Albin Zehe

Albin Zehe

Auto-generated profile, based on publicly available information

Albin Zehe

Center for Artificial Intelligence and Data Science (CAIDAS), Julius-Maximilians-Universität Würzburg

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 April 2020 in ACM Transactions on Social Computing
Verified
Journal Article
Open access
KKKonstantin Kobs
Konstantin Kobs
AZAlbin Zehe
Albin Zehe
ABArmin Bernstetter
Armin Bernstetter
JCJulian Chibane
Julian Chibane
JPJan Pfister
Jan Pfister
AHAndreas Hotho
Andreas Hotho
In recent years, streaming platforms for video games have seen increasingly large interest, as so-called esports have developed into a lucrative branch of business. Like for other sports, watching esports has become a new kind of entertainment medium, which is possible due to platforms that allow gamers to live stream their gameplay, the most popular platform being Twitch.tv. On these platforms, users can comment on streams in real time and thereby express their opinion about the events in the stream. Due to the popularity of Twitch.tv, this can be a valuable source of feedback for streamers aiming to improve their reception in a gaming-oriented audience. In this work, we explore the possibility of deriving feedback for video streams on Twitch.tv by analyzing the sentiment of live text comments made by stream viewers in highly active channels. Automatic sentiment analysis on these comments is a challenging task, as one can compare the language used in Twitch.tv with that used by an audience in a stadium, shouting as loud as possible in sometimes nonorganized ways. This language is very different from common English, mixing Internet slang and gaming-related language with abbreviations, intentional and unintentional grammatical and orthographic mistakes, and emoji-like images called emotes . Classic lexicon-based sentiment analysis techniques therefore fail when applied to Twitch comments. To overcome the challenge posed by the nonstandard language, we propose two unsupervised lexicon-based approaches that make heavy use of the information encoded in emotes, as well as a weakly supervised neural network–based classifier trained on the lexicon-based outputs, which is supposed to help generalization to unknown words by use of domain-specific word embeddings. To enable better understanding of Twitch.tv comments, we analyze a large dataset of comments, uncovering specific properties of their language, and provide a smaller set of comments labeled with sentiment information by crowdsourcing. We present two case studies showing the effectiveness of our methods in generating sentiment trajectories for events live streamed on Twitch.tv that correlate well with specific topics in the given stream. This allows for a new kind of implicit real-time feedback gathering for Twitch streamers and companies producing games or streaming content on Twitch. We make our datasets and code publicly available for further research. 1

16 April 2020 in Proceedings of the IEEE
Verified
Journal Article
Closed access
André BauerAndré Bauer
André Bauer
MZMarwin Zufle
Marwin Zufle
NHNikolas Herbst
Nikolas Herbst
AZAlbin Zehe
Albin Zehe
AHAndreas Hotho
Andreas Hotho
SKSamuel Kounev
Samuel Kounev
Modern distributed systems and Internet-of-Things applications are governed by fast living and changing requirements. Moreover, they have to struggle with huge amounts of data that they create or have to process. To improve the self-awareness of such systems and enable proactive and autonomous decisions, reliable time series forecasting methods are required. However, selecting a suitable forecasting method for a given scenario is a challenging task. According to the ``No-Free-Lunch Theorem,'' there is no general forecasting method that always performs best. Thus, manual feature engineering remains to be a mandatory expert task to avoid trial and error. Furthermore, determining the expected time-to-result of existing forecasting methods is a challenge. In this article, we extensively assess the state-of-the-art in time series forecasting. We compare existing methods and discuss the issues that have to be addressed to enable their use in a self-aware computing context. To address these issues, we present a step-by-step approach to fully automate the feature engineering and forecasting process. Then, following the principles from benchmarking, we establish a level-playing field for evaluating the accuracy and time-to-result of automated forecasting methods for a broad set of application scenarios. We provide results of a benchmarking competition to guide in selecting and appropriately using existing forecasting methods for a given self-aware computing context. Finally, we present a case study in the area of self-aware data-center resource management to exemplify the benefits of fully automated learning and reasoning processes on time series data.

