Publications
Explore the publications from our RecSys laboratory.
1.
Kolb, Thomas Elmar; Nalis-Neuner, Irina; Neidhardt, Julia
Like a Skilled DJ – an Expert Study on News Recommendations Beyond Accuracy Proceedings Article
In: Kille, Benjamin (Ed.): CEUR-WS.org, 2023.
@inproceedings{ 20.500.12708_191170,
title = {Like a Skilled DJ - an Expert Study on News Recommendations Beyond Accuracy},
author = {Thomas Elmar Kolb and Irina Nalis-Neuner and Julia Neidhardt},
editor = {Benjamin Kille},
doi = {10.34726/5332},
year = {2023},
date = {2023-01-01},
volume = {3561},
publisher = {CEUR-WS.org},
series = {CEUR Workshop Proceedings},
abstract = {In the past, recommender systems were primarily focused on optimizing accuracy. However, in recent years, there has been an increasing awareness that considerations beyond accuracy are necessary. The definition of what constitutes a good recommendation is a crucial issue. The most precise prediction may not always be the recommendation that satisfies the user best. This study offers a comprehensive investigation into the present advancements within the realm of beyond-accuracy measurements, especially the metrics diversity, serendipity, and novelty. Collaborative efforts between algorithmic models and domain experts can enrich recommendation quality, particularly in labeling and categorizing content. To address this, we present a study conducted by experts in the news domain. This study provides new insights into the multifaceted nature of this challenge. Employing an interdisciplinary approach, we underscore the significance of constructing a system that revolves around the user. Recent discussions about algorithmic content filtering and its societal implications underscore the importance of maintaining human involvement in the decision-making loop.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In the past, recommender systems were primarily focused on optimizing accuracy. However, in recent years, there has been an increasing awareness that considerations beyond accuracy are necessary. The definition of what constitutes a good recommendation is a crucial issue. The most precise prediction may not always be the recommendation that satisfies the user best. This study offers a comprehensive investigation into the present advancements within the realm of beyond-accuracy measurements, especially the metrics diversity, serendipity, and novelty. Collaborative efforts between algorithmic models and domain experts can enrich recommendation quality, particularly in labeling and categorizing content. To address this, we present a study conducted by experts in the news domain. This study provides new insights into the multifaceted nature of this challenge. Employing an interdisciplinary approach, we underscore the significance of constructing a system that revolves around the user. Recent discussions about algorithmic content filtering and its societal implications underscore the importance of maintaining human involvement in the decision-making loop.
2.
Kolb, Thomas Elmar; Wagne, Ahmadou; Sertkan, Mete; Neidhardt, Julia
Potentials of Combining Local Knowledge and LLMs for Recommender Systems Proceedings Article
In: Anelli, Vito Walter; Basile, Pierpaolo; Melo, Gerard De; Donini, Francesco; Ferrara, Antonio; Musto, Cataldo; Narducci, Fedelucio; Ragone, Azzurra; Zanker, Markus (Ed.): pp. 61–64, CEUR-WS.org, 2023.
@inproceedings{ 20.500.12708_191183,
title = {Potentials of Combining Local Knowledge and LLMs for Recommender Systems},
author = {Thomas Elmar Kolb and Ahmadou Wagne and Mete Sertkan and Julia Neidhardt},
editor = {Vito Walter Anelli and Pierpaolo Basile and Gerard De Melo and Francesco Donini and Antonio Ferrara and Cataldo Musto and Fedelucio Narducci and Azzurra Ragone and Markus Zanker},
doi = {10.34726/5334},
year = {2023},
date = {2023-01-01},
volume = {3560},
pages = {61–64},
publisher = {CEUR-WS.org},
series = {CEUR Workshop Proceedings},
abstract = {LLMs have revolutionized the understanding and generation of natural language, offering new possibilities for enhancing recommendation systems. In previous studies, LLMs exploit their global knowledge to provide zero- or few-shot recommendations. In this work, we aim to highlight the opportunities that LLMs pose to enrich the field of recommender systems combined with local knowledge. We propose to view recommender systems combined with LLMs from a broader perspective, recognizing them not merely as another method to replace existing recommendation approaches, but rather as a complementary and powerful approach to enhance and augment the overall recommendation process.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
LLMs have revolutionized the understanding and generation of natural language, offering new possibilities for enhancing recommendation systems. In previous studies, LLMs exploit their global knowledge to provide zero- or few-shot recommendations. In this work, we aim to highlight the opportunities that LLMs pose to enrich the field of recommender systems combined with local knowledge. We propose to view recommender systems combined with LLMs from a broader perspective, recognizing them not merely as another method to replace existing recommendation approaches, but rather as a complementary and powerful approach to enhance and augment the overall recommendation process.
