Mixed-initiative recommender systems
•
2 likes•496 views
Slides of my talk at Data Science Leuven 21 November 2018
Recommended
Recommended
More Related Content
Similar to Mixed-initiative recommender systems
Similar to Mixed-initiative recommender systems (20)
More from Katrien Verbert
More from Katrien Verbert (14)
Recently uploaded
Recently uploaded (20)
Mixed-initiative recommender systems
- 1. Mixed-initiative Recommender Systems Katrien Verbert Augment/HCI - KU Leuven @katrien_v
- 2. Human-Computer Interaction group recommender systems – visualization – intelligent user interfaces Learning analytics Media consumption Research Information Systems Wellness & health Augment prof. Katrien Verbert ARIA prof. Adalberto Simeone Computer Graphics prof. Phil Dutré Language Intelligence & Information Retrieval prof. Sien Moens
- 3. Augment/HCI team Robin De Croon Postdoc researcher Katrien Verbert Associate Professor Francisco Gutiérrez PhD researcher Tom Broos PhD researcher Martijn Millecamp PhD researcher Sven Charleer Postdoc researcher Nyi Nyi Htun Postdoc researcher Houda Lamqaddam PhD researcher Yucheng Jin PhD researcher Oscar Alvarado PhD researcher http://augment.cs.kuleuven.be/ Diego Rojo Carcia PhD researcher
- 4. Collaborative filtering – Content-based filtering Knowledge-based filtering - Hybrid Recommendation techniques
- 5. Mixed-initiative recommender systems Core objectives: • Explaining recommendations to increase user trust and acceptance • Enable users to interact with the recommendation process
- 8. 8
- 9. Interactive recommender systems Transparency: explaining the rational of recommendations User control: closing the gap between browse and search Diversity – novelty Cold start Context-aware interfaces 9 He, C., Parra, D. and 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, pp.9-27.
- 10. Flexible interaction with RecSys Research visit Host: Carnegie Mellon University & University of Pittsburg Collaboration: John Stamper, Peter Brusilovsky, Denis Parra Period: April 2012 – June 2012 Second post-doctoral fellowship FWO host university: KU Leuven, Belgium supervisor: Erik Duval period: Oct 2012 – Sept 2015 10
- 11. Overview research topics 11 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 Learning Analytics - Media Consumption – Research Information Systems - Healthcare
- 13. Also recommendations personalized relevance prospect 13
- 14. Contributions new approach to support exploration, transparency and controllability recommender systems are shown as agents in parallel to real users and tags users can interrelate entities to find items evaluation study that assesses effectiveness probability of item selection 14 Verbert, K., Parra, D., Brusilovsky, P., & Duval, E. (2013). Visualizing recommendations to support exploration, transparency and controllability. In Proceedings of the IUI 2013 international conference on Intelligent user interfaces (pp. 351-362). ACM.
- 15. TalkExplorer - I Entities Tags, Recommender Agents, Users 15
- 17. Results of studies 1 & 2 Effectiveness: # bookmarked items / #explorations Effectiveness increases with intersections of more entities Effectiveness wasn’t affected in the field study (study 2) … but exploration distribution was affected 17 Average effectiveness Total number of explorations Verbert, K., Parra, D., & Brusilovsky, P. (2016). Agents vs. users: Visual recommendation of research talks with multiple dimension of relevance. ACM Transactions on Interactive Intelligent Systems (TIIS), 6(2), 11.
- 19. Three user studies Study 1: Within-subjects study with 20 users baseline: exploration of recommendations in CN3 Second condition: exploration of recommendations in IEx Data from two conferences EC-TEL 2014, EC-TEL 2015 Study 2: Field study at Digital Humanities conference + 1000 participants, less technically oriented Study 3: Field study at IUI conference Smaller scale, technical audience 19
- 20. Study 1 vs Study 2 vs Study 3 Overall ”augmented agents” were used in all three studies Precision scores significantly higher for augmented agents in study 1 and study 3 Participants of study 2 (Digital Humanities) more interested in content perspective Rated several dimensions lower (use intention, fun, information sufficiency, control) 20 Cardoso, B., Sedrakyan, G., Gutiérrez, F., Parra, D., Brusilovsky, P., & Verbert, K. (2018). IntersectionExplorer, a multi- perspective approach for exploring recommendations. International Journal of Human-Computer Studies.