4 April 2024 in Journal on Multimodal User Interfaces
Verified
Journal Article
Open access
PDF file
SSSophia C. Steinhaeusser
Sophia C. Steinhaeusser
AZAlbin Zehe
Albin Zehe
PSPeggy Schnetter
Peggy Schnetter
AHAndreas Hotho
Andreas Hotho
BLBirgit Lugrin
Birgit Lugrin
Storytelling is a long-established tradition and listening to stories is still a popular leisure activity. Caused by technization, storytelling media expands, e.g., to social robots acting as multi-modal storytellers, using different multimodal behaviours such as facial expressions or body postures. With the overarching goal to automate robotic storytelling, we have been annotating stories with emotion labels which the robot can use to automatically adapt its behavior. With it, three different approaches are compared in two studies in this paper: 1) manual labels by human annotators (MA), 2) software-based word-sensitive annotation using the Linguistic Inquiry and Word Count program (LIWC), and 3) a machine learning based approach (ML). In an online study showing videos of a storytelling robot, the annotations were validated, with LIWC and MA achieving the best, and ML the worst results. In a laboratory user study, the three versions of the story were compared regarding transportation and cognitive absorption, revealing no significant differences but a positive trend towards MA. On this empirical basis, the Automated Robotic Storyteller was implemented using manual annotations. Future iterations should include other robots and modalities, fewer emotion labels and their probabilities.

11 May 2026 in IEEE Transactions on Affective Computing
Verified
Journal Article
Closed access
SSSophia C. Steinhaeusser
Sophia C. Steinhaeusser
AZAlbin Zehe
Albin Zehe
SMSophia Maier
Sophia Maier
AHAndreas Hotho
Andreas Hotho
BLBirgit Lugrin
Birgit Lugrin
Storytelling has accompanied mankind since its beginnings and receiving stories is still a popular leisure activity. Today, new storytelling media are emerging. For example, social robots can act as storytellers, using different modalities such as facial expressions or body postures. With the goal of automating robotic storytelling, we annotated stories with emotion labels which a robot can use to automatically adapt its behavior to. Doing so, three different approaches and a baseline are compared in two studies in this paper: 1) manual labels by human annotators (MA), 2) software-based word-sensitive annotation using the Linguistic Inquiry and Word Count program (LIWC), 3) a LLM approach using ChatGPT (GPT), and 4) random annotations as baseline (RBL). In an online study showing videos of a storytelling robot, the annotations were validated using a backwards annotation task, with GPT closely followed by MA achieving the best, and the RBL the worst results. In a subsequent lab user study, all versions of the story were compared regarding storytelling experience, emotions, and robot perception, revealing no significant differences for any of these variables based on the annotation approach, but a significant decrease of negative affect and fatigue over the storytelling in all conditions. Taken together, the results indicate suitability of both the manual and GPT approaches. On this empirical basis, the Automatic Robotic Storyteller was implemented using GPT-based annotations and bodily expressions. The reduced programming effort of our system makes robotic storytelling accessible to a broader audience, strengthening its deployment in the field of affective computing.

17 March 2025
Verified
Book Chapter
Closed access
THThora Hagen
Thora Hagen
JOJanna Omeliyanenko
Janna Omeliyanenko
AEAnton Ehrmanntraut
Anton Ehrmanntraut
AHAndreas Hotho
Andreas Hotho
AZAlbin Zehe
Albin Zehe
FJFotis Jannidis
Fotis Jannidis
In this chapter, we explore the integration of knowledge graphs into large language models (LLMs) to enhance computational literary studies, a key sub-field of computational humanities. Computational humanities have evolved significantly over the past decades, driven by the increasing digitization of cultural heritage data, advancements in computing power, and the development of advanced analytical methods. This digitization progress has particularly enabled the creation of knowledge graphs that capture the semantic relationships embedded in humanities data, including texts. These graphs represent structured knowledge that can enrich LLMs, enabling them to generate semantically rich representations even in domains with limited computational resources. We explore how integrating knowledge graphs can enhance natural language processing (NLP) applications, specifically for the analysis of literary texts. To benefit the humanities and support the integration of KGs with LLMs for computational humanities, we discuss the specific content new KGs should ideally encompass. This necessitates a broader conceptualization of knowledge graphs, and supplies novel challenges for the field of knowledge graph creation, including for example diachronic concept alignment. Alongside this new perspective, we also propose the automatic creation of knowledge graphs from literary texts, such as graph-based plot representation, to allow for graph-based text analyses but also, again, create literary-informed LLMs through the integration of such graphs. By demonstrating these techniques through the lens of computational literary studies, we illustrate the significant impact that knowledge graphs can have on enriching LLMs and advancing the humanities.

1 January 2022
Verified
Conference Paper
Open access
PDF file
SHStefan Heil
Stefan Heil
KKKarina Kopp
Karina Kopp
AZAlbin Zehe
Albin Zehe
KKKonstantin Kobs
Konstantin Kobs
AHAndreas Hotho
Andreas Hotho
Stefan Heil, Karina Kopp, Albin Zehe, Konstantin Kobs, Andreas Hotho. Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022). 2022.