Zum Inhalt springen
Technische Universität München

Seminar in Summer Semester 2024:

Human-Centered Recommender Systems (IN2107, IN45014)

(Dr. Wolfgang Wörndl)

News

[19.06.2024] The presentation schedule has been added below

[16.04.2024] The seminar is fully booked, no more topics available. First meeting is Wednesday, 17.04.2024 at 16:00 in room MI 01.07.023

[22.03.2024] Please find the assignments of topics below. There is one topic available (Update: 9. Beyond Matrix Completion and Accuracy), registration by Email to woerndl@cit.tum.de (first-come first-serve)

[15.03.2024] I have added two topics due to high demand for seminar places, but the seminar is full now after the matching process

[22.01.2024] Web page online

Information

  • This seminar is for students in the Informatics Master program (module IN2107) or related study porgrams
    • The seminar will be conducted in English language
    • Prerequisites are a Bachelor's degree in computer science or related field
  • Students are expected to write a paper and give a presentation about the given topic area
    • Just summarizing related work is only the foundation of your paper, you need to show your own contribution beyond the summarization of other work. This contribution can be a new classification scheme, assessment and/or comparison of existing work, coming up with some novel conceptual ideas, design of an algorithm idea, mock-up of a new user interface, sketching new application areas and/or similar contributions
    • Your paper can be very focused, concentrating on a (small) subset of the given area. You can for example first give a brief overview of the topic area and then dig deeper on a selected aspect
      • The title of the paper should be adapted to the chosen focus, it does not have to be the original title of the topic
    • You need to search for suitable literature, the stated reference is intended as a starting point. Orient yourself to the given references or other research papers for the structure/contents of your own paper. You need to cite the literature your work is based on and clearly indicate when you are adopting or paraphrasing other work

    • To find additional literature, Google Scholar (scholar.google.com) is a valuable source. You can not only find literature based on keywords, but also find papers that have cited a certain paper. To do so, follow the "cited by ..." links of your most relevant articles

  • We expect all parts of the paper to be written by the student and not generative AI. We will check all submitted documents with automated tools for plagiarism and AI-generated content.

  • Detailed information for paper
    • 12-14 pages (including references) in English in the ACM format with a single column, not the old one with two columns
    • State your name, affiliation and email address (as the only author), include an abstract and selected CCS concepts and keywords
    • References and citations need to be in the correct format (but usage of a LaTex BIB file is optional)
    • Acknowledgments or appendix are optional and not expected
  • Detailed information for presentation
    • Duration is 25-30 minutes, plus questions&answers
    • The talk should be given freely, i.e. not completely read out from a script (in English)
    • You should present slides electronically in any format (e.g. Powerpoint or PDF)
    • You can use the TUM powerpoint template or your own format for the slides
  • All topics are advised by Wolfgang Wörndl, please send an Email to woerndl(at)in.tum.de for support or an appointment
  • Grading will be based on both the paper and the presentation (approximately equal weight)
  • Prerequisites for credits:
    • Submit the paper in acceptable quality until the stated deadline
    • Give a presentation of acceptable quality on the assigned date
    • Attend all presentation meetings and participate in the discussion

Procedure

  • Information meeting on Wed, 17.04.2024, 16:00 (room MI 01.07.023) with focus on the paper (but you are free to start earlier working on your paper after the matching is completed, and you have been assigned a place and topic!)
  • Intermediate meeting with short presentations of paper outlines and informal discussions on Wed, 22.05.2024, 16:30 (room tbd)
  • Submission of your paper in the correct format until Tue, 18.06.2024, 23:59 (no extensions!)
    • Submit the PDF and also the LaTeX source file(s) via Email to woerndl(at)in.tum.de (sending a link to a repository, dropbox or something similar is also possible)
  • Information meeting on Wed, 19.06.2024, 16:00 (room MI 01.07.023) with focus on the presentations
  • The presentations will likely be held on 08.07. (if needed), 09.07., 10.07. and 11.07.2024 each starting at 16:00 (room tbd)

Registration

Presentation Schedule

Tuesday 09.07.2024 (from 16:00, room 02.09.014)

  • Ahmet Erkan Turan: Examining Recommender Systems: Implicit Feedback and the Potential of External Factors
  • Ellen Saft: Visualization Challenges of Recommender Systems
  • Muhammad Haseeb Chaudhry: Personalized tourist recommender with help of Large Language Models (LLMs)
  • Evtim Kostadinov: Recommending and Presenting Sequences of Items