3.
4.
Sertkan, Mete; Althammer, Sophia; Hofstätter, Sebastian; Knees, Peter; Neidhardt, Julia
Exploring Effect-Size-Based Meta-Analysis for Multi-Dataset Evaluation Proceedings Article
In: CEUR-WS.org, 2023.
@inproceedings{ 20.500.12708_191697,
title = {Exploring Effect-Size-Based Meta-Analysis for Multi-Dataset Evaluation},
author = {Mete Sertkan and Sophia Althammer and Sebastian Hofstätter and Peter Knees and Julia Neidhardt},
doi = {10.34726/5352},
year = {2023},
date = {2023-01-01},
volume = {3476},
publisher = {CEUR-WS.org},
series = {CEUR Workshop Proceedings},
abstract = {In this paper, we address the essential yet complex task of evaluating Recommender Systems (RecSys) across multiple datasets. This is critical for gauging their overall performance and applicability in various contexts. Owing to the unique characteristics of each dataset and the variability in algorithm performance, we propose the adoption of effect-size-based meta-analysis, a proven tool in comparative research. This approach enables us to compare a “treatment model” and a “control model” across multiple datasets, offering a comprehensive evaluation of their performance. Through two case studies, we highlight the flexibility and effectiveness of this method in multi-dataset evaluations, irrespective of the metric utilized. The power of forest plots in providing an intuitive and concise summarization of our analysis is also demonstrated, which significantly aids in the communication of research findings. Our work provides valuable insights into leveraging these methodologies to draw more reliable and validated conclusions on the generalizability and robustness of RecSys models.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper, we address the essential yet complex task of evaluating Recommender Systems (RecSys) across multiple datasets. This is critical for gauging their overall performance and applicability in various contexts. Owing to the unique characteristics of each dataset and the variability in algorithm performance, we propose the adoption of effect-size-based meta-analysis, a proven tool in comparative research. This approach enables us to compare a “treatment model” and a “control model” across multiple datasets, offering a comprehensive evaluation of their performance. Through two case studies, we highlight the flexibility and effectiveness of this method in multi-dataset evaluations, irrespective of the metric utilized. The power of forest plots in providing an intuitive and concise summarization of our analysis is also demonstrated, which significantly aids in the communication of research findings. Our work provides valuable insights into leveraging these methodologies to draw more reliable and validated conclusions on the generalizability and robustness of RecSys models.
5.
Basso, Linda; Nalis-Neuner, Irina; Neidhardt, Julia
News Diversity and Well-Being – An Experimental Exploration Of Diversity-Aware Recommender Systems Proceedings Article
In: 2023.
@inproceedings{ 20.500.12708_193212,
title = {News Diversity and Well-Being – An Experimental Exploration Of Diversity-Aware Recommender Systems},
author = {Linda Basso and Irina Nalis-Neuner and Julia Neidhardt},
year = {2023},
date = {2023-01-01},
abstract = {The demand for socially responsible designs for news recommender systems is currently of the utmost relevance. This paper presents a novel and interdisciplinary approach, bringing together psychol- ogists and computer scientists, to examine the impact of diverse news recommendations on individual users. In this experimental study, participants were divided into two groups, interacting with either a diverse news recommender system (experimental group) or a non-diversified system (control group). Subjective well-being and personal evaluations of the recommender system were measured. Although the study did not find a significant positive impact on participants’ subjective well-being after consuming more diverse news, this preliminary investigation opens avenues for further re- search. This study sets the stage for future investigations, providing valuable insights and highlighting the complexities of promoting diverse news consumption through recommender systems. Further research is warranted to explore potential enhancements and refine the understanding of the relationship between diversified news recommendations and user well-being. This contribution lays a groundstone for further research on responsibilities and how to implement basic human values, which are important to sustain and advance the democratic society we live in.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The demand for socially responsible designs for news recommender systems is currently of the utmost relevance. This paper presents a novel and interdisciplinary approach, bringing together psychol- ogists and computer scientists, to examine the impact of diverse news recommendations on individual users. In this experimental study, participants were divided into two groups, interacting with either a diverse news recommender system (experimental group) or a non-diversified system (control group). Subjective well-being and personal evaluations of the recommender system were measured. Although the study did not find a significant positive impact on participants’ subjective well-being after consuming more diverse news, this preliminary investigation opens avenues for further re- search. This study sets the stage for future investigations, providing valuable insights and highlighting the complexities of promoting diverse news consumption through recommender systems. Further research is warranted to explore potential enhancements and refine the understanding of the relationship between diversified news recommendations and user well-being. This contribution lays a groundstone for further research on responsibilities and how to implement basic human values, which are important to sustain and advance the democratic society we live in.
6.
Neidhardt, Julia; Wörndl, Wolfgang; Kuflik, Tsvi; Goldenberg, Dmitri; Zanker, Markus
Workshop on Recommenders in Tourism (RecTour) 2023 Proceedings Article
In: Zhang, Jie; Chen, Li; Berkovsky, Shlomo (Ed.): pp. 1274–1275, Association for Computing Machinery, New York, 2023.
@inproceedings{ 20.500.12708_191202,
title = {Workshop on Recommenders in Tourism (RecTour) 2023},
author = {Julia Neidhardt and Wolfgang Wörndl and Tsvi Kuflik and Dmitri Goldenberg and Markus Zanker},
editor = {Jie Zhang and Li Chen and Shlomo Berkovsky},
doi = {10.1145/3604915.3608764},
year = {2023},
date = {2023-01-01},
pages = {1274–1275},
publisher = {Association for Computing Machinery},
address = {New York},
abstract = {The Workshop on Recommenders in Tourism (RecTour) 2023, which is held in conjunction with the 17th issue of the ACM Conference on Recommender Systems (RecSys) in Singapore, addresses specific challenges for recommender systems in the tourism domain. In this overview paper, we summarize our motivations to organize the RecTour workshop and present the main topic areas of RecTour submissions. These include context-aware recommendations, group recommender systems, recommending composite items, decision making and user interaction issues, different information sources and various application scenarios.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The Workshop on Recommenders in Tourism (RecTour) 2023, which is held in conjunction with the 17th issue of the ACM Conference on Recommender Systems (RecSys) in Singapore, addresses specific challenges for recommender systems in the tourism domain. In this overview paper, we summarize our motivations to organize the RecTour workshop and present the main topic areas of RecTour submissions. These include context-aware recommendations, group recommender systems, recommending composite items, decision making and user interaction issues, different information sources and various application scenarios.
7.
Nalis, Irina; Neidhardt, Julia
Not Facial Expression, nor Fingerprint – Acknowledging Complexity and Context in Emotion Research for Human-Centered Personalization and Adaptation Proceedings Article
In: Adjunct Proceedings of the 31st ACM Conference on User Modeling, Adaptation and Personalization, pp. 325–330, Association for Computing Machinery, New York, 2023.
@inproceedings{ 20.500.12708_192197,
title = {Not Facial Expression, nor Fingerprint – Acknowledging Complexity and Context in Emotion Research for Human-Centered Personalization and Adaptation},
author = {Irina Nalis and Julia Neidhardt},
doi = {10.1145/3563359.3596990},
year = {2023},
date = {2023-01-01},
booktitle = {Adjunct Proceedings of the 31st ACM Conference on User Modeling, Adaptation and Personalization},
pages = {325–330},
publisher = {Association for Computing Machinery},
address = {New York},
abstract = {While research on emotion has emerged as a crucial area in studying this relationship, the use of classical psychological concepts in human emotion detection and sentiment analysis has been challenged by the cognitive sciences and psychology. This paper argues that the uncritical adoption of concepts that overlook the complexity and context of emotions may hinder progress in this field. To overcome this limitation, the theory of constructed emotion is reviewed, which suggests that emotions are not distinct categories but rather dimensions that require dynamic, rather than static, contextualized models. By prioritizing digital wellbeing in emotion studies and acknowledging complexity and context, future research can develop more effective models for emotion detection and sentiment analysis. The aim is to provide valuable insights for researchers seeking to advance our understanding of the relationship between technology and wellbeing for human centered-adaptation and personalization.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
While research on emotion has emerged as a crucial area in studying this relationship, the use of classical psychological concepts in human emotion detection and sentiment analysis has been challenged by the cognitive sciences and psychology. This paper argues that the uncritical adoption of concepts that overlook the complexity and context of emotions may hinder progress in this field. To overcome this limitation, the theory of constructed emotion is reviewed, which suggests that emotions are not distinct categories but rather dimensions that require dynamic, rather than static, contextualized models. By prioritizing digital wellbeing in emotion studies and acknowledging complexity and context, future research can develop more effective models for emotion detection and sentiment analysis. The aim is to provide valuable insights for researchers seeking to advance our understanding of the relationship between technology and wellbeing for human centered-adaptation and personalization.
8.
Knees, Peter; Neidhardt, Julia; Nalis-Neuner, Irina
Recommender Systems: Techniques, Effects, and Measures Toward Pluralism and Fairness Book Section
In: Werthner, Hannes; Ghezzi, Carlo; Kramer, Jeff (Ed.): pp. 417–434, Springer, Cham, 2023.
