Müllner Peter , Lex Elisabeth, Schedl Markus, Kowald Dominik
2023
State-of-the-art recommender systems produce high-quality recommendations to support users in finding relevant content. However, through the utilization of users' data for generating recommendations, recommender systems threaten users' privacy. To alleviate this threat, often, differential privacy is used to protect users' data via adding random noise. This, however, leads to a substantial drop in recommendation quality. Therefore, several approaches aim to improve this trade-off between accuracy and user privacy. In this work, we first overview threats to user privacy in recommender systems, followed by a brief introduction to the differential privacy framework that can protect users' privacy. Subsequently, we review recommendation approaches that apply differential privacy, and we highlight research that improves the trade-off between recommendation quality and user privacy. Finally, we discuss open issues, e.g., considering the relation between privacy and fairness, and the users' different needs for privacy. With this review, we hope to provide other researchers an overview of the ways in which differential privacy has been applied to state-of-the-art collaborative filtering recommender systems.
Duricic Tomislav, Kowald Dominik, Emanuel Lacic, Lex Elisabeth
2023
By providing personalized suggestions to users, recommender systems have become essential to numerous online platforms. Collaborative filtering, particularly graph-based approaches using Graph Neural Networks (GNNs), have demonstrated great results in terms of recommendation accuracy. However, accuracy may not always be the most important criterion for evaluating recommender systems' performance, since beyond accuracy aspects such as recommendation diversity, serendipity, and fairness can strongly influence user engagement and satisfaction. This review paper focuses on addressing these dimensions in GNN-based recommender systems, going beyond the conventional accuracy-centric perspective. We begin by reviewing recent developments in approaches that improve not only the accuracy-diversity trade-off, but also promote serendipity and fairness in GNN-based recommender systems. We discuss different stages of model development including data preprocessing, graph construction, embedding initialization, propagation layers, embedding fusion, score computation, and training methodologies. Furthermore, we present a look into the practical difficulties encountered in assuring diversity, serendipity, and fairness, while retaining high accuracy. Finally, we discuss potential future research directions for developing more robust GNN-based recommender systems that go beyond the unidimensional perspective of focusing solely on accuracy. This review aims to provide researchers and practitioners with an in-depth understanding of the multifaceted issues that arise when designing GNN-based recommender systems.
Müllner Peter , Lex Elisabeth, Schedl Markus, Kowald Dominik
2023
User-based KNN recommender systems (UserKNN) utilize the rating data of a target user’s k nearest neighbors in the recommendation process. This, however, increases the privacy risk of the neighbors since their rating data might be exposed to other users or malicious parties. To reduce this risk, existing work applies differential privacy by adding randomness to the neighbors’ ratings, which reduces the accuracy of UserKNN. In this work, we introduce ReuseKNN, a novel differentially-private KNN-based recommender system. The main idea is to identify small but highly reusable neighborhoods so that (i) only a minimal set of users requires protection with differential privacy, and (ii) most users do not need to be protected with differential privacy, since they are only rarely exploited as neighbors. In our experiments on five diverse datasets, we make two key observations: Firstly, ReuseKNN requires significantly smaller neighborhoods, and thus, fewer neighbors need to be protected with differential privacy compared to traditional UserKNN. Secondly, despite the small neighborhoods, ReuseKNN outperforms UserKNN and a fully differentially private approach in terms of accuracy. Overall, ReuseKNN leads to significantly less privacy risk for users than in the case of UserKNN.
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.