Wednesday 10.07.2024 (from 16:00, room 01.07.023)

  • Patrik Zander: Effective Use of Explanations in Recommender Systems
  • Başak Balci: Fairness and Multi-Stakeholder Recommender Systems
  • Janka Marschalková: Beyond Matrix Completion and Accuracy
  • René Jung: Privacy-Preserving Recommender Systems

Thursday 11.07.2024 (from 16:00, room 01.13.010)

  • Agata Andrysiak: Impact of Visual Layouts on User Experience and Decision Making in Human-Centered Recommender Systems: A Comparative Study of Grid and List Views
  • Quanyi Liu: Interactive Recommender Systems for the Internet of Things
  • Saltuk Bugra Karacan: Decentralized Recommender Systems: A Comparative Analysis with Traditional Recommender Systems

Topics and Literature

Introduction to recommender systems (optional):

* Ricci, F., Rokach, L., & Shapira, B. (2021). Recommender systems: Techniques, applications, and challenges. Recommender Systems Handbook, 1-35. (https://link.springer.com/chapter/10.1007/978-1-0716-2197-4_1)

1. Collecting Implicit and Explicit Feedback for Recommendations (Ahmet Erkan Turan)

Many recommendation techniques such as collaborative filtering need user ratings to recommend items. User ratings can be created by collecting explicit and explicit feedback. Explicit feedback, e.g., a user rating on a scale from 1 to 5 is very accurate but it demands an effort from the user. On the other hand, collecting implicit feedback is a bigger challenge but observing the user's browsing behavior or eye movements when interacting with a recommender system allows to identify recommendations that better satisfy the user's needs without annoying the user. The focus should be on how to better use implicit feedback for recommendation.

  • Rendle, S. (2022). Item recommendation from implicit feedback. In Recommender Systems Handbook (pp. 143-171). Springer, New York, NY.
  • Jawaheer, G., Weller, P., & Kostkova, P. (2014). Modeling user preferences in recommender systems: A classification framework for explicit and implicit user feedback. ACM Transactions on Interactive Intelligent Systems (TiiS), 4(2), 1-26.
  • Jannach, D., Lerche, L., & Zanker, M. (2018). Recommending based on implicit feedback. In Social Information Access (pp. 510-569). Springer, Cham.

2. Conversational and Critique-Based Recommender Systems (Tai Zhang)

Structured dialogues between the user and the recommender system promise better recommendations than traditional recommendations based on a one-time request/response. Recommender system providing dialogues to iteratively learn the user's preferences and improve the recommendations are called conversational. A conversational recommender system can either suggest two or more alternatives for concrete recommendations and let the user indicate her or his preferences for one item over the others or iteratively ask the user questions about the most important features of the expected recommendations.

  • Pramod, D., & Bafna, P. (2022). Conversational Recommender Systems Techniques, Tools, Acceptance, and Adoption: A State of the Art Review. Expert Systems with Applications, 117539.
  • Xie, H., Wang, D. D., Rao, Y., Wong, T. L., Raymond, L. Y., Chen, L., & Wang, F. L. (2018). Incorporating user experience into critiquing-based recommender systems: a collaborative approach based on compound critiquing. International Journal of Machine Learning and Cybernetics, 9(5), 837-852.
  • Jannach, D., Manzoor, A., Cai, W., & Chen, L. (2021). A survey on conversational recommender systems. ACM Computing Surveys (CSUR), 54(5), 1-36.

3. Visualization Challenges of Recommender Systems (Ellen Saft)

The traditional focus in recommender systems research has been on the algorithms to predict ratings and generate recommendations, but has shifted more towards the user experience in recent years. So is increasingly important how users can interact with recommender systems. One particular challenge is how to visualize item spaces to assist users in exploring more complex domains such as travel recommendation.

  • He, C., Parra, D., & Verbert, K. (2016). Interactive recommender systems: A survey of the state of the art and future research challenges and opportunities. Expert Systems with Applications, 56, 9-27.
  • Keck, M., & Kammer, D. (2018). Exploring Visualization Challenges for Interactive Recommender Systems. VisBIA@ AVI, 22-31.
  • Kunkel, J., Loepp, B., & Ziegler, J. (2017, March). A 3D item space visualization for presenting and manipulating user preferences in collaborative filtering. In Proceedings of the 22nd international conference on intelligent user interfaces (pp. 3-15).

4. Tourism Recommender Systems (Muhammad Haseeb Chaudhry)

Recommendation for travel and tourism pose increased challenges for systems because of the complex nature of the domain. Recommender systems need to combine heterogenous data sources such as flights, hotels and attractions, take user preferences into account, but also consider context such as the best times to travel in a region. In addition, people do relatively few trips per year, so it hard to compile enough data for destination recommender systems.

  • Massimo, D., & Ricci, F. (2022). Building effective recommender systems for tourists. AI Magazine.
  • Chaudhari, K., & Thakkar, A. (2020). A comprehensive survey on travel recommender systems. Archives of Computational Methods in Engineering, 27(5), 1545-1571.
  • Sarkar, J. L., Majumder, A., Panigrahi, C. R., Roy, S., & Pati, B. (2022). Tourism recommendation system: a survey and future research directions. Multimedia Tools and Applications, 1-45.

5. Recommending and Presenting Sequences of Items (Evtim Kostadinov)

Many recommender systems recommend single items such as movies or restaurants. Recommending a sequence of items, for example a music playlist or a tourist trip composed of multiple points of interest, is a more complicated issue. Not only the choice of items but also the sequence order influences the quality of the recommendation. For example, a strong ending with a very well-liked item at the end of a sequence might maximize the user satisfaction as the user tends to remember the end of recommendation most. When a recommendation is found, the system has to decide how to present the sequence to the user. The whole sequence could be recommended at a time but in certain scenarios, it might be appreciated if only the upcoming item is presented to the user at the right time.

  • Wörndl, W., Hefele, A., & Herzog, D. (2017). Recommending a sequence of interesting places for tourist trips. Information Technology & Tourism, 17(1), 31-54.
  • Lim, K. H., Chan, J., Karunasekera, S., & Leckie, C. (2019). Tour recommendation and trip planning using location-based social media: A survey. Knowledge and Information Systems, 60(3), 1247-1275.
  • Gavalas, D., Konstantopoulos, C., Mastakas, K., & Pantziou, G. (2014). A survey on algorithmic approaches for solving tourist trip design problems. Journal of Heuristics, 20(3), 291-328.

6. Recommendations for Groups (Bo Shao)

When recommending items to a group of users instead of a single user, the preferences of all group members have to be taken into account. Different preference aggregation strategies exist for this purpose. Basic approaches such as calculating an average preference of all users are easy to realize but might not be optimal. For example, if one user really dislikes a certain item, this item should not be recommended even if the majority of group members likes the item. It is important to say that there is no perfect way to aggregate the individual preferences. Instead, the group's intrinsic characteristics and the problem's nature have to be considered. It is also important how users can interact with these group recommender systems.

  • Masthoff, J., & Delić, A. (2022). Group Recommender Systems: Beyond Preference Aggregation. In Recommender Systems Handbook (pp. 381-420). Springer, New York, NY.
  • Jameson, A., & Smyth, B. (2007). Recommendation to groups. In The adaptive web (pp. 596-627). Springer, Berlin, Heidelberg.
  • Alvarado Rodriguez, O. L., Htun, N. N., Jin, Y., & Verbert, K. (2022). A systematic review of interaction design strategies for group recommendation systems. Proceedings of the ACM on Human-Computer Interaction.

7. Explanations and User Control in Recommender Systems (Patrik Zander)

Recommender systems excel in providing personalized items recommendations to users but it is not always clear why certain items were recommended. Often, the system and corresponding algorithms are black boxes for users. Explanations and justifications can help making the recommendation more understandable for users. Furthermore, it is important to provide means for users to control and influence the recommendation process. This is not only important to facilitate more accurate recommendations but also to increase the trust of users in the system.

  • Jannach, D., Jugovac, M. & Nunes, I. (2023). Explanations and user control in recommender systems. In M. Augstein, E. Herder & W. Wörndl (Ed.), Personalized Human-Computer Interaction (pp. 129-152). Berlin, Boston: De Gruyter Oldenbourg. https://doi.org/10.1515/9783110988567-006.
  • Tintarev, N., & Masthoff, J. (2022). Beyond explaining single item recommendations. In Recommender Systems Handbook (pp. 711-756). Springer, New York, NY.
  • Kouki, P., Schaffer, J., Pujara, J., O’Donovan, J., & Getoor, L. (2020). Generating and understanding personalized explanations in hybrid recommender systems. ACM Transactions on Interactive Intelligent Systems (TiiS), 10(4), 1-40

8. Fairness and Multi-Stakeholder Recommender Systems (Başak Balci)

Often, recommender systems are optimized for recommendations to best suit the users' interests and preferences only. However, other stakeholders are involved in real work recommender applications. For example, in e-Commerce, sellers may want to maximize their profit, and in tourism recommendation, suggesting the most popular places may lead to overtourism. In addition, objectives such as coverage, diversity and fairness play an important role when generating recommendations. Therefore, recommender systems need to balance possible conflicting interests of different parties and take ethical considerations into account.

  • Abdollahpouri, H., & Burke, R. (2022). Multistakeholder recommender systems. In Recommender systems handbook(pp. 647-677). Springer, New York, NY.
  • Ekstrand, M. D., Das, A., Burke, R., & Diaz, F. (2022). Fairness in recommender systems. In Recommender systems handbook (pp. 679-707). Springer, New York, NY.
  • Wang, Y., Ma, W., Zhang, M., Liu, Y., & Ma, S. (2022). A survey on the fairness of recommender systems. ACM Journal of the ACM (JACM).

9. Beyond Matrix Completion and Accuracy (Janka Marschalková)

The main goal of recommender systems is to predict unknown ratings of items for users. This can be seen as the task to complete the user-item matrix. Methods such as matrix factorization can solve this task and have been successfully applied in various domains. However, for some scenarios these general approaches work not as well. It is important to consider other aspects and metrics such as diversity, coverage, novelty, and serendipity.

  • Jannach, D., Resnick, P., Tuzhilin, A., & Zanker, M. (2016). Recommender systems - beyond matrix completion. Communications of the ACM, 59(11), 94-102.
  • Kaminskas, M., & Bridge, D. (2016). Diversity, serendipity, novelty, and coverage: a survey and empirical analysis of beyond-accuracy objectives in recommender systems. ACM Transactions on Interactive Intelligent Systems (TiiS), 7(1), 1-42.
  • Knees, P., Neidhardt, J., & Nalis, I. (2023). Recommender Systems: Techniques, Effects, and Measures Toward Pluralism and Fairness. In Introduction to Digital Humanism: A Textbook (pp. 417-434). Cham: Springer Nature Switzerland. (https://link.springer.com/chapter/10.1007/978-3-031-45304-5_27)

10. Privacy-Preserving Recommender Systems (René Jung)

Recommender systems generate personalized recommendation based on information about users. The more accurate this information is, the better recommendations can be tailored towards user needs and interests. However, collecting and utilizing personal data raises privacy issues. Users may be unaware which data is collected and do not want systems to acquire information about themselves. There are existing solutions to generate personalized recommendation while still respecting user privacy. Thus, this is topic about the trade-off between personalization and privacy in recommender systems.

  • Ogunseyi, T. B., Avoussoukpo, C. B., & Jiang, Y. (2023). A systematic review of privacy techniques in recommendation systems. International Journal of Information Security, 1-14.
  • Chen, C., Liu, Z., Zhao, P., Zhou, J., & Li, X. (2018, April). Privacy preserving point-of-interest recommendation using decentralized matrix factorization. In Proceedings of the AAAI conference on artificial intelligence (Vol. 32, No. 1).
  • Himeur, Y., Sohail, S. S., Bensaali, F., Amira, A., & Alazab, M. (2022). Latest trends of security and privacy in recommender systems: a comprehensive review and future perspectives. Computers & Security, 118, 102746.

11. Human-Centered Recommender Systems and Decision Making (Agata Agnieszka Andrysiak)

The user interface influences how people make decision. For example, when searching for products it is not only important to list appropriate items but also consider how to present them. This topic should discuss some aspects of the interplay between user interface design and human decision making in recommender systems. In addition, psychological factors such as personality, emotions, and decision biases can significantly affect the outcome of a decision process.