- 21. Overview research topics 21 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 Learning analytics - Media Consumption – Research Information Systems - Healthcare
- 22. Explanations 22
- 23. Personal characteristics Need for cognition • Measurement of the tendency for an individual to engage in, and enjoy, effortful cognitive activities • Measured by test of Cacioppo et al. [1984] Visualisation literacy • Measurement of the ability to interpret and make meaning from information presented in the form of images and graphs • Measured by test of Boy et al. [2014] Locus of control (LOC) • Measurement of the extent to which people believe they have power over events in their lives • Measured by test of Rotter et al. [1966] Visual working memory • Measurement of the ability to recall visual patterns [Tintarev and Mastoff, 2016] • Measured by Corsi block-tapping test Musical experience • Measurement of the ability to engage with music in a flexible, effective and nuanced way [Müllensiefen et al., 2014] • Measured using the Goldsmiths Musical Sophistication Index (Gold-MSI) Tech savviness • Measured by confidence in trying out new technology 23
- 24. User study Within-subjects design: 105 participants recruited with Amazon Mechanical Turk Baseline version (without explanations) compared with explanation interface Pre-study questionnaire for all personal characteristics Task: Based on a chosen scenario for creating a play-list, explore songs and rate all songs in the final playlist Post-study questionnaire: Recommender effectiveness Trust Good understanding Use intentions Novelty Satisfaction Confidence
- 25. Results 25 The interaction effect between NFC (divided into 4 quartiles Q1-Q4) and interfaces in terms of confidence
- 26. Design implications Explanations should be personalised for different groups of end- users. Users should be able to choose whether or not they want to see explanations. Explanation components should be flexible enough to present varying levels of details depending on a user’s preference. 26
- 27. User control Users tend to be more satisfied when they have control over how recommender systems produce suggestions for them (Konstan and Riedl, 2012) Control recommendations Douban FM Control user profile Spotify Control algorithm parameters TasteWeights
- 28. Controllability Cognitive load Additional controls may increase cognitive load (Andjelkovic et al. 2016; Jin et al. 2017)
- 29. Different levels of user control 29 Level Recommender components Controls low Recommendations (REC) Rating, removing, and sorting medium User profile (PRO) Select which user profile data will be considered by the recommender high Algorithm parameters (PAR) Modify the weight of different parameters Jin, Y., Tintarev, N., & Verbert, K. (2018, September). Effects of personal characteristics on music recommender systems with different levels of controllability. In Proceedings of the 12th ACM Conference on Recommender Systems (pp. 13-21). ACM.
- 30. User profile (PRO) Algorithm parameters (PAR) Recommendations (REC) 8 control settings No control REC PAR PRO REC*PRO REC*PAR PRO*PAR REC*PRO*PAR
- 31. Evaluation method Between-subjects – 240 participants recruited with AMT Independent variable: settings of user control 2x2x2 factorial design Dependent variables: Acceptance (ratings) Cognitive load (NASA-TLX), Musical Sophistication, Visual Memory Framework Knijnenburg et al. [2012]
- 32. Results Main effects: from REC to PRO to PAR → higher cognitive load Two-way interaction: does not necessarily result in higher cognitive load. Adding an additional control component to PAR increases the acceptance. PRO*PAR has less cognitive load than PRO and PAR High MS leads to higher quality, and thereby result in higher acceptance 32
- 33. Overview research topics 33 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 Learning Analytics - Media Consumption – Research Information Systems - Healthcare
- 34. Recommender systems for food 34
- 35. Augmented reality 35 Gutiérrez, Francisco, Htun, Nyi Nyi, Charleer, Sven, De Croon, Robin, Verbert, Katrien (2019) Designing Augmented Reality Applications for Personal Health Decision-Making. Proceedings of HICSS-52.
- 36. Job explorer 36
- 38. RECOMMENDE R ALGORITHMS MACHINE LEARNING INTERACTIVE DASHBOARDS SMART ALERTS RICH CARE PLANS OPEN IoT ARCHITECTURE
- 39. Peter Brusliovsky NavaTintarev CristinaConatiDenis Parra Collaborations Bart Knijnenburg Jurgen Ziegler
- 41. References Boy, J., Rensink, R. A., Bertini, E., & Fekete, J. D. (2014). A principled way of assessing visualization literacy. IEEE transactions on visualization and computer graphics, 20(12), 1963-1972. Cacioppo, J.T., Petty, R.E. and Feng Kao, C., 1984. The efficient assessment of need for cognition. Journal of personality assessment, 48(3), pp.306-307. Konstan, J.A. and Riedl, J., 2012. Recommender systems: from algorithms to user experience. User modeling and user-adapted interaction, 22(1-2), pp.101-123. Müllensiefen, D., Gingras, B., Musil, J., & Stewart, L. (2014). The musicality of non-musicians: an index for assessing musical sophistication in the general population. PloS one, 9(2), e89642. Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological monographs: General and applied, 80(1), 1. Tintarev, N., & Masthoff, J. (2016). Effects of Individual Differences in Working Memory on Plan Presentational Choices. Frontiers in psychology, 7, 1793.
