Academic Positions

  • Present 2014

    PhD Student

    Eindhoven University, Human-Technology Interaction Group

    Royal Philips, Data Science Department

Commercial Roles

  • 2014 2013

    Data Scientist

    O2mc, Uden, the Netherlands

  • 2013 2011

    Technical Consultant Online Analytics

    Adversitement, Uden, the Netherlands

Education & Training

  • M.Sc. 2011

    Master of Science in Human-Technology Interaction

    Eindhoven University of Technology

  • B.Sc.2008

    Bachelor of Science in Innovation Sciences

    Eindhoven University of Technology

Research Summary

My main research interest is online personalization. While most research in this field puts emphasis either on the machine learning aspect or the human-computer interaction aspects, I feel that personalization requires a closer link between these two. My goal is to find a right balance in research and preventing on the one hand research on algorithms that is based on over-simplifications of users and their behavior or on the other hand in more user-focussed research that is under-engineered.

Personalization involves on the one hand mining or analyzing large amounts of data, which requires a thorough understanding of the current methods. On the other hand, as personalization aims to help people, it requires both understanding the user, as actually verifying that the personalization indeed helps the user.

My research direction was established during my master thesis on recommender systems. The observation that current collaborative filtering recommender algorithms share similarities with the way decision making psychologists operationalize preferences, lead to a study to see if that similarity can be used to make recommender systems more understandable to their users. This study lead to more work on diversification in recommender systems.

I applied the same approach in the domain of website adaptation. I investigated to what extent we can incorporate knowledge of website owners on their audience in real-time online personalization. I did this by not only considering the observable data, but relating this data to visitor segments website owners assumed to have. This allowed for a more controlled, transparent implementation of the website adaptations, as well as verifying that the hypothesized segments actually exist.

Interests

  • Recommender Systems
  • Website Personalization/Adaptation
  • Machine Learning
  • Measuring Psychological Traits

Filter by type:

Sort by year:

How item discovery enabled by diversity leads to increased recommendation list attractiveness

Bruce Ferwerda, Mark P. Graus, Andreu Vall, Marko Tkalcic, Markus Schedl
Conference PapersProceedings of the Symposium on Applied Computing, 2017, 1693-1696

Abstract

We study how a website adaptation based on segment predictions from click streams affects visitor behavior and user experience. Through statistical analysis we investigate how the adaptation changed actual behavior. Through structural equation modeling of subjective experience we answer why the change in behavior occurred. The study shows the value of using survey data for constructing and evaluating predictive models. It additionally shows how a website adaptation influences user experience and how this in turn influences visitor behavior.

Can trailers help to alleviate popularity bias in choice-based preference elicitation?

Mark P. Graus, Martijn C. Willemsen
Workshop PapersJoint Workshop on Interfaces and Human Decision Making for Recommender Systems

Abstract

Previous research showed that choice-based preference elicitation can be successfully used to reduce effort during user cold start, resulting in an improved user satisfaction with the recommender system. However, it has also been shown to result in highly popular recommendations. In the present study we investigate if trailers reduce this bias to popular recommendations by informing the user and enticing her to choose less popular movies. In a user study we show that users that watched trailers chose relatively less popular movies and how trailers affected the overall user experience with the recommender system.

The Influence of Users’ Personality Traits on Satisfaction and Attractiveness of Diversified Recommendation Lists

Bruce Ferwerda, Mark P. Graus, Andreu Vall, Marko Tkalcic, Markus Schedl
Workshop Papers4th Workshop on Emotions and Personality in Personalized Systems (EMPIRE) 2016

Abstract

Diversifying recommendations has shown to be a good means to counteract on choice difficulties and overload, and is able to positively influence subjective evaluations, such as satisfaction and attractiveness. Personal characteristics (e.g., domain expertise, prior preference strength) have shown to influence the desired level of diversity in a recommendation list. However, only personal characteristics that are directly related to the domain have been investigated so far. In this work we take personality traits as a general user model and show that specific traits are related to a preference for different levels of diversity (in terms of recommendation satisfaction and attractiveness). Among 103 participants we show that conscientiousness is related to a preference for a higher degree of diversification, while agreeableness is related to a mