Trügler Andreas, Scher Sebastian, Kopeinik Simone, Kowald Dominik
2023
The use of data-driven decision support by public agencies is becoming more widespread and already influences the allocation of public resources. This raises ethical concerns, as it has adversely affected minorities and historically discriminated groups. In this paper, we use an approach that combines statistics and data-driven approaches with dynamical modeling to assess long-term fairness effects of labor market interventions. Specifically, we develop and use a model to investigate the impact of decisions caused by a public employment authority that selectively supports job-seekers through targeted help. The selection of who receives what help is based on a data-driven intervention model that estimates an individual’s chances of finding a job in a timely manner and rests upon data that describes a population in which skills relevant to the labor market are unevenly distributed between two groups (e.g., males and females). The intervention model has incomplete access to the individual’s actual skills and can augment this with knowledge of the individual’s group affiliation, thus using a protected attribute to increase predictive accuracy. We assess this intervention model’s dynamics—especially fairness-related issues and trade-offs between different fairness goals- over time and compare it to an intervention model that does not use group affiliation as a predictive feature. We conclude that in order to quantify the trade-off correctly and to assess the long-term fairness effects of such a system in the real-world, careful modeling of the surrounding labor market is indispensable.
Reiter-Haas Markus, Kopeinik Simone, Lex Elisabeth
2021
In this paper, we study the moral framing of political content on Twitter. Specifically, we examine differences in moral framing in two datasets: (i) tweets from US-based politicians annotated with political affiliation and (ii) COVID-19 related tweets in German from followers of the leaders of the five major Austrian political parties. Our research is based on recent work that introduces an unsupervised approach to extract framing bias and intensity in news using a dictionary of moral virtues and vices. In this paper, we use a more extensive dictionary and adapt it to German-language tweets. Overall, in both datasets, we observe a moral framing that is congruent with the public perception of the political parties. In the US dataset, democrats have a tendency to frame tweets in terms of care, while loyalty is a characteristic frame for republicans. In the Austrian dataset, we find that the followers of the governing conservative party emphasize care, which is a key message and moral frame in the party’s COVID-19 campaign slogan. Our work complements existing studies on moral framing in social media. Also, our empirical findings provide novel insights into moral-based framing on COVID19 in Austria
Rekabsaz Navi, Kopeinik Simone, Schedl Markus
2021
Societal Biases in Retrieved Contents: Measurement Framework and Adversarial Mitigation of BERT Ranker
Lucija Krusic, Barbara Schuppler, Martin Hagmüller, Kopeinik Simone
2021
Due to recent advances in digitalisation and an emergence of new technologies, the STEM job market is further growing. This leads to higher salaries and lower unemployment rates. Despite these advantages, a pressing economic need for qualified STEM personal and many initiatives for increasing interest in STEM subjects, Austrian technical universities have consistently had issues with recruiting engineering students. Particularly women remain strongly underrepresented in STEM careers. Possible causes of this gender gap can be found in the effects of stereotype threat and the influence of role models, as stereotypical representations affect young people in various phases of their personal and professional development. As a part of the project proposal “Gender differences in career choices: Does the language matter?“, we investigated gender biases that potential students of Austrian STEM universities might face, and conducted two pilot studies: i) the analysis of EFL textbooks used in Austrian high schools, and ii) viewbooks used as promotional material for Austrian universities. EFL (English as a foreign language) textbooks are often used in teaching. We consider them as particularly relevant, since each of these books includes a dedicated section on careers. In the course of the first pilot study, we conducted a content analysis of eight textbooks for gender biases of personas in the context of careers and jobs. While results point to a nearly equal distribution of male and female personifications i.e., we found 9% more male characters, they were, however, not equally distributed among the different careers. Female personas were commonly associated with traditionally female careers (“stay-at-home mom”, “housewife”), which can be classified as indoor and domestic, while male personas tended to be associated with more prestigious, outdoor occupations (“doctor”, “police officer”). STEM occupations were predominantly (80%) associated with the male gender. Thus, the analysis of the Austrian EFL textbooks clearly points to the existence of gender stereotyping and gender bias as to the relationship of gender and career choice. In the second pilot study, we analyzed the symbolic portrayal of gender diversity in 52 Austrian university viewbooks, one for each bachelor programme at five Universities covering fields such as STEM, economy and law. As part of the analysis, we compared the representations of male and female students and professors with the actual student and faculty body. Results show a rather equal numeric gender representation in the non- technical university viewbooks but not in those of technical universities analysed. The comparison to real-life students’ gender distribution, revealed instances of underrepresentation of the male student body and overrepresentation of the female student body in technical university viewbooks (e.g., 15.4% underrepresentation of male students and 15.3% overrepresentation of female students in TUGraz viewbooks). We consider this a positive finding, as we believe that a diverse and gender neutral representation of people in educational and career information materials has the potential to entice a desired change in prospective students’ perception towards STEM subjects and engineering sciences.
Kopeinik Simone, Seitlinger Paul, Lex Elisabeth
2019
Kopeinik Simone, Lex Elisabeth, Kowald Dominik, Albert Dietrich, Seitlinger Paul
2019
When people engage in Social Networking Sites, they influence one another through their contributions. Prior research suggests that the interplay between individual differences and environmental variables, such as a person’s openness to conflicting information, can give rise to either public spheres or echo chambers. In this work, we aim to unravel critical processes of this interplay in the context of learning. In particular, we observe high school students’ information behavior (search and evaluation of Web resources) to better understand a potential coupling between confirmatory search and polarization and, in further consequence, improve learning analytics and information services for individual and collective search in learning scenarios. In an empirical study, we had 91 high school students performing an information search in a social bookmarking environment. Gathered log data was used to compute indices of confirmatory search and polarisation as well as to analyze the impact of social stimulation. We find confirmatory search and polarization to correlate positively and social stimulation to mitigate, i.e., reduce the two variables’ relationship. From these findings, we derive practical implications for future work that aims to refine our formalism to compute confirmatory search and polarisation indices and to apply it for depolarizing information services
Kowald Dominik, Kopeinik Simone , Lex Elisabeth
2017
Recommender systems have become important tools to supportusers in identifying relevant content in an overloaded informationspace. To ease the development of recommender systems, a numberof recommender frameworks have been proposed that serve a widerange of application domains. Our TagRec framework is one of thefew examples of an open-source framework tailored towards developingand evaluating tag-based recommender systems. In this paper,we present the current, updated state of TagRec, and we summarizeand reƒect on four use cases that have been implemented withTagRec: (i) tag recommendations, (ii) resource recommendations,(iii) recommendation evaluation, and (iv) hashtag recommendations.To date, TagRec served the development and/or evaluation processof tag-based recommender systems in two large scale Europeanresearch projects, which have been described in 17 research papers.‘us, we believe that this work is of interest for both researchersand practitioners of tag-based recommender systems.
Kopeinik Simone, Lex Elisabeth, Seitlinger Paul, Ley Tobias, Albert Dietrich
2017
In online social learning environments, tagging has demonstratedits potential to facilitate search, to improve recommendationsand to foster reflection and learning.Studieshave shown that shared understanding needs to be establishedin the group as a prerequisite for learning. We hypothesisethat this can be fostered through tag recommendationstrategies that contribute to semantic stabilization.In this study, we investigate the application of two tag recommendersthat are inspired by models of human memory:(i) the base-level learning equation BLL and (ii) Minerva.BLL models the frequency and recency of tag use while Minervais based on frequency of tag use and semantic context.We test the impact of both tag recommenders on semanticstabilization in an online study with 56 students completinga group-based inquiry learning project in school. Wefind that displaying tags from other group members contributessignificantly to semantic stabilization in the group,as compared to a strategy where tags from the students’individual vocabularies are used. Testing for the accuracyof the different recommenders revealed that algorithms usingfrequency counts such as BLL performed better whenindividual tags were recommended. When group tags wererecommended, the Minerva algorithm performed better. Weconclude that tag recommenders, exposing learners to eachother’s tag choices by simulating search processes on learners’semantic memory structures, show potential to supportsemantic stabilization and thus, inquiry-based learning ingroups.
Kowald Dominik, Lex Elisabeth, Kopeinik Simone
2016
In recent years, a number of recommendation algorithmshave been proposed to help learners find suitable learning resources online.Next to user-centered evaluations, offline-datasets have been usedto investigate new recommendation algorithms or variations of collaborativefiltering approaches. However, a more extensive study comparinga variety of recommendation strategies on multiple TEL datasets ismissing. In this work, we contribute with a data-driven study of recommendationstrategies in TEL to shed light on their suitability forTEL datasets. To that end, we evaluate six state-of-the-art recommendationalgorithms for tag and resource recommendations on six empiricaldatasets: a dataset from European Schoolnets TravelWell, a dataset fromthe MACE portal, which features access to meta-data-enriched learningresources from the field of architecture, two datasets from the socialbookmarking systems BibSonomy and CiteULike, a MOOC dataset fromthe KDD challenge 2015, and Aposdle, a small-scale workplace learningdataset. We highlight strengths and shortcomings of the discussed recommendationalgorithms and their applicability to the TEL datasets.Our results demonstrate that the performance of the algorithms stronglydepends on the properties and characteristics of the particular dataset.However, we also find a strong correlation between the average numberof users per resource and the algorithm performance. A tag recommenderevaluation experiment reveals that a hybrid combination of a cognitiveinspiredand a popularity-based approach consistently performs best onall TEL datasets we utilized in our study.
Kopeinik Simone, Kowald Dominik, Hasani-Mavriqi Ilire, Lex Elisabeth
2016
Classic resource recommenders like Collaborative Filteringtreat users as just another entity, thereby neglecting non-linear user-resource dynamics that shape attention and in-terpretation. SUSTAIN, as an unsupervised human cate-gory learning model, captures these dynamics. It aims tomimic a learner’s categorization behavior. In this paper, weuse three social bookmarking datasets gathered from Bib-Sonomy, CiteULike and Delicious to investigate SUSTAINas a user modeling approach to re-rank and enrich Collab-orative Filtering following a hybrid recommender strategy.Evaluations against baseline algorithms in terms of recom-mender accuracy and computational complexity reveal en-couraging results. Our approach substantially improves Col-laborative Filtering and, depending on the dataset, success-fully competes with a computationally much more expen-sive Matrix Factorization variant. In a further step, we ex-plore SUSTAIN’s dynamics in our specific learning task andshow that both memorization of a user’s history and clus-tering, contribute to the algorithm’s performance. Finally,we observe that the users’ attentional foci determined bySUSTAIN correlate with the users’ level of curiosity, iden-tified by the SPEAR algorithm. Overall, the results ofour study show that SUSTAIN can be used to efficientlymodel attention-interpretation dynamics of users and canhelp improve Collaborative Filtering for resource recommen-dations.
Seitlinger Paul, Kowald Dominik, Kopeinik Simone, Hasani-Mavriqi Ilire, Ley Tobias, Lex Elisabeth
2015
Classic resource recommenders like Collaborative Filtering(CF) treat users as being just another entity, neglecting non-linear user-resource dynamics shaping attention and inter-pretation. In this paper, we propose a novel hybrid rec-ommendation strategy that re nes CF by capturing thesedynamics. The evaluation results reveal that our approachsubstantially improves CF and, depending on the dataset,successfully competes with a computationally much moreexpensive Matrix Factorization variant.
Kowald Dominik, Seitlinger Paul, Kopeinik Simone, Ley Tobias, Trattner Christoph
2015
We assume that recommender systems are more successful,when they are based on a thorough understanding of how people processinformation. In the current paper we test this assumption in the contextof social tagging systems. Cognitive research on how people assign tagshas shown that they draw on two interconnected levels of knowledge intheir memory: on a conceptual level of semantic fields or LDA topics,and on a lexical level that turns patterns on the semantic level intowords. Another strand of tagging research reveals a strong impact oftime-dependent forgetting on users' tag choices, such that recently usedtags have a higher probability being reused than "older" tags. In thispaper, we align both strands by implementing a computational theory ofhuman memory that integrates the two-level conception and the processof forgetting in form of a tag recommender. Furthermore, we test theapproach in three large-scale social tagging datasets that are drawn fromBibSonomy, CiteULike and Flickr.As expected, our results reveal a selective effect of time: forgetting ismuch more pronounced on the lexical level of tags. Second, an extensiveevaluation based on this observation shows that a tag recommender interconnectingthe semantic and lexical level based on a theory of humancategorization and integrating time-dependent forgetting on the lexicallevel results in high accuracy predictions and outperforms other wellestablishedalgorithms, such as Collaborative Filtering, Pairwise InteractionTensor Factorization, FolkRank and two alternative time-dependentapproaches. We conclude that tag recommenders will benefit from goingbeyond the manifest level of word co-occurrences, and from includingforgetting processes on the lexical level.
Kowald Dominik, Kopeinik S., Seitlinger Paul, Trattner Christoph, Ley Tobias
2015
In this paper, we introduce a tag recommendation algorithmthat mimics the way humans draw on items in their long-term memory.Based on a theory of human memory, the approach estimates a tag'sprobability being applied by a particular user as a function of usagefrequency and recency of the tag in the user's past. This probability isfurther refined by considering the inuence of the current semantic contextof the user's tagging situation. Using three real-world folksonomiesgathered from bookmarks in BibSonomy, CiteULike and Flickr, we showhow refining frequency-based estimates by considering usage recency andcontextual inuence outperforms conventional "most popular tags" approachesand another existing and very effective but less theory-driven,time-dependent recommendation mechanism.By combining our approach with a simple resource-specific frequencyanalysis, our algorithm outperforms other well-established algorithms,such as FolkRank, Pairwise Interaction Tensor Factorization and CollaborativeFiltering. We conclude that our approach provides an accurateand computationally efficient model of a user's temporal tagging behavior.We demonstrate how effective principles of recommender systemscan be designed and implemented if human memory processes are takeninto account.