pith. sign in

arxiv: 2605.01400 · v1 · submitted 2026-05-02 · 💻 cs.HC · cs.AI· cs.CY· cs.IR

Investigating the Effects of Different Levels of User Control in an Interactive Educational Recommender System

Qurat Ul Ain , Mohamed Amine Chatti , William Kana Tsoplefack , Rawaa Alatrash , Shoeb Joarder This is my paper

Pith reviewed 2026-05-09 18:18 UTC · model grok-4.3

classification 💻 cs.HC cs.AIcs.CYcs.IR
keywords educational recommender systemsuser controlperceived controltransparencytrustsatisfactionperceived qualityuser study
0
0 comments X

The pith

Enabling users to build and refine their profiles in an educational recommender system is sufficient to promote positive perceptions of control, transparency, trust, satisfaction, and quality.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper tests the effects of giving learners different amounts of control over an educational recommender system. Users could adjust their own profiles as input, tweak the recommendation process, or edit the final output list. The results indicate that profile control by itself drives most of the gains in how users view the system, while extra controls over the algorithm and recommendations add only modest reinforcement. A reader would care because these systems aim to personalize learning paths in online courses, and knowing the minimal effective level of control could simplify interfaces without sacrificing benefits. If the pattern holds, builders can concentrate on clear profile tools rather than full algorithmic interfaces.

Core claim

The work shows that in an interactive educational recommender system, allowing users to build and refine their profiles is sufficient to generate positive perceptions of the recommendation goals, while providing additional control over the algorithm and recommendations mainly reinforces those impressions. Perceived control is the only goal significantly affected by the different levels of control, and input control produces the strongest effect on it. The levels of control influence transparency, trust, satisfaction, and perceived quality in distinct yet interconnected patterns, and user control in general positively shapes these perceptions to varying degrees.

What carries the argument

The staged control options over input (user profile construction), process (recommendation algorithm), and output (displayed recommendations).

If this is right

  • Input control alone promotes positive perceptions of the educational recommender system.
  • Additional process and output controls mainly reinforce existing positive impressions rather than creating substantial new effects.
  • Perceived control is the only recommendation goal significantly affected by control levels, with input control exerting the strongest influence.
  • Different control levels affect transparency, trust, satisfaction, and perceived quality in distinct but interconnected ways.
  • User control overall positively shapes transparency, trust, satisfaction, and perceived quality, though to varying extents.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Interface designers could prioritize simple profile-editing features over complex algorithm controls to achieve most perception benefits.
  • The observed perception gains might support higher completion rates in online courses if profile control reduces the time users spend searching for suitable materials.
  • Similar control patterns could be tested in non-educational recommender systems to check whether input control remains the dominant stage.
  • Longer deployments might reveal whether the reinforced impressions translate into sustained use or measurable learning improvements.

Load-bearing premise

Self-reported perceptions collected in a short-term between-subjects experiment reflect the real effects of control levels during extended everyday use across different educational platforms.

What would settle it

A follow-up study that finds no difference in actual engagement or satisfaction between users limited to profile control and users given full control over multiple weeks of real course activity would undermine the claim that input control alone is sufficient.

Figures

Figures reproduced from arXiv: 2605.01400 by Mohamed Amine Chatti, Qurat Ul Ain, Rawaa Alatrash, Shoeb Joarder, William Kana Tsoplefack.

Figure 1
Figure 1. Figure 1: A sample PKG for a learner. 3.1.1 PKG-based recommendations: In this recommendation approach, we use the constructed PKG of the learner. First, the SBERT embedding of the Wikipedia content of DNU concepts of the learner are generated to represent the learner model. Then the DNU concepts of the learner are used to query YouTube and Wkipedia APIs to get candidates for recommendations. After that, from the ti… view at source ↗
Figure 2
Figure 2. Figure 2: User Interface of the ERS in CourseMapper with three levels of user control: input (A), process (B), and output (C). view at source ↗
Figure 3
Figure 3. Figure 3: Prototypes for different levels of user control in the ERS. view at source ↗
Figure 4
Figure 4. Figure 4: User control with the input of the ERS. (a) User control with the process of the ERS (b) User control with the output of the ERS view at source ↗
Figure 5
Figure 5. Figure 5: User control with the process and output of the ERS. view at source ↗
Figure 6
Figure 6. Figure 6: Boxplots of all dependent variables (DVs) by condition ( view at source ↗
Figure 7
Figure 7. Figure 7: Pearson Correlation between different goals, with 95% confidence interval, where statistically significant correlations are view at source ↗
Figure 8
Figure 8. Figure 8: Structural path model with standardized coefficients ( view at source ↗
Figure 9
Figure 9. Figure 9: Condition-wise Correlation analysis between different goals, with 95% confidence interval, where statistically significant view at source ↗
read the original abstract