Lex Elisabeth, Kowald Dominik, Seitlinger Paul, Tran Tran, Felfernig Alexander, Schedl Markus
2021
Psychology-informed Recommender Systems
Kowald Dominik, Müllner Peter , Zangerle Eva, Bauer Christine, Schedl Markus, Lex_KC Elisabeth
2021
Support the Underground: Characteristics of Beyond-Mainstream Music Listeners. EPJ Data Scienc
Schedl Markus, Bauer Christine, Reisinger Wolfgang, Kowald Dominik, Lex_KC Elisabeth
2021
Listener Modeling and Context-Aware Music Recommendation Based on Country Archetyp
Lacic Emanuel, Markus Reiter-Haas, Kowald Dominik, Reddy Dareddy Mano, Cho Junghoo, Lex Elisabeth
2020
In this work, we address the problem of providing job recommendations in an online session setting, in which we do not have full user histories. We propose a recom-mendation approach, which uses different autoencoder architectures to encode ses-sions from the job domain. The inferred latent session representations are then used in a k-nearest neighbor manner to recommend jobs within a session. We evaluate our approach on three datasets, (1) a proprietary dataset we gathered from the Austrian student job portal Studo Jobs, (2) a dataset released by XING after the RecSys 2017 Challenge and (3) anonymized job applications released by CareerBuilder in 2012. Our results show that autoencoders provide relevant job recommendations as well as maintain a high coverage and, at the same time, can outperform state-of-the-art session-based recommendation techniques in terms of system-based and session-based novelty
Lex Elisabeth, Kowald Dominik, Schedl Markus
2020
In this paper, we address the problem of modeling and predicting the music genre preferences of users. We introduce a novel user modeling approach, BLLu, which takes into account the popularity of music genres as well as temporal drifts of user listening behavior. To model these two factors, BLLu adopts a psychological model that describes how humans access information in their memory. We evaluate our approach on a standard dataset of Last.fm listening histories, which contains fine-grained music genre information. To investigate performance for different types of users, we assign each user a mainstreaminess value that corresponds to the distance between the user’s music genre preferences and the music genre preferences of the (Last.fm) mainstream. We adopt BLLu to model the listening habits and to predict the music genre preferences of three user groups: listeners of (i) niche, low-mainstream music, (ii) mainstream music, and (iii) medium-mainstream music that lies in-between. Our results show that BLLu provides the highest accuracy for predicting music genre preferences, compared to five baselines: (i) group-based modeling, (ii) user-based collaborative filtering, (iii) item-based collaborative filtering, (iv) frequency-based modeling, and (v) recency-based modeling. Besides, we achieve the most substantial accuracy improvements for the low-mainstream group. We believe that our findings provide valuable insights into the design of music recommender systems
Adolfo Ruiz Calleja, Dennerlein Sebastian, Kowald Dominik, Theiler Dieter, Lex Elisabeth, Tobias Ley
2019
In this paper, we propose the Social Semantic Server (SSS) as a service-based infrastructure for workplace andprofessional Learning Analytics (LA). The design and development of the SSS has evolved over 8 years, startingwith an analysis of workplace learning inspired by knowledge creation theories and its application in differentcontexts. The SSS collects data from workplace learning tools, integrates it into a common data model based ona semantically-enriched Artifact-Actor Network and offers it back for LA applications to exploit the data. Further,the SSS design promotes its flexibility in order to be adapted to different workplace learning situations. Thispaper contributes by systematizing the derivation of requirements for the SSS according to the knowledge creationtheories, and the support offered across a number of different learning tools and LA applications integrated to it.It also shows evidence for the usefulness of the SSS extracted from four authentic workplace learning situationsinvolving 57 participants. The evaluation results indicate that the SSS satisfactorily supports decision making indiverse workplace learning situations and allow us to reflect on the importance of the knowledge creation theoriesfor such analysis.
Kowald Dominik
2018
Social tagging systems enable users to collaboratively assign freely chosen keywords (i.e.,tags) to resources (e.g., Web links). In order to support users in nding descriptive tags, tagrecommendation algorithms have been proposed. One issue of current state-of-the-art tagrecommendation algorithms is that they are often designed in a purely data-driven way andthus, lack a thorough understanding of the cognitive processes that play a role when peopleassign tags to resources. A prominent example is the activation equation of the cognitivearchitecture ACT-R, which formalizes activation processes in human memory to determineif a specic memory unit (e.g., a word or tag) will be needed in a specic context. It is theaim of this thesis to investigate if a cognitive-inspired approach, which models activationprocesses in human memory, can improve tag recommendations.For this, the relation between activation processes in human memory and usage prac-tices of tags is studied, which reveals that (i) past usage frequency, (ii) recency, and (iii)semantic context cues are important factors when people reuse tags. Based on this, acognitive-inspired tag recommendation approach termed BLLAC+MPr is developed based onthe activation equation of ACT-R. An extensive evaluation using six real-world folksonomydatasets shows that BLLAC+MPr outperforms current state-of-the-art tag recommendationalgorithms with respect to various evaluation metrics. Finally, BLLAC+MPr is utilized forhashtag recommendations in Twitter to demonstrate its generalizability in related areas oftag-based recommender systems. The ndings of this thesis demonstrate that activationprocesses in human memory can be utilized to improve not only social tag recommendationsbut also hashtag recommendations. This opens up a number of possible research strands forfuture work, such as the design of cognitive-inspired resource recommender systems
Hasani-Mavriqi Ilire, Kowald Dominik, Helic Denis, Lex Elisabeth
2018
In this paper, we study the process of opinion dynamics and consensus building inonline collaboration systems, in which users interact with each other followingtheir common interests and their social proles. Specically, we are interested inhow users similarity and their social status in the community, as well as theinterplay of those two factors inuence the process of consensus dynamics. Forour study, we simulate the diusion of opinions in collaboration systems using thewell-known Naming Game model, which we extend by incorporating aninteraction mechanism based on user similarity and user social status. Weconduct our experiments on collaborative datasets extracted from the Web. Ourndings reveal that when users are guided by their similarity to other users, theprocess of consensus building in online collaboration systems is delayed. Asuitable increase of inuence of user social status on their actions can in turnfacilitate this process. In summary, our results suggest that achieving an optimalconsensus building process in collaboration systems requires an appropriatebalance between those two factors.
