chapter
Existing Semantic Desktops are still reproached for being too complicated to use or not scaling well. Besides, a real "killer app" is still missing. In this paper, we present a new prototype inspired by NEPOMUK and its successors having a semantic graph and ontologies as its basis. In addition, we introduce the idea of context spaces that users can...
Preprint
Full-text available
Existing Semantic Desktops are still reproached for being too complicated to use or not scaling well. Besides, a real "killer app" is still missing. In this paper, we present a new prototype inspired by NEPOMUK and its successors having a semantic graph and ontologies as its basis. In addition, we introduce the idea of context spaces that users can...
Chapter
Chapter 1 detailed the need for preservation especially for our personal life. Even so, it is not part of most users’ regular practice. Preservation challenges users with a manual effort and a somewhat stern discipline to continuously perform the required steps. Being aware of these challenges, the chapter will present the novel approach realized i...
Conference Paper
With an increasing amount of available sensors lifelogging produces more and more data. Thus, realizing necessary condensation and forgetting processes becomes a challenge. In the last three years we investigated Managed Forgetting, Synergetic Preservation and Contextualized Remembering in the so-called ForgetIT project. Using the Semantic Desktop...
Conference Paper
With the growing amount of content stored in personal and organizational information spaces, finding and re-finding documents becomes both more crucial and challenging. In this work, we propose an approach to reduce information overload in navigation by automatically focusing on important documents, adaptively to the tasks at hand. Based on the ide...
Conference Paper
The World Wide Web has been rapidly developing in the last decade. In recent years, the Semantic Web has gained a lot of traction. It is a vision of the Web where data is understandable by machines as well as humans. Developments in the Semantic Web made way for the creation of massive knowledge bases containing a wealth of structured information....
Chapter
The paper presents recent developments in our research on Semantic Desktop for personal knowledge management supported by an ecosystem of applications and plug-ins using the knowledge worker’s Personal Information Model (PIMO) – a formal representation of his mental model for knowledge work – in everyday applications. We explain how the infrastruct...
Conference Paper
In this paper, we present an efficient comparison of text and RDF data for recognizing named entities. Here, a named entity is a text sequence that refers to a URI reference within an RDF graph. We present suffix arrays as representation format for text and a relational database scheme to represent SemanticWeb data. Using these representation facil...
Conference Paper
Although the Semantic Desktop paradigm has great potential, new users have to face the cold-start problem. Having to start with empty models is a barrier to any semantic technology and filling them with world-known concepts does not work for personal models. We propose to analyze the email database of a user and extract concepts of multiple types t...
Article
This paper presents the RDFHomepage project, a frame- work for using a person's structured data sources to auto-generate an HTML homepage. RDFHomepage uses RDF files as input, and currently supports several well-known RDF schemas, such as FOAF. In addition to these we have RDF converters for other structured file-formats, like Bibtex. RDFHomepage p...
Conference Paper
In knowledge work, users are confronted with difficulties in remembering, retrieving, and accessing relevant information and knowledge for their task at hand. In addition, knowledge-intensive, task-oriented work is highly fragmented and, therefore, requires knowledge workers to effectively handle and recover from interruptions. The Semantic Deskto...
Conference Paper
We discuss the idea of text responsive to reading and argue that the combination of eye tracking, text and real time interaction offers various possibilities to en- hance the reading experience. We present a number of prototypes and applications facilitating the user's gaze in order to assist comprehension difficulties and show their benefit in a p...
Article
We created a simple-to-use framework to construct gaze-responsive applications using web technology focussing on text. A plugin enables any compatible browser to interpret a new set of gaze handlers that behave similar to existing HTML and JavaScript mouse and keyboard event facilities. Keywords like onFixation, onGazeOver, and onRead can be attach...
Conference Paper
The paper presents an approach to support knowledge-intensive tasks with a context-sensitive task management system that is integrated into the user's personal knowledge space represented in the Nepomuk Semantic Desktop. The context-sensitive assistance is based on the combination of user observation, agile task modelling, automatic task prediction...
Article
This article describes the Semantic Desktop. We give insights into the core services that aim to improve personal knowledge management on the desktop. We describe these core components of our Semantic Desktop system and give evaluation results. Results of a long-term study reveal effects of using the Semantic Desktop on personal knowledge work.
Article
This article describes the Semantic Desktop. We give insights into the core services that aim to improve personal knowledge management on the desktop. We describe these core components of our Semantic Desktop system and give evaluation results. Results of a long-term study reveal effects of using the Semantic Desktop on personal knowledge work.
Conference Paper
Full-text available
The use of semantic representations in document-oriented environments – as formal annotations or embedded instances of a formal knowledge base – is seen as an enabling technology for intelligent services which may help knowledge workers in tasks like finding, structuring, or assessing information. Also, a high level of formalization has potential t...
Conference Paper
Full-text available
For document-centric work, meta-information in form of annotations has proven useful to enhance search and other retrieval tasks. The Mymory project uses a web-based workbench based on the semantic wiki Kaukolu that allows annotating texts both with concepts modeled in the user’s personal information model and other ontologies in a flexible way. An...
Conference Paper
Full-text available
For document-centric work, meta-information in form of an- notations has proven useful to enhance search and other retrieval tasks. Since creating annotations manually is a lot of work, it is desirable to also tap less obtrusive sources of meta-information such as the user's context (projects the user is working on, currently relevant topics, etc.)...
Conference Paper
Skipforward is a distributed recommendation system us-ing a lightweight ontology approach for formalizing opinions about item features. Items can be things such as songs or board games; describing and discriminating item features are the genre of a song or the degree of chance in a board game. Every user of the system is free to add new items and s...
Article
Full-text available
In this work we present the results of the EPOS project with regard to the needs of personalization in the Semantic Web. Focus of this work is the subjective view of an individual person, expressed in a Personal Information Model (PIMO). It is matched both with personal resources (files, e-mails, and websites) of the user and organizational knowled...
Conference Paper
The integration of heterogenous data sources is a crucial step for the upcoming semantic web - if existing information is not integrated, where will the data come from that the semantic web builds on? In this paper we present the gnowsis adapter framework, an implementation of an RDF graph system that can be used to integrate structured data source...
Article
In the research project EPOS1 we build a pro-active, context- sensitive support system to aid the user with his knowledge work, which is mostly about searching, reading, creating, and archiving of documents. In order to avoid distracting the user, the context gathering is realized by installable user observation plugins for standard applications su...
Conference Paper
In the research project EPOS we build a pro-active, context-sensitive support system to aid the user with his knowledge work, which is mostly about searching, reading, creating, and archiving of documents. In order to avoid distracting the user, the context gathering is realized by installable user observation plugins for standard applications such...
Article
Instant messaging(IM) has become more and more common these days, and is complementing e- mail and other means of electronic communication. However, due to its heavily context-dependent na- ture, searching archives of instant messages using only full text search is a tedious task. Also, in contrast to mails, files, and other electronic me- dia, ins...
Conference Paper
Full-text available
Knowledge-intensive activities can typically not be modeled sufficiently by classical, static process models and workflows. To enable a process-oriented knowledge management approach under these circumstances, the concept of weakly-structured workflows has been developed. This approach integrates process modeling and workflow enactment and facilita...
Conference Paper
Full-text available
Wissensmanagement zielt darauf ab, die Durchführung wissensintensiver Geschäftsprozesse im Unternehmen effizienter und effektiver zu machen. Während Software-Unterstützung für das WM momentan vorwiegend auf Kommunikations- und Kollaborations-Unterstützung bzw. verbessertes Informationsmanagement baut, war ein gezielter Geschäftsprozess- bzw. Workfl...
Article
Full-text available
Knowledge workers are typically processing multiple tasks in parallel. As humans can concentrate and work on but one task at a time, users have to switch between them. The required task switching causes a high cognitive load and decreases eciency. To reduce the costs of these task switches, we have to nd ways to cope with a multitude of dormant tas...

Citations

... information model (PIMO) (Maus et al., 2013;Sauermann et al., 2007), which reflects a user's mental model in a semantic network and is based on the so-called Semantic Desktop approach (Sauermann et al., 2005; most recent prototype in Jilek et al., 2018a). In the most recent implementation called cSpaces (short for Context Spaces Semantic Desktop; Jilek et al., 2018c), all things belonging to the same context in the user's mind-calendar events, files, e-mails, bookmarks, topics etc. (see Fig. 3 for an example)-are thus actually stored within a corresponding context in their PIMO. ...
... Notwithstanding the notable contribution of the mentioned research, especially for proposing the evaluation criteria for such RS and also providing an accessible dataset, many further steps are still to be taken toward bringing such RS to real-world applications. An essential step which is of great importance is moving from the simple characterization of context by Sappelli et al. [22] which is actually limited to eight predefined topics, to more complex models of context provided by Schwarz et al. [4,25] beyond which the innovative self-organizing contexts (briefly discussed in section 3.1 as a useful concept) has also been proposed by Jilek et al. [26]. ...
... The semantic desktop has proven to be a good solution for semantifying the work environment for KW [30], making the semantic dimension of CS practical. As defined by Sauermann et al. [28], a semantic desktop is an enlarged supplement to the user's memory in which an individual stores all her digital information like documents, multimedia, and messages interpreted as semantic web resources forming a KG using an individual vocabulary consisting of, e.g., topics, persons, events, and tasks embedded on the desktop (like provided by Maus et al. [31]). The performance of KG-based recommendations basically depends on the construction quality of such KG beneath the system which can be evaluated through innovative methods like proposed by Schröder et al. [32]. ...
... belonging to the same event or involving the same persons) would activate the "remembering" of the resource under consideration, making its memory buoyancy higher. This is the case represented by the dashed line of Figure 1 (Tran et al., 2016;Jilek et al., 2016). The memory buoyancy curves show activation and decay of concepts of a user's personal semantic network over a period of time. ...
... In further detail, a PIMO contains a semantic graph that consists of many thousand so-called interrelated "things" (e.g., files, tasks, events, persons, web pages, etc.)which connect the resources on the desktop with the representation in the graph as well as "virtual" concepts such as topicsthat shall represent a person's mental model. Each of these things is continuously assigned a specific value that serves as an indicator for its current relevancethe so-called "Memory Buoyancy", a means for a forgetful digital memory (Nieder ee et al., 2015;Tuan et al., 2016;Logie et al., 2018). Thus, the system is able to only show information that is, in a given moment, more relevant while at the same time inhibiting less relevant information Tempel et al., 2019). ...
... found in file names, mail subjects or in the (content) body of these items, were not extracted. On the one hand, we cannot spare these concepts, since SD systems may underperform without such additional vocabulary (for example experienced in [4]). On the other hand, accepting concepts that are not relevant for the user will clutter the created knowledge graph with irrelevant instances and non-helpful relations between items. ...
... SD's WebDAV 7 -based calendar PimoDAV is depicted in Figure 2: events entered into PimoDAV are automatically analyzed and added to the user's PIMO. Writing textual annotations (comments, notes, etc.) is possible using Seed [4], a semantic text editor (see Figure 3). While typing, Seed scans the text for entities found in the user's PIMO or other sources like DBpedia 8 . ...
... Other authors preferred a hybrid approach, where web surfing is performed through a combination of gaze and voice input [Hakkani-Tür et al. 2014;Sengupta et al. 2018]. Under another perspective, Biedert et al. [2010] developed a framework to create web applications that are natively gaze-responsive. ...
... This allows for context-aware applications such as a computer desktop allowing to switch between contexts. The interface is then adapted accordingly by providing the application windows which were active in that context, thus also hiding non-relevant ones (Schwarz, Kiesel, & van Elst, 2008). ...
... The Personal Information Model of a user (PIMO) [24] was proposed in the NEPOMUK project [57] and is now maintained by OSCAF members as a response to the need of having a formal representation of the users Mental Model 50. The EPOS project uses the PIMO to create the user's own domain ontology, called PIMO-User [60]. Personal concepts, ideas, projects, contacts and other resources that are of direct interest to the user represent the user's Personal Information Models (PIM). ...