Educational recommender systems (ERSs) are becoming increasingly important in enhancing educational outcomes and personalizing learning experiences by providing recommendations of personalized resources and activities to learners, tailored to their individual learning needs. While user control is widely assumed to improve user experience, the effects of different levels of control in ERSs remain underexplored. To address this gap, we designed and evaluated an interactive ERS within the MOOC platform CourseMapper, where learners could interact with the input (i.e., user profile), process (i.e., recommendation algorithm), and output (i.e., recommendations) of the system. We conducted a between-subjects user study (N=184) to examine how varying levels of user control in an ERS influenced users' perceptions of the recommendation goals of perceived control, transparency, trust, satisfaction, and perceived quality. Our results show that enabling users to build and refine their profile is sufficient to promote positive perceptions of the ERS, while additional control options mainly reinforce these impressions. Moreover, perceived control is the only goal significantly affected by providing different levels of user control in the ERS, with input control exerting the strongest influence. Furthermore, different levels of control affect transparency, trust, satisfaction, and perceived quality in distinct yet interconnected ways. Overall, the findings provide empirical evidence that user control positively shapes transparency, trust, satisfaction, and perceived quality, though to varying extents.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper reports a between-subjects user study (N=184) on an interactive educational recommender system (ERS) embedded in the CourseMapper MOOC platform. Participants received varying levels of control over the input (user profile construction and refinement), process (recommendation algorithm), and output (recommendations). The central claims are that input control alone is sufficient to produce positive perceptions of the ERS, that additional process and output controls mainly reinforce those impressions, that perceived control is the only recommendation goal showing statistically significant differences across conditions (with input control exerting the strongest influence), and that the other goals (transparency, trust, satisfaction, perceived quality) are affected in distinct but interconnected ways.

Significance. If the results hold under stronger validation, the work supplies actionable design guidance for ERSs by indicating that profile-building controls can deliver most of the perceptual benefits without the overhead of full algorithmic or output control. This could simplify interfaces in educational settings while still supporting transparency, trust, and satisfaction. The study is strengthened by its use of a deployed platform and a reasonably powered sample; it adds empirical data to the HCI/recommender-systems literature on control granularity in learning contexts.

major comments (2)
  1. [User Study and Results] The headline claim that input control is 'sufficient to promote positive perceptions' (abstract and results) rests exclusively on immediate post-task self-reported Likert ratings collected after a single scripted interaction. No behavioral logs (profile-edit counts retained, recommendation acceptance rates, click-through data) or longitudinal follow-up are reported, leaving open whether the sufficiency finding generalizes beyond lab demand characteristics or novelty effects.
  2. [Results and Discussion] The assertion that 'perceived control is the only goal significantly affected' and that 'input control exert[s] the strongest influence' requires full statistical reporting (exact p-values, effect sizes, power analysis, and multiple-comparison corrections) to be load-bearing; the abstract supplies none, and the between-subjects design leaves prior MOOC experience unaccounted for as a potential confound on the transparency/trust/satisfaction scales.
minor comments (2)
  1. The abstract should include at least the key statistical outcomes (p-values, effect sizes) that support the significance claims.
  2. [Methodology] Clarify the exact operationalization of the three control levels (e.g., which UI elements were enabled/disabled in each condition) and how the between-subjects assignment was randomized.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed feedback on our manuscript. We address each major comment below, indicating where revisions will be made to strengthen the paper while maintaining the integrity of our study design and findings.

read point-by-point responses

  1. Referee: [User Study and Results] The headline claim that input control is 'sufficient to promote positive perceptions' (abstract and results) rests exclusively on immediate post-task self-reported Likert ratings collected after a single scripted interaction. No behavioral logs (profile-edit counts retained, recommendation acceptance rates, click-through data) or longitudinal follow-up are reported, leaving open whether the sufficiency finding generalizes beyond lab demand characteristics or novelty effects.