Seitlinger Paul, Ley Tobias, Kowald Dominik, Theiler Dieter, Hasani-Mavriqi Ilire, Dennerlein Sebastian, Lex Elisabeth, Albert D.
2017
Creative group work can be supported by collaborative search and annotation of Web resources. In this setting, it is important to help individuals both stay fluent in generating ideas of what to search next (i.e., maintain ideational fluency) and stay consistent in annotating resources (i.e., maintain organization). Based on a model of human memory, we hypothesize that sharing search results with other users, such as through bookmarks and social tags, prompts search processes in memory, which increase ideational fluency, but decrease the consistency of annotations, e.g., the reuse of tags for topically similar resources. To balance this tradeoff, we suggest the tag recommender SoMe, which is designed to simulate search of memory from user-specific tag-topic associations. An experimental field study (N = 18) in a workplace context finds evidence of the expected tradeoff and an advantage of SoMe over a conventional recommender in the collaborative setting. We conclude that sharing search results supports group creativity by increasing the ideational fluency, and that SoMe helps balancing the evidenced fluency-consistency tradeoff.
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.
Trattner Christoph, Kowald Dominik, Seitlinger Paul, Ley Tobias
2016
Several successful tag recommendation mechanisms have been developed, including algorithms built upon Collaborative Filtering, Tensor Factorization, graph-based and simple "most popular tags" approaches. From an economic perspective, the latter approach has been convincing since calculating frequencies is computationally efficient and effective with respect to different recommender evaluation metrics. In this paper, we introduce a tag recommendation algorithm that mimics the way humans draw on items in their long-term memory in order to extend these conventional "most popular tags" approaches. Based on a theory of human memory, the approach estimates a tag's reuse probability as a function of usage frequency and recency in the user's past (base-level activation) as well as of the current semantic context (associative component).Using four real-world folksonomies gathered from bookmarks in BibSonomy, CiteULike, Delicious and Flickr, we show how refining frequency-based estimates by considering recency and semantic context outperforms conventional "most popular tags" approaches and another existing and very effective but less theory-driven, time-dependent recommendation mechanism. By combining our approach with a simple resource-specific frequency analysis, our algorithm outperforms other well-established algorithms, such as Collaborative Filtering, FolkRank and Pairwise Interaction Tensor Factorization with respect to recommender accuracy and runtime. We conclude that our approach provides an accurate and computationally efficient model of a user's temporal tagging behavior. Moreover, we demonstrate how effective principles of recommender systems can be designed and implemented if human memory processes are taken into account.
Santos Patricia, Dennerlein Sebastian, Theiler Dieter, Cook John, Treasure-Jones Tamsin, Holley Debbie, Kerr Micky , Atwell Graham, Kowald Dominik, Lex Elisabeth
2016
Social learning networks enable the sharing, transfer and enhancement of knowledge in the workplace that builds the ground to exchange informal learning practices. In this work, three healthcare networks are studied in order to understand how to enable the building, maintaining and activation of new contacts at work and the exchange of knowledge between them. By paying close attention to the needs of the practitioners, we aimed to understand how personal and social learning could be supported by technological services exploiting social networks and the respective traces reflected in the semantics. This paper presents a case study reporting on the results of two co-design sessions and elicits requirements showing the importance of scaffolding strategies in personal and shared learning networks. Besides, the significance of these strategies to aggregate trust among peers when sharing resources and decision-support when exchanging questions and answers. The outcome is a set of design criteria to be used for further technical development for a social tool. We conclude with the lessons learned and future work.