Editor's Notes
- Amazon.com gebruikt een collaborative filtering techniek: zoekt gelijkenissen tussen gebruikers en gaat dan op basis van wat gelijkaardige gebruikers kopen aanbevelingen doen.
- Bring tools that have been researched for decades within reach of lay-humanists – traditional scholar not well versed with computational methods
- - In field study, users explore mainly clusters connected to one entity Users don’t even get to explore interactions between 4 items enabling users to explore interrelationships between prospects increases probability of finding a relevant item interrelating tags with other entities increase their effectiveness significantly
- The procedure contains the following steps: \begin{enumerate} \item \textit{Tutorial of study} - Participants were invited to read the description of the user study and to choose a scenario for generating a play-list. Then, they were asked to watch a task tutorial. Only the features of the particular setting were shown in this video. The ``Start'' button of the study was only activated after finishing the tutorial. Users logged in with their Spotify accounts to our experimental system, so that our recommenders could leverage the Spotify API and user listening history to generate ``real'' recommendations. \item \textit{Pre-study questionnaire} - This questionnaire collects user demographics and measures user's personal characteristics such as musical sophistication and visual memory capacity. %and their trust in recommender systems. The visual memory capacity is measured by ``Corsi block-tapping test''. In the test, a number of tiles are highlighted one at a time, and participants are asked to select the tiles in the correct order afterward. The number of highlighted tiles increases until the user makes too many errors. In Experiments 1 and 3, we used a test with a more sophisticated implementation of the Corsi test~\footnote{\url{https://www.humanbenchmark.com/tests/memory}, accessed June 2018}, which allows us to better distinguish participants by the level of visual memory capacity. In Experiment 2, to control the workload of participants in the within-subjects design, we chose a simple version of the Corsi test~\footnote{\url{http://www.psytoolkit.org/experiment-library/corsi.html}, accessed June 2018} for measuring visual short-term memory. \end{itemize} \item \textit{Manipulating Recommender and rating songs} - To ensure that participants spent enough time to explore recommendations, the questionnaire link was only activated after 10 minutes. After tweaking the recommender, participants were asked to rate the top-20 recommended songs that resulted from their interactions. \item \textit{Post-study questionnaire} - Participants were asked to evaluate the perceived quality, perceived accuracy, perceived diversity, satisfaction, effectiveness, and choice difficulty of the recommender system. After answering all the questions, participants were given opportunities to provide free-text comments of their opinions and suggestions about our recommender. \end{enumerate}
- Figure 2 shows that participants with low NFC are reporting a higher confident in their playlist with the explanations in- terface than in the baseline. Participants with a high NFC reported the opposite. Hence, the participants with low NFC have more confidence in the explanation interface than in the baseline, in contrast to user with low NFC. An explanation might be that low NFC participants benefited from the expla- nations because they did not spontaneously engage in much extra reasoning to justify the recommendations they received, and when they received the rational from the explanation this increased their confidence in their songs selection. Figure 2 also indicates that as the NFC increased, the con- fidence of participants in the playlist created in the baseline also increased. This result indicated that participants with a high NFC were more willing to understand their own musical preference in relation to the attributes of the recommended songs. This may have resulted in a higher confidence in their playlist. We did not see the same increase in trust as NFC increases in the explanation interface. As Figure 2 shows, the NFC scores in the third quartile were almost the same for both interfaces. At the highest NFC level, participants had a higher confidence in the baseline than in the explanation interface. The reduced confidence within the explanation interface could be an indication that users with a high NFC have less need for explanations.
- We employed a between-subjects study to investigate the effects of interactions among different user control on acceptance, perceived diversity, and cognitive load. We consider each of three user control components as a variable. By following the 2x2x2 factorial design we created eight experimental settings (Table~\ref{tab:table1}), which allows us to analyze three main effects, three two-way interactions, and one three-way interaction. We also investigate which specific \textit{personal characteristics} (musical sophistication, visual memory capacity) influence acceptance and perceived diversity. Each experimental setting is evaluated by a group of participants (N=30). Of note, to minimize the effects of UI layout, all settings have the same UI and disable the unsupported UI controls, e.g., graying out sliders. As shown in section ~\ref{evaluation questions}, we employed Knijnenburg et al.'s framework~\citep{knijnenburg2012explaining} to measure the six subjective factors, perceived quality, perceived diversity, perceived accuracy, effectiveness, satisfaction, and choice difficulty~\citep{knijnenburg2012explaining}. In addition, we measured cognitive load by using a classic cognitive load testing questionnaire, the NASA-TLX~\footnote{https://humansystems.arc.nasa.gov/groups/tlx}. It assesses the cognitive load on six aspects: mental demand, physical demand, temporal demand, performance, effort, and frustration. The procedure follows the design outlined in the general methodology (c.f., Section \ref{sec:general-procedure}). The \textit{experimental task} is to compose a play-list for the chosen scenario by interacting with the recommender system. Participants were presented with play-list style recommendations (Figure~\ref{fig:vis1}c). Conditions were altered on a between-subjects basis. Each participant was presented with only one setting of user control. For each setting, initial recommendations are generated based on the selected top three artists, top two tracks, and top one genre. According to the controls provided in a particular setting, participants were able to manipulate the recommendation process.