    Authors: We acknowledge that the study relies on immediate post-interaction self-reported Likert ratings without accompanying behavioral logs or longitudinal data. This design was chosen to isolate perceptual effects in a controlled between-subjects setup, consistent with common HCI evaluation practices for recommender systems. We agree that behavioral metrics and follow-up measures would provide additional robustness. In the revised manuscript, we will expand the limitations and future work sections to explicitly discuss the absence of behavioral data, potential novelty or demand effects, and the need for longitudinal validation. We will also moderate phrasing in the abstract and results to avoid overgeneralization while preserving the core contribution of the perceptual findings. revision: partial

  2. Referee: [Results and Discussion] The assertion that 'perceived control is the only goal significantly affected' and that 'input control exert[s] the strongest influence' requires full statistical reporting (exact p-values, effect sizes, power analysis, and multiple-comparison corrections) to be load-bearing; the abstract supplies none, and the between-subjects design leaves prior MOOC experience unaccounted for as a potential confound on the transparency/trust/satisfaction scales.

    Authors: We appreciate the call for complete statistical transparency. The revised manuscript will include exact p-values, effect sizes, a post-hoc power analysis, and details on multiple-comparison corrections. The abstract will be updated to summarize these statistics. On the potential confound of prior MOOC experience, we will report the distribution across conditions and, where relevant, include it as a covariate or discuss it explicitly as a limitation. Our randomization procedure aimed to balance such factors, but we will add this analysis to address the concern directly. revision: yes

Circularity Check

0 steps flagged

No circularity: purely empirical between-subjects study

full rationale

The paper presents results from a controlled user study (N=184) comparing three levels of user control in an ERS via post-task Likert-scale questionnaires. No equations, model fits, predictions, or derivations appear anywhere in the manuscript. Central claims (e.g., input control suffices for positive perceptions; only perceived control shows significant differences) are obtained directly from statistical tests on the collected survey responses. No self-citations are invoked as load-bearing uniqueness theorems or ansatzes; prior work is cited only for background on recommender systems and user control. The derivation chain is therefore self-contained and non-circular.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

This is an empirical HCI user study with no mathematical derivations, free parameters, or invented entities. It rests on standard domain assumptions of HCI methodology.

axioms (1)
  • domain assumption Self-reported Likert-scale measures validly capture users' perceptions of control, transparency, trust, satisfaction, and quality
    Invoked implicitly when interpreting survey results as evidence for the effects of control levels

pith-pipeline@v0.9.0 · 5581 in / 1159 out tokens · 85605 ms · 2026-05-09T18:18:36.389092+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

106 extracted references · 106 canonical work pages · 1 internal anchor

  1. [1]

    Solmaz Abdi, Hassan Khosravi, Shazia Sadiq, and Dragan Gasevic. 2020. Complementing educational recommender systems with open learner models. InProceedings of the tenth international conference on learning analytics & knowledge. 360–365

    work page 2020

  2. [2]

    Sunita B Aher and LMRJ Lobo. 2013. Combination of machine learning algorithms for recommendation of courses in E-Learning System based on historical data.Knowledge-Based Systems51 (2013), 1–14

    work page 2013

  3. [3]

    Muneeb Imtiaz Ahmad, Jasmin Bernotat, Katrin Lohan, and Friederike Eyssel. 2019. Trust and cognitive load during human-robot interaction. arXiv preprint arXiv:1909.05160(2019)

    work page arXiv 2019

  4. [4]

    Qurat Ul Ain, Mohamed Amine Chatti, Komlan Gluck Charles Bakar, Shoeb Joarder, and Rawaa Alatrash. 2023. Automatic Construction of Educational Knowledge Graphs: A Word Embedding-Based Approach.Information14, 10 (2023). https://doi.org/10.3390/info14100526

    work page doi:10.3390/info14100526 2023

  5. [5]

    Qurat Ul Ain, Mohamed Amine Chatti, Shoeb Joarder, Ilia Nassif, Benjamine Stella Wobiwo Teda, Mouadh Guesmi, and Rawaa Alatrash. 2023. Learning Channels to Support Interaction and Collaboration in CourseMapper. InProceedings of the 14th International Conference on Education Technology and Computers(Barcelona, Spain)(ICETC ’22)

    work page 2023

  6. [6]

    Qurat Ul Ain, Mohamed Amine Chatti, Paul Arthur Meteng Kamdem, Rawaa Alatrash, Shoeb Joarder, and Clara Siepmann. 2024. Learner Modeling and Recommendation of Learning Resources using Personal Knowledge Graphs. InProceedings of the 14th Learning Analytics and Knowledge Conference. 273–283

    work page 2024

  7. [7]

    Qurat Ul Ain, Mohamed Amine Chatti, William Kana Tsoplefack, Rawaa Alatrash, and Shoeb Joarder. 2025. Designing and Evaluating an Educational Recommender System with Different Levels of User Control.arXiv preprint arXiv:2501.12894(2025)

    work page arXiv 2025

  8. [8]

    Rawaa Alatrash, Mohamed Amine Chatti, Qurat Ul Ain, Yipeng Fang, Shoeb Joarder, and Clara Siepmann. 2024. ConceptGCN: Knowledge concept recommendation in MOOCs based on knowledge graph convolutional networks and SBERT.Computers and Education: Artificial Intelligence6 (2024), 100193

    work page 2024

  9. [9]

    Houssam El Aouifi, Mohamed El Hajji, Youssef Es-Saady, and Hassan Douzi. 2024. Video-Based Learning Recommender Systems: A Systematic Literature Review.IEEE Transactions on Learning Technologies17 (2024), 485–497. https://doi.org/10.1109/TLT.2023.3313391

    work page doi:10.1109/tlt.2023.3313391 2024

  10. [10]

    Krisztian Balog and Filip Radlinski. 2020. Measuring recommendation explanation quality: The conflicting goals of explanations. InProceedings of the 43rd international ACM SIGIR conference on research and development in information retrieval. 329–338. Manuscript submitted to ACM 38 Ain et al

    work page 2020

  11. [11]

    Jordan Barria-Pineda and Peter Brusilovsky. 2019. Explaining educational recommendations through a concept-level knowledge visualization. In Companion Proceedings of the 24th International Conference on Intelligent User Interfaces

    work page 2019

  12. [12]

    Peter M Bentler and Douglas G Bonett. 1980. Significance tests and goodness of fit in the analysis of covariance structures.Psychological bulletin 88, 3 (1980), 588

    work page 1980

  13. [13]

    Svetlin Bostandjiev, John O’Donovan, and Tobias Höllerer. 2012. TasteWeights: a visual interactive hybrid recommender system. InProceedings of the 6th ACM Conference on Recommender Systems. Association for Computing Machinery, New York, USA

    work page 2012

  14. [14]

    Svetlin Bostandjiev, John O’Donovan, and Tobias Höllerer. 2013. LinkedVis: Exploring social and semantic career recommendations.International Conference on Intelligent User Interfaces, Proceedings IUI, 107–116. https://doi.org/10.1145/2449396.2449412

    work page doi:10.1145/2449396.2449412 2013

  15. [15]

    Fatiha Bousbahi and Henda Chorfi. 2015. MOOC-Rec: A Case Based Recommender System for MOOCs.Procedia - Social and Behavioral Sciences 195 (2015), 1813–1822. World Conference on Technology, Innovation and Entrepreneurship

    work page 2015

  16. [16]

    Virginia Braun and Victoria Clarke. 2006. Using thematic analysis in psychology.Qualitative research in psychology3, 2 (2006), 77–101

    work page 2006

  17. [17]

    Simon Bruns, André Calero Valdez, Christoph Greven, Martina Ziefle, and Ulrik Schroeder. 2015. What should i read next? a personalized visual publication recommender system. InInternational Conference on Human Interface and the Management of Information. Springer, 89–100

    work page 2015

  18. [18]

    Maritza Bustos López, Giner Alor-Hernández, José Luis Sánchez-Cervantes, Mario Andrés Paredes-Valverde, and María del Pilar Salas-Zárate. 2020. EduRecomSys: an educational resource recommender system based on collaborative filtering and emotion detection.Interacting with Computers32, 4 (2020), 407–432

    work page 2020

  19. [19]

    1994.Structural equation modeling with EQS and EQS/Windows: Basic concepts, applications, and programming