@incollection{ 20.500.12708_191188,
title = {Recommender Systems: Techniques, Effects, and Measures Toward Pluralism and Fairness},
author = {Peter Knees and Julia Neidhardt and Irina Nalis-Neuner},
editor = {Hannes Werthner and Carlo Ghezzi and Jeff Kramer},
doi = {10.1007/978-3-031-45304-5_27},
year = {2023},
date = {2023-01-01},
pages = {417–434},
publisher = {Springer},
address = {Cham},
abstract = {Recommender systems are widely used in various applications, such as online shopping, social media, and news personalization. They can help systems by delivering only the most relevant and promising information to their users and help people by mitigating information overload. At the same time, algorithmic recommender systems are a new form of gatekeeper that preselects and controls the information being presented and actively shapes users’ choices and behavior. This becomes a crucial aspect, as, if unaddressed and not safeguarded, these systems are susceptible to perpetuate and even amplify existing biases, including unwanted societal biases, leading to unfair and discriminatory outcomes. In this chapter, we briefly introduce recommender systems, their basic mechanisms, and their importance in various applications. We show how their outcomes and performance are assessed and discuss approaches to addressing pluralism and fairness in recommender systems. Finally, we highlight recently emerging directions within recommender systems research, pointing out opportunities for digital humanism to contribute interdisciplinary expertise.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Recommender systems are widely used in various applications, such as online shopping, social media, and news personalization. They can help systems by delivering only the most relevant and promising information to their users and help people by mitigating information overload. At the same time, algorithmic recommender systems are a new form of gatekeeper that preselects and controls the information being presented and actively shapes users’ choices and behavior. This becomes a crucial aspect, as, if unaddressed and not safeguarded, these systems are susceptible to perpetuate and even amplify existing biases, including unwanted societal biases, leading to unfair and discriminatory outcomes. In this chapter, we briefly introduce recommender systems, their basic mechanisms, and their importance in various applications. We show how their outcomes and performance are assessed and discuss approaches to addressing pluralism and fairness in recommender systems. Finally, we highlight recently emerging directions within recommender systems research, pointing out opportunities for digital humanism to contribute interdisciplinary expertise.
9.
Grossmann, Wilfried; Sertkan, Mete; Neidhardt, Julia; Werthner, Hannes
Pictures as a tool for matching tourist preferences with destinations Book Section
In: Augstein, Mirjam; Herder, Eelco; Wörndl, Wolfgang (Ed.): pp. 337–354, De Gruyter Oldenbourg, Berlin ; Boston, 2023.
@incollection{ 20.500.12708_191182,
title = {Pictures as a tool for matching tourist preferences with destinations},
author = {Wilfried Grossmann and Mete Sertkan and Julia Neidhardt and Hannes Werthner},
editor = {Mirjam Augstein and Eelco Herder and Wolfgang Wörndl},
doi = {10.1515/9783110988567-013},
year = {2023},
date = {2023-01-01},
pages = {337–354},
publisher = {De Gruyter Oldenbourg},
address = {Berlin ; Boston},
series = {De Gruyter Textbook},
abstract = {Usually descriptions of touristic products comprise information about accommodation, tourist attractions or leisure activities. Tourist decisions for a product are based on personal characteristics, planned vacation activities and specificities of potential touristic products. The decision should guarantee a high level of emotional and physical well-being, considering also some hard constraints like temporal and monetary resources, or travel distance. The starting point for the design of the described recommender system is a unified description of the preferences of the tourist and the opportunities offered by touristic products using the so-called seven-factor model. For the assignment of the values in the seven-factor model a predefined set of pictures is the pivotal instrument. These pictures represent various aspects of the personality and preferences of the tourist as well as general categories for the description of destinations, i. e., certain tourist attractions like landscape, cultural facilities, different leisure activities or emotional aspects associated with tourism. Based on the picture selection of a customer a so-called factor algorithm calculates values for each factor of the seven-factor model. This is a rather fast and intuitive method for acquisition of information about personality and preferences. The evaluation of the factors of the products is obtained by mapping descriptive attributes of touristic products onto the predefined pictures and afterwards applying the factor algorithm to the pictures characterizing the product. Based on this unified description of tourists and touristic products a recommendation can be defined by measuring the similarity between the user attributes and the product attributes. The approach is evaluated using data from a travel agency. Furthermore, other possible applications are discussed.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Usually descriptions of touristic products comprise information about accommodation, tourist attractions or leisure activities. Tourist decisions for a product are based on personal characteristics, planned vacation activities and specificities of potential touristic products. The decision should guarantee a high level of emotional and physical well-being, considering also some hard constraints like temporal and monetary resources, or travel distance. The starting point for the design of the described recommender system is a unified description of the preferences of the tourist and the opportunities offered by touristic products using the so-called seven-factor model. For the assignment of the values in the seven-factor model a predefined set of pictures is the pivotal instrument. These pictures represent various aspects of the personality and preferences of the tourist as well as general categories for the description of destinations, i. e., certain tourist attractions like landscape, cultural facilities, different leisure activities or emotional aspects associated with tourism. Based on the picture selection of a customer a so-called factor algorithm calculates values for each factor of the seven-factor model. This is a rather fast and intuitive method for acquisition of information about personality and preferences. The evaluation of the factors of the products is obtained by mapping descriptive attributes of touristic products onto the predefined pictures and afterwards applying the factor algorithm to the pictures characterizing the product. Based on this unified description of tourists and touristic products a recommendation can be defined by measuring the similarity between the user attributes and the product attributes. The approach is evaluated using data from a travel agency. Furthermore, other possible applications are discussed.