- Main effects: REC has lowest cgload and highest acceptance Two-way: All the settings that combine two control components do not lead to significantly higher cognitive load than using only one control component. combing multiple control components potentially increases acceptance without increasing cognitive load significantly. visual memory is not a significant factor that affects the cognitive load of controlling recommender systems. In other words, controlling the more advanced recommendation components in this study does not seem to demand a high visual memory. In addition, we did not find an effect of visual memory on acceptance (or perceived accuracy and quality). One possible explanation is that users with higher musical so- phistication are able to leverage different control components to explore songs, and this influences their perception of recommenda- tion quality, thereby accepting more songs. Our results show that the settings of user control significantly influence cognitive load and recommendation acceptance. We discuss the results by the main effects and interaction effects in a 2x2x2 factorial design. Moreover, we discuss how visual memory and musical sophistication affect cognitive load, perceived diversity, and recommendation acceptance. \subsubsection{Main effects} We discuss the main effects of three control components. Increased control level; from control of recommendations (REC) to user profile (PRO) to algorithm parameters (PAR); leads to higher cognitive load (see Figure \ref{fig:margin}c). The increased cognitive load, in turn, leads to lower interaction times. Compared to the control of algorithm parameters (PAR) or user profile (PRO), the control of recommendations (REC) introduces the least cognitive load and supports users in finding songs they like. We observe that most existing music recommender systems only allow users to manipulate the recommendation results, e.g., users provide feedback to a recommender through acceptance. However, the control of recommendations is a limited operation that does not allow users to understand or control the deep mechanism of recommendations. \subsubsection{Two-way interaction effects} Adding multiple controls allows us to improve on existing systems w.r.t. control, and do not necessarily result in higher cognitive load. Adding an additional control component to algorithm parameters increases the acceptance of recommended songs significantly. Interestingly, all the settings that combine two control components do \textit{not} lead to significantly higher cognitive load than using only one control component. We even find that users' cognitive load is significantly \textit{lower} for (PRO*PAR) than (PRO, PAR), which shows a benefit of combining user profile and algorithm parameters in user control. Moreover, combing multiple control components potentially increases acceptance without increasing cognitive load significantly. Arguably, it is beneficial to combine multiple control components in terms of acceptance and cognitive load. \subsubsection{Three-way interaction effects} The interaction of PRO*PAR*REC tends to increase acceptance (see Figure \ref{fig:margin}a), and it does not lead to higher cognitive load (see Figure \ref{fig:margin}c). Moreover, it also tends to increase interaction times and accuracy. Therefore, we may consider having three control components in a system. Consequently, we answer the research question. \textbf{RQ1}: \textit{The UI setting (user control, visualization, or both) has a significant effect on recommendation acceptance?} It seems that combining PAR with a second control component or combing three control components increases acceptance significantly. %KV: this paragraph refers to different RQs: either rephrase or omit? -SOLVED \subsubsection{Effects of personal characteristics} Having observed the trends across all users, we survey the difference in cognitive load and item acceptance due to personal characteristics. We study two kinds of characteristics: visual working memory and musical sophistication. \paragraph{Visual working memory} The SEM model suggests that visual memory is not a significant factor that affects the cognitive load of controlling recommender systems. The cognitive load for the type of controls used may not be strongly affected by individual differences in visual working memory. In other words, controlling the more advanced recommendation components in this study does not seem to demand a high visual memory. In addition, we did not find an effect of visual memory on acceptance (or perceived accuracy and quality). Finally, the question items for diversity did not converge in our model, so we are not able to make a conclusion about the influence of visual working memory on diversity. \paragraph{Musical sophistication} Our results imply that high musical sophistication allows users to perceive higher recommendation quality, and may thereby be more likely to accept recommended items. However, higher musical sophistication also increases choice difficulty, which may negatively influence acceptance. One possible explanation is that users with higher musical sophistication are able to leverage different control components to explore songs, and this influences their perception of recommendation quality, thereby accepting more songs. Finally, the question items for diversity did not converge in our model, so we are not able to make a conclusion about the influence of musical sophistication on diversity.