    Barbara M Byrne. 1994.Structural equation modeling with EQS and EQS/Windows: Basic concepts, applications, and programming. Sage

    work page 1994

  20. [20]

    Mohamed Amine Chatti, Mouadh Guesmi, and Arham Muslim. 2024. Visualization for recommendation explainability: a survey and new perspectives.ACM Transactions on Interactive Intelligent Systems14, 3 (2024), 1–40

    work page 2024

  21. [21]

    Mohamed Amine Chatti, Mouadh Guesmi, Laura Vorgerd, Thao Ngo, Shoeb Joarder, Qurat Ul Ain, and Arham Muslim. 2022. Is more always better? The effects of personal characteristics and level of detail on the perception of explanations in a recommender system. InProceedings of the 30th ACM Conference on User Modeling, Adaptation and Personalization. 254–264

    work page 2022

  22. [22]

    Hung Chau, Jordan Barria-Pineda, and Peter Brusilovsky. 2018. Learning content recommender system for instructors of programming courses. In Artificial Intelligence in Education: 19th International Conference, AIED 2018, London, UK, June 27–30, 2018, Proceedings, Part II 19. Springer, 47–51

    work page 2018

  23. [23]

    Yu Chen and Pearl Pu. 2012. CoFeel: Using emotions for social interaction in group recommender systems. (2012)

    work page 2012

  24. [24]

    2013.Statistical power analysis for the behavioral sciences

    Jacob Cohen. 2013.Statistical power analysis for the behavioral sciences. routledge

    work page 2013

  25. [25]

    2013.Applied multiple regression/correlation analysis for the behavioral sciences

    Jacob Cohen, Patricia Cohen, Stephen G West, and Leona S Aiken. 2013.Applied multiple regression/correlation analysis for the behavioral sciences. Routledge

    work page 2013

  26. [26]

    Felipe Leite da Silva, Bruna Kin Slodkowski, Ketia Kellen Araújo da Silva, and Sílvio César Cazella. 2023. A systematic literature review on educational recommender systems for teaching and learning: research trends, limitations and opportunities.Education and Information Technologies 28, 3 (2023), 3289–3328

    work page 2023

  27. [27]

    Hendrik Drachsler, Katrien Verbert, Olga C Santos, and Nikos Manouselis. 2015. Panorama of recommender systems to support learning. In Recommender systems handbook. Springer, 421–451

    work page 2015

  28. [28]

    Ekstrand, Daniel Kluver, F

    Michael D. Ekstrand, Daniel Kluver, F. Maxwell Harper, and Joseph A. Konstan. 2015. Letting Users Choose Recommender Algorithms: An Experimental Study.Proceedings of the 9th ACM Conference on Recommender Systems(2015)

    work page 2015

  29. [29]

    Rosta Farzan and Peter Brusilovsky. 2011. Encouraging user participation in a course recommender system: An impact on user behavior.Computers in Human Behavior27, 1 (2011), 276–284

    work page 2011

  30. [30]

    Eleni Fotopoulou, Anastasios Zafeiropoulos, Michalis Feidakis, Dimitrios Metafas, and Symeon Papavassiliou. 2020. An interactive recommender system based on reinforcement learning for improving emotional competences in educational groups. InInternational Conference on Intelligent Tutoring Systems. Springer

    work page 2020

  31. [31]

    Daniel Gallego, Enrique Barra, Aldo Gordillo, and Gabriel Huecas. 2013. Enhanced recommendations for e-Learning authoring tools based on a proactive context-aware recommender. In2013 IEEE frontiers in education conference (FIE). IEEE, 1393–1395

    work page 2013

  32. [32]

    Enrique García, Cristóbal Romero, Sebastián Ventura, and Carlos de Castro. 2009. An architecture for making recommendations to courseware authors using association rule mining and collaborative filtering.User Modeling and User-Adapted Interaction19, 1 (2009), 99–132

    work page 2009

  33. [33]

    Fatih Gedikli, Dietmar Jannach, and Mouzhi Ge. 2014. How should I explain? A comparison of different explanation types for recommender systems.International Journal of Human-Computer Studies72, 4 (2014), 367–382

    work page 2014

  34. [34]

    David Graus, Maya Sappelli, and D Manh Chu. 2018. Let me tell you who you are. InExplaining recommender systems by opening black box user profiles. In Proceedings of the 2nd Fatrec Workshop on Responsible Recommendation, Vancouver, BC, Canada. 2–7

    work page 2018

  35. [35]