  • l. Jameson, A., Willemsen, M. C., & Felfernig, A. (2022). Individual and Group Decision Making and Recommender Systems. In Recommender Systems Handbook (pp. 789-832). Springer, New York, NY.
  • Konstan, J., & Terveen, L. (2021). Human-centered recommender systems: Origins, advances, challenges, and opportunities. AI Magazine, 42(3), 31-42.
  • Tran, T. N. T., Felfernig, A., & Tintarev, N. (2021). Humanized recommender systems: State-of-the-art and research issues. ACM Transactions on Interactive Intelligent Systems (TiiS), 11(2), 1-41.

12. Evaluating the User Experience of Recommender Systems (Christian Briegel)

The quality of a recommender system is not only determined by the accuracy of the recommendations. The overall user experience (UX) when interacting with the recommender system is a very important factor which decides if people like using the recommender system. According to Hassenzahl (2008), UX is "a momentary, primarily evaluative feeling (good-bad) while interacting with a product or service. Good UX is the consequence of fulfilling the human needs for autonomy, competence, stimulation (self-oriented) through interacting with the product or service (i.e. hedonic quality)." Many methods exist to evaluate the UX of recommender systems but also the usability of the system's user interfaces which is one very important aspect of UX.

  • Knijnenburg, B. P., Willemsen, M. C., Gantner, Z., Soncu, H., & Newell, C. (2012). Explaining the user experience of recommender systems. User modeling and user-adapted interaction, 22(4), 441-504.
  • Pu, P., Chen, L., & Hu, R. (2011, October). A user-centric evaluation framework for recommender systems. In Proceedings of the fifth ACM conference on Recommender systems (pp. 157-164).
  • Champiri, Z. D., Mujtaba, G., Salim, S. S., & Chong, C. Y. (2019, January). User experience and recommender systems. In 2019 2nd International Conference on Computing, Mathematics and Engineering Technologies (iCoMET) (pp. 1-5). IEEE.

13. Interactive Recommender Systems for the Internet of Things (Quanyi Liu)

The Internet of Things (IoT) is about connecting physical devices and smart object and thus enable a new range of services such as automating logistics. But is also important to consider how people can interact with these kind of smart objects. It is needed to come up with suitable interaction patterns both for administering and configurating IoT devices and also (end) users having to handle smart objects. One if the problems is that these devices often do not have a build-in user interface for interaction, such as a display, but have to be managed using a smartphone app, for example, that only shows an abstract representation of the object. Recommender systems could play an important role in supporting users discovering and interacting with IoT services.

* Felfernig, A., Polat-Erdeniz, S., Uran, C., Reiterer, S., Atas, M., Tran, T. N. T., ... & Dolui, K. (2019). An overview of recommender systems in the internet of things. Journal of Intelligent Information Systems, 52(2), 285-309.
* Palaiokrassas, G., Karlis, I., Litke, A., Charlaftis, V., & Varvarigou, T. (2017, July). An IoT architecture for personalized recommendations over big data oriented applications. In 2017 IEEE 41st Annual Computer Software and Applications Conference (COMPSAC) (Vol. 2, pp. 475-480). IEEE.
* Bergman, J., Olsson, T., Johansson, I., & Rassmus-Gröhn, K. (2018, March). An exploratory study on how Internet of Things developing companies handle User Experience Requirements. In International Working Conference on Requirements Engineering: Foundation for Software Quality (pp. 20-36). Springer, Cham.

14. Decentralized Recommender Systems (Saltuk Bugra Karacan)

Usually, recommender systems are centralized, i.e. user ratings and all other information is stored and managed on centralized servers. However, there are reasons to consider decentralization of recommender systems. Potential advantages include better scalability and robustness, but also potential benefits concerning privacy. In mobile scenarios, users could manage their preferences, ratings, and other personal information on their mobile devices to improve control over their data. Collaborative filtering could still be employed by allowing selective data exchange from device to device in a local neighborhood, for example. 

* Wang, Z., Liu, X., Chang, S., Zhou, J., Qi, G. J., & Huang, T. S. (2015). Decentralized recommender systems. arXiv preprint arXiv:1503.01647.
* Beierle, F., & Egger, S. (2020). MobRec—mobile platform for decentralized recommender systems. IEEE Access, 8, 185311-185329.
* Belal, Y., Bellet, A., Mokhtar, S. B., & Nitu, V. (2022). Pepper: Empowering user-centric recommender systems over gossip learning. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 6(3), 1-27.

Contact

To top