10.
Baumann, Andreas; Hofmann, Klaus; Marakasova, Anna; Neidhardt, Julia; Wissik, Tanja
Semantic micro-dynamics as a reflex of occurrence frequency: a semantic networks approach Journal Article
In: Cognitive Linguistics, vol. 34, iss. 3-4, pp. 533–568, 2023, ISSN: 0936-5907.
@article{ 20.500.12708_191531,
title = {Semantic micro-dynamics as a reflex of occurrence frequency: a semantic networks approach},
author = {Andreas Baumann and Klaus Hofmann and Anna Marakasova and Julia Neidhardt and Tanja Wissik},
url = {https://api.elsevier.com/content/abstract/scopus_id/85175416448},
doi = {10.1515/cog-2022-0008},
issn = {0936-5907},
year = {2023},
date = {2023-01-01},
journal = {Cognitive Linguistics},
volume = {34},
issue = {3-4},
pages = {533–568},
publisher = {De Gruyter},
abstract = {This article correlates fine-grained semantic variability and change with measures of occurrence frequency to investigate whether a word's degree of semantic change is sensitive to how often it is used. We show that this sensitivity can be detected within a short time span (i.e., 20 years), basing our analysis on a large corpus of German allowing for a high temporal resolution (i.e., per month). We measure semantic variability and change with the help of local semantic networks, combining elements of deep learning methodology and graph theory. Our micro-scale analysis complements previous macro-scale studies from the field of natural language processing, corroborating the finding that high token frequency has a negative effect on the degree of semantic change in a lexical item. We relate this relationship to the role of exemplars for establishing form-function pairings between words and their habitual usage contexts.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This article correlates fine-grained semantic variability and change with measures of occurrence frequency to investigate whether a word’s degree of semantic change is sensitive to how often it is used. We show that this sensitivity can be detected within a short time span (i.e., 20 years), basing our analysis on a large corpus of German allowing for a high temporal resolution (i.e., per month). We measure semantic variability and change with the help of local semantic networks, combining elements of deep learning methodology and graph theory. Our micro-scale analysis complements previous macro-scale studies from the field of natural language processing, corroborating the finding that high token frequency has a negative effect on the degree of semantic change in a lexical item. We relate this relationship to the role of exemplars for establishing form-function pairings between words and their habitual usage contexts.
11.
Pachinger, Pia; Hanbury, Allan; Neidhardt, Julia; Planitzer, Anna Maria
Toward Disambiguating the Definitions of Abusive, Offensive, Toxic, and Uncivil Comments Proceedings Article
In: pp. 107–113, 2023.
@inproceedings{ 20.500.12708_191532,
title = {Toward Disambiguating the Definitions of Abusive, Offensive, Toxic, and Uncivil Comments},
author = {Pia Pachinger and Allan Hanbury and Julia Neidhardt and Anna Maria Planitzer},
doi = {10.18653/v1/2023.c3nlp-1.11},
year = {2023},
date = {2023-01-01},
pages = {107–113},
abstract = {The definitions of abusive, offensive, toxic and uncivil comments used for annotating corpora for automated content moderation are highly intersected and researchers call for their disambiguation. We summarize the definitions of these terms as they appear in 23 papers across different fields. We compare examples given for uncivil, offensive, and toxic comments, attempting to foster more unified scientific resources. Additionally, we stress that the term incivility that frequently appears in social science literature has hardly been mentioned in the literature we analyzed that focuses on computational linguistics and natural language processing.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The definitions of abusive, offensive, toxic and uncivil comments used for annotating corpora for automated content moderation are highly intersected and researchers call for their disambiguation. We summarize the definitions of these terms as they appear in 23 papers across different fields. We compare examples given for uncivil, offensive, and toxic comments, attempting to foster more unified scientific resources. Additionally, we stress that the term incivility that frequently appears in social science literature has hardly been mentioned in the literature we analyzed that focuses on computational linguistics and natural language processing.
12.
Wagne, Ahmadou
COVID-19 and populism in Austrian news user comments – A machine learning approach Masters Thesis
Technische Universität Wien, Wien, 2023.
@mastersthesis{ 20.500.12708_176675,
title = {COVID-19 and populism in Austrian news user comments - A machine learning approach},
author = {Ahmadou Wagne},
doi = {10.34726/hss.2023.105940},
year = {2023},
date = {2023-01-01},
address = {Wien},
school = {Technische Universität Wien},
abstract = {The COVID-19 pandemic and the resulting government measures have triggered a wave of protests and demonstrations in Austria. We saw some protestors resorting to populist rhetoric to express their dissatisfaction, which in some cases, led to anti-democratic tendencies. Populist talking points have not been confined to the public sphere but rather emerged from social media and other online platforms, including news comments. Thus, there is a need to develop automated methods to detect populist statements in those texts. While previous research on populism has focused on politicians, recent studies have emphasized the role of citizens as populist actors. However, most of the current methods to detect populism in text rely on manual coding or dictionary-based approaches. Only a few scholars have attempted to employ machine learning for this task, resulting in a shortage of annotated data, particularly for the German language. To address this gap, this thesis adopts a minimalistic ideational definition of populism and performs various experiments using BERT-based transformer models to enhance the detection of populist user-generated content. Additionally, the thesis presents the first annotated dataset for populist news user comments in the German language by conducting an annotation study. The proposed model outperforms the state-of-the-art in this area and is applied in a case study analyzing the correlation between COVID-19 and populism in Austrian news user comments. A large-scale analysis of comments in the news forum of the Austrian daily newspaper Der Standard is conducted and the study reveals that the topic of COVID-19 in news articles attracted more populist comments than other topics during the pandemic. From these findings, implications can be drawn for future crisis management and communication.},
keywords = {},
pubstate = {published},
tppubtype = {mastersthesis}
}
The COVID-19 pandemic and the resulting government measures have triggered a wave of protests and demonstrations in Austria. We saw some protestors resorting to populist rhetoric to express their dissatisfaction, which in some cases, led to anti-democratic tendencies. Populist talking points have not been confined to the public sphere but rather emerged from social media and other online platforms, including news comments. Thus, there is a need to develop automated methods to detect populist statements in those texts. While previous research on populism has focused on politicians, recent studies have emphasized the role of citizens as populist actors. However, most of the current methods to detect populism in text rely on manual coding or dictionary-based approaches. Only a few scholars have attempted to employ machine learning for this task, resulting in a shortage of annotated data, particularly for the German language. To address this gap, this thesis adopts a minimalistic ideational definition of populism and performs various experiments using BERT-based transformer models to enhance the detection of populist user-generated content. Additionally, the thesis presents the first annotated dataset for populist news user comments in the German language by conducting an annotation study. The proposed model outperforms the state-of-the-art in this area and is applied in a case study analyzing the correlation between COVID-19 and populism in Austrian news user comments. A large-scale analysis of comments in the news forum of the Austrian daily newspaper Der Standard is conducted and the study reveals that the topic of COVID-19 in news articles attracted more populist comments than other topics during the pandemic. From these findings, implications can be drawn for future crisis management and communication.
13.
14.
Nalis-Neuner, Irina
Sentiment Analysis – psychological perspective Miscellaneous
2023.
@misc{ 20.500.12708_193816,
title = {Sentiment Analysis - psychological perspective},
author = {Irina Nalis-Neuner},
year = {2023},
date = {2023-01-01},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
15.
Sertkan, Mete; Althammer, Sophia; Hofstätter, Sebastian
Ranger: A Toolkit for Effect-Size Based Multi-Task Evaluation Proceedings Article
In: Danushka, Bollegala (Ed.): pp. 581–587, Association for Computational Linguistics, 2023.