    Brynjar Gretarsson, John O’Donovan, Svetlin Bostandjiev, Christopher Hall, and Tobias Höllerer. 2010. SmallWorlds: Visualizing Social Recommen- dations.Computer Graphics Forum29, 3 (2010), 833–842

    work page 2010

  36. [36]

    Mouadh Guesmi, Mohamed Amine Chatti, Shoeb Joarder, Qurat Ul Ain, Rawaa Alatrash, Clara Siepmann, and Tannaz Vahidi. 2023. Interactive explanation with varying level of details in an explainable scientific literature recommender system.International Journal of Human–Computer Interaction(2023), 1–22

    work page 2023

  37. [37]

    Mouadh Guesmi, Mohamed Amine Chatti, Shoeb Joarder, Qurat Ul Ain, Clara Siepmann, Hoda Ghanbarzadeh, and Rawaa Alatrash. 2023. Justification vs. transparency: Why and how visual explanations in a scientific literature recommender system.Information14, 7 (2023), 401. Manuscript submitted to ACM Investigating the Effects of Different Levels of User Control ...

    work page 2023

  38. [38]

    Jaron Harambam, Dimitrios Bountouridis, Mykola Makhortykh, and Joris van Hoboken. 2019. Designing for the better by taking users into account: a qualitative evaluation of user control mechanisms in (news) recommender systems. InProceedings of 13th ACM Conference on Recommender Systems (RecSys ’19)

    work page 2019

  39. [39]

    Maxwell Harper, Funing Xu, Harmanpreet Kaur, Kyle Condiff, Shuo Chang, and Loren Terveen

    F. Maxwell Harper, Funing Xu, Harmanpreet Kaur, Kyle Condiff, Shuo Chang, and Loren Terveen. 2015. Putting Users in Control of their Recommendations. InProceedings of the 9th ACM Conference on Recommender Systems(Vienna, Austria)(RecSys ’15). Association for Computing Machinery, New York, NY, USA, 3–10

    work page 2015

  40. [40]

    Chen He, Denis Parra, and Katrien Verbert. 2016. Interactive recommender systems: A survey of the state of the art and future research challenges and opportunities.Expert Systems with Applications56 (2016), 9–27

    work page 2016

  41. [41]

    Marco Hellmann, Diana C Hernandez-Bocanegra, and Jürgen Ziegler. 2022. Development of an instrument for measuring users’ perception of transparency in recommender systems.system12 (2022), 7

    work page 2022

  42. [42]

    Dietmar Jannach, Michael Jugovac, and Ingrid Nunes. 2019. Explanations and user control in recommender systems. InProceedings of the 23rd International Workshop on Personalization and Recommendation on the Web and Beyond. 31–31

    work page 2019

  43. [43]

    Dietmar Jannach, Sidra Naveed, and Michael Jugovac. 2017. User control in recommender systems: Overview and interaction challenges. In E-Commerce and Web Technologies: 17th International Conference, EC-Web 2016, Porto, Portugal, September 5-8, 2016, Revised Selected Papers 17. Springer, 21–33

    work page 2017

  44. [44]

    Yucheng Jin, Bruno De Lemos Ribeiro Pinto Cardoso, and Katrien Verbert. 2017. How do different levels of user control affect cognitive load and acceptance of recommendations?. InIntRS@ RecSys. 35–42

    work page 2017

  45. [45]

    Yucheng Jin, Karsten Seipp, Erik Duval, and Katrien Verbert. 2016. Go with the flow: effects of transparency and user control on targeted advertising using flow charts. InProceedings of the international working conference on advanced visual interfaces. 68–75

    work page 2016

  46. [46]

    Yucheng Jin, Nava Tintarev, and Katrien Verbert. 2018. Effects of personal characteristics on music recommender systems with different levels of controllability. InProceedings of the 12th ACM Conference on Recommender Systems. 13–21

    work page 2018

  47. [47]

    Michael Jugovac and Dietmar Jannach. 2017. Interacting with recommenders—overview and research directions.ACM Transactions on Interactive Intelligent Systems (TiiS)7, 3 (2017), 1–46

    work page 2017

  48. [48]

    Seungwon Jung, Minjae Son, Chung-il Kim, Jehyeok Rew, and Eenjun Hwang. 2019. Video-based learning assistant scheme for sustainable education.New Review of Hypermedia and Multimedia25, 3 (2019), 161–181

    work page 2019

  49. [49]

    Antti Kangasrääsiö, Dorota Glowacka, and Samuel Kaski. 2015. Improving Controllability and Predictability of Interactive Recommendation Interfaces for Exploratory Search. InProceedings of the 20th International Conference on Intelligent User Interfaces (IUI ’15)

    work page 2015

  50. [50]

    Kenny, Burak Kaniskan, and D

    David A. Kenny, Burak Kaniskan, and D. Betsy McCoach. 2015. The Performance of RMSEA in Models with Small Degrees of Freedom.Sociological Methods & Research44, 3 (2015), 486–507. https://doi.org/10.1177/0049124114543236

    work page doi:10.1177/0049124114543236 2015

  51. [51]

    Shristi Shakya Khanal, PWC Prasad, Abeer Alsadoon, and Angelika Maag. 2020. A systematic review: machine learning based recommendation systems for e-learning.Education and Information Technologies(2020)

    work page 2020

  52. [52]

    René F Kizilcec. 2016. How much information? Effects of transparency on trust in an algorithmic interface. InProceedings of the 2016 CHI conference on human factors in computing systems. 2390–2395

    work page 2016

  53. [53]

    Rex B. Kline. 2016.Principles and Practice of Structural Equation Modeling(4 ed.). The Guilford Press, New York

    work page 2016

  54. [54]

    Bart P Knijnenburg, Svetlin Bostandjiev, John O’Donovan, and Alfred Kobsa. 2012. Inspectability and control in social recommenders. InProceedings of the sixth ACM conference on Recommender systems

    work page 2012

  55. [55]

    Bart P Knijnenburg, Nikhil Rao, and Alfred Kobsa. 2012. Experimental materials used in the study on inspectability and control in social recommender systems.Institute for Software Research, University of California, Irvine(2012)

    work page 2012

  56. [56]

    Bart P Knijnenburg, Niels JM Reijmer, and Martijn C Willemsen. 2011. Each to his own: how different users call for different interaction methods in recommender systems. InProceedings of the fifth ACM conference on Recommender systems. 141–148

    work page 2011

  57. [57]

    Bart P Knijnenburg, Martijn C Willemsen, Zeno Gantner, Handan Soncu, and Chris Newell. 2012. Explaining the user experience of recommender systems. InProceedings of the 10th ACM conference on Recommender systems. ACM, 287–290

    work page 2012

  58. [58]

    2007.Einführung in die computergestützte Analyse qualitativer Daten

    Udo Kuckartz. 2007.Einführung in die computergestützte Analyse qualitativer Daten. Springer

    work page 2007

  59. [59]

    Todd Kulesza, Margaret Burnett, Weng-Keen Wong, and Simone Stumpf. 2015. Principles of explanatory debugging to personalize interactive machine learning. InProceedings of the 20th international conference on intelligent user interfaces. 126–137

    work page 2015

  60. [60]

    Todd Kulesza, Simone Stumpf, Margaret Burnett, Sherry Yang, Irene Kwan, and Weng-Keen Wong. 2013. Too much, too little, or just right? Ways explanations impact end users’ mental models. In2013 IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC). IEEE, San Jose, CA, USA, 3–10. https://doi.org/10.1109/VLHCC.2013.6645235

    work page doi:10.1109/vlhcc.2013.6645235 2013

  61. [61]

    Johannes Kunkel, Tim Donkers, Lisa Michael, Catalin-Mihai Barbu, and Jürgen Ziegler. 2019. Let me explain: Impact of personal and impersonal explanations on trust in recommender systems. InProceedings of the 2019 CHI conference on human factors in computing systems. 1–12

    work page 2019

  62. [62]

    Dominik J Leiner. 2025. SoSci Survey (version 3.7.06)[computer software].München: SoSci Survey GmbH(2025)

    work page 2025

  63. [63]

    Q Vera Liao and S Shyam Sundar. 2022. Designing for responsible trust in AI systems: A communication perspective. InProceedings of the 2022 ACM conference on fairness, accountability, and transparency. 1257–1268

    work page 2022

  64. [64]