@inproceedings{ 20.500.12708_192941,
title = {Ranger: A Toolkit for Effect-Size Based Multi-Task Evaluation},
author = {Mete Sertkan and Sophia Althammer and Sebastian Hofstätter},
editor = {Bollegala Danushka},
doi = {10.18653/v1/2023.acl-demo.56},
year = {2023},
date = {2023-01-01},
volume = {Volume 3: System Demonstrations},
pages = {581–587},
publisher = {Association for Computational Linguistics},
abstract = {In this paper, we introduce Ranger - a toolkit to facilitate the easy use of effect-size-based meta-analysis for multi-task evaluation in NLP and IR. We observed that our communities often face the challenge of aggregating results over incomparable metrics and scenarios, which makes conclusions and take-away messages less reliable. With Ranger, we aim to address this issue by providing a task-agnostic toolkit that combines the effect of a treatment on multiple tasks into one statistical evaluation, allowing for comparison of metrics and computation of an overall summary effect. Our toolkit produces publication-ready forest plots that enable clear communication of evaluation results over multiple tasks. Our goal with the ready-to-use Ranger toolkit is to promote robust, effect-size-based evaluation and improve evaluation standards in the community. We provide two case studies for common IR and NLP settings to highlight Ranger’s benefits.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper, we introduce Ranger – a toolkit to facilitate the easy use of effect-size-based meta-analysis for multi-task evaluation in NLP and IR. We observed that our communities often face the challenge of aggregating results over incomparable metrics and scenarios, which makes conclusions and take-away messages less reliable. With Ranger, we aim to address this issue by providing a task-agnostic toolkit that combines the effect of a treatment on multiple tasks into one statistical evaluation, allowing for comparison of metrics and computation of an overall summary effect. Our toolkit produces publication-ready forest plots that enable clear communication of evaluation results over multiple tasks. Our goal with the ready-to-use Ranger toolkit is to promote robust, effect-size-based evaluation and improve evaluation standards in the community. We provide two case studies for common IR and NLP settings to highlight Ranger’s benefits.
16.
Afzaal, Muhammad; Zia, Aayesha; Nouri, Jalal; Fors, Uno
Informative Feedback and Explainable AI-Based Recommendations to Support Students’ Self-regulation Journal Article
In: Technology, Knowledge and Learning, pp. 1–24, 2023.
@article{afzaal2023informative,
title = {Informative Feedback and Explainable AI-Based Recommendations to Support Students’ Self-regulation},
author = {Muhammad Afzaal and Aayesha Zia and Jalal Nouri and Uno Fors},
year = {2023},
date = {2023-01-01},
journal = {Technology, Knowledge and Learning},
pages = {1–24},
publisher = {Springer},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
17.
Afzaal, Muhammad; Nouri, Jalal; Aayesha, Aayesha
A Transformer-Based Approach for the Automatic Generation of Concept-Wise Exercises to Provide Personalized Learning Support to Students Proceedings Article
In: European Conference on Technology Enhanced Learning, pp. 16–31, Springer 2023.
@inproceedings{afzaal2023transformer,
title = {A Transformer-Based Approach for the Automatic Generation of Concept-Wise Exercises to Provide Personalized Learning Support to Students},
author = {Muhammad Afzaal and Jalal Nouri and Aayesha Aayesha},
year = {2023},
date = {2023-01-01},
booktitle = {European Conference on Technology Enhanced Learning},
pages = {16–31},
organization = {Springer},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
18.
Hussak, Melanie; Neidhardt, Julia; Stilz, Melanie
Bildung für den Frieden in einer digitalisierten Welt Miscellaneous
2022.
@misc{ 20.500.12708_153931,
title = {Bildung für den Frieden in einer digitalisierten Welt},
author = {Melanie Hussak and Julia Neidhardt and Melanie Stilz},
year = {2022},
date = {2022-01-01},
abstract = {In Artikel 4 der Erklärung der UN-Generalversammlung über eine Kultur des Friedens steht:
Bildung auf allen Ebenen ist eines der wichtigsten Instrumente zum Aufbau einer Kultur des Friedens. Dabei kommt der Menschenrechtserziehung eine besondere Bedeutung zu.
Frieden ist ein nie abgeschlossenes Projekt, das auf die stetige Abnahme von Gewalt und die gleichzeitige Zunahme von Gerechtigkeit zielt. Die Friedenspädagogik als inter- und transdisziplinäre Wissenschaft besitzt den weiten Blick, das Potential der Informatik für den “Ewigen Frieden” (Kant) in einer Paneldiskussion anschaulich und konkret darzustellen.},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
In Artikel 4 der Erklärung der UN-Generalversammlung über eine Kultur des Friedens steht:
Bildung auf allen Ebenen ist eines der wichtigsten Instrumente zum Aufbau einer Kultur des Friedens. Dabei kommt der Menschenrechtserziehung eine besondere Bedeutung zu.
Frieden ist ein nie abgeschlossenes Projekt, das auf die stetige Abnahme von Gewalt und die gleichzeitige Zunahme von Gerechtigkeit zielt. Die Friedenspädagogik als inter- und transdisziplinäre Wissenschaft besitzt den weiten Blick, das Potential der Informatik für den “Ewigen Frieden” (Kant) in einer Paneldiskussion anschaulich und konkret darzustellen.
Bildung auf allen Ebenen ist eines der wichtigsten Instrumente zum Aufbau einer Kultur des Friedens. Dabei kommt der Menschenrechtserziehung eine besondere Bedeutung zu.
Frieden ist ein nie abgeschlossenes Projekt, das auf die stetige Abnahme von Gewalt und die gleichzeitige Zunahme von Gerechtigkeit zielt. Die Friedenspädagogik als inter- und transdisziplinäre Wissenschaft besitzt den weiten Blick, das Potential der Informatik für den “Ewigen Frieden” (Kant) in einer Paneldiskussion anschaulich und konkret darzustellen.
19.
20.
Sertkan, Mete; Althammer, Sophia; Hofstätter, Sebastian; Neidhardt, Julia
Diversifying Sentiments in News Recommendation Proceedings Article
In: 2022.
@inproceedings{ 20.500.12708_175977,
title = {Diversifying Sentiments in News Recommendation},
author = {Mete Sertkan and Sophia Althammer and Sebastian Hofstätter and Julia Neidhardt},
doi = {10.34726/3903},
year = {2022},
date = {2022-01-01},
volume = {3228},
series = {CEUR Workshop Proceedings},
abstract = {Personalized news recommender systems are widely deployed to filter the information overload caused by the sheer amount of news produced daily. Recommended news articles usually have a sentiment similar to the sentiment orientation of the previously consumed news, creating a self-reinforcing cycle of sentiment chambers around people. Wu et al. introduced SentiRec – a sentiment diversity-aware neural news recommendation model to counter this lack of diversity.
In this work, we reproduce SentiRec without access to the original source code and data sample. We re-implement SentiRec from scratch and use the Microsoft MIND dataset (same source but different subset as in the original work) for our experiments. We evaluate and discuss our reproduction from different perspectives. While the original paper mainly has a user-centric perspective on sentiment diversity by comparing the recommendation list to the user’s interaction history, we also analyze the intra-list sentiment diversity of the recommendation list. Additionally, we study the effect of sentiment diversification on topical diversity. Our results suggest that SentiRec does not generalize well to other data since the compared baselines already perform well, opposing the original work’s findings. While the original SentiRec utilizes a rule-based sentiment analyzer, we also study a pre-trained neural sentiment analyzer. However, we observe no improvements in effectiveness nor in sentiment diversity. To foster reproducibility, we make our source code publicly available.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Personalized news recommender systems are widely deployed to filter the information overload caused by the sheer amount of news produced daily. Recommended news articles usually have a sentiment similar to the sentiment orientation of the previously consumed news, creating a self-reinforcing cycle of sentiment chambers around people. Wu et al. introduced SentiRec – a sentiment diversity-aware neural news recommendation model to counter this lack of diversity.
In this work, we reproduce SentiRec without access to the original source code and data sample. We re-implement SentiRec from scratch and use the Microsoft MIND dataset (same source but different subset as in the original work) for our experiments. We evaluate and discuss our reproduction from different perspectives. While the original paper mainly has a user-centric perspective on sentiment diversity by comparing the recommendation list to the user’s interaction history, we also analyze the intra-list sentiment diversity of the recommendation list. Additionally, we study the effect of sentiment diversification on topical diversity. Our results suggest that SentiRec does not generalize well to other data since the compared baselines already perform well, opposing the original work’s findings. While the original SentiRec utilizes a rule-based sentiment analyzer, we also study a pre-trained neural sentiment analyzer. However, we observe no improvements in effectiveness nor in sentiment diversity. To foster reproducibility, we make our source code publicly available.
In this work, we reproduce SentiRec without access to the original source code and data sample. We re-implement SentiRec from scratch and use the Microsoft MIND dataset (same source but different subset as in the original work) for our experiments. We evaluate and discuss our reproduction from different perspectives. While the original paper mainly has a user-centric perspective on sentiment diversity by comparing the recommendation list to the user’s interaction history, we also analyze the intra-list sentiment diversity of the recommendation list. Additionally, we study the effect of sentiment diversification on topical diversity. Our results suggest that SentiRec does not generalize well to other data since the compared baselines already perform well, opposing the original work’s findings. While the original SentiRec utilizes a rule-based sentiment analyzer, we also study a pre-trained neural sentiment analyzer. However, we observe no improvements in effectiveness nor in sentiment diversity. To foster reproducibility, we make our source code publicly available.