    Benedikt Loepp, Tim Hussein, and Jüergen Ziegler. 2014. Choice-based preference elicitation for collaborative filtering recommender systems. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 3085–3094. Manuscript submitted to ACM 40 Ain et al

    work page 2014

  65. [65]

    Nikos Manouselis, Hendrik Drachsler, Riina Vuorikari, Hans Hummel, and Rob Koper. 2011. Recommender systems in technology enhanced learning.Recommender systems handbook(2011), 387–415

    work page 2011

  66. [66]

    D Harrison McKnight, Vivek Choudhury, and Charles Kacmar. 2002. Developing and validating trust measures for e-commerce: An integrative typology.Information systems research13, 3 (2002), 334–359

    work page 2002

  67. [67]

    Harrison McKnight, Michelle Carter, and Paul Clay. 2009. Trust in technology: Development of a set of constructs and measures. (2009)

    work page 2009

  68. [68]

    Pavel Michlík and Mária Bieliková. 2010. Exercises recommending for limited time learning.Procedia Computer Science1, 2 (2010), 2821–2828

    work page 2010

  69. [69]

    Tim Miller. 2022. Are we measuring trust correctly in explainability, interpretability, and transparency research?arXiv preprint arXiv:2209.00651 (2022)

    work page arXiv 2022

  70. [70]

    Thao Ngo, Johannes Kunkel, and Jürgen Ziegler. 2020. Exploring mental models for transparent and controllable recommender systems: a qualitative study. InProceedings of the 28th ACM Conference on User Modeling, Adaptation and Personalization. 183–191

    work page 2020

  71. [71]

    Ingrid Nunes and Dietmar Jannach. 2017. A systematic review and taxonomy of explanations in decision support and recommender systems.User Modeling and User-Adapted Interaction27, 3 (2017), 393–444

    work page 2017

  72. [72]

    John O’Donovan, Barry Smyth, Brynjar Gretarsson, Svetlin Bostandjiev, and Tobias Höllerer. 2008. PeerChooser: visual interactive recommendation. InProceedings of the SIGCHI Conference on Human Factors in Computing Systems(Florence, Italy)(CHI ’08). Association for Computing Machinery, NY, USA, 1085–1088

    work page 2008

  73. [73]

    Jeroen Ooge, Leen Dereu, and Katrien Verbert. 2023. Steering Recommendations and Visualising Its Impact: Effects on Adolescents’ Trust in E-Learning Platforms. InProceedings of the 28th International Conference on Intelligent User Interfaces(Sydney, NSW, Australia)(IUI ’23). 156–170

    work page 2023

  74. [74]

    Denis Parra and Peter Brusilovsky. 2015. User-controllable personalization: A case study with SetFusion.International Journal of Human-Computer Studies78 (2015), 43–67

    work page 2015

  75. [75]

    Denis Parra, Peter Brusilovsky, and Christoph Trattner. 2014. See what you want to see: visual user-driven approach for hybrid recommendation. InProceedings of the 19th International Conference on Intelligent User Interfaces(Haifa, Israel)(IUI ’14). Association for Computing Machinery, New York, USA, 235–240

    work page 2014

  76. [76]

    Prolific Academic Ltd. 2025. Prolific. https://www.prolific.com

    work page 2025

  77. [77]

    Pearl Pu, Li Chen, and Rong Hu. 2011. A user-centric evaluation framework for recommender systems. InACM Conference on Recommender Systems

    work page 2011

  78. [78]

    Pearl Pu, Li Chen, and Rong Hu. 2011. User-involved preference elicitation for product search and recommender systems.AI magazine32, 3 (2011), 69–81

    work page 2011

  79. [79]

    Nils Reimers and Iryna Gurevych. 2019. Sentence-BERT: Sentence embeddings using Siamese BERT-networks.arXiv preprint arXiv:1908.10084 (2019)

    work page internal anchor Pith review arXiv 2019

  80. [80]

    Yuri Saito and Takayuki Itoh. 2011. MusiCube: a visual music recommendation system featuring interactive evolutionary computing. InProceedings of the 2011 Visual Information Communication - International Symposium(Hong Kong, China)(VINCI ’11). Association for Computing Machinery, New York, USA, Article 5, 6 pages

    work page 2011

Showing first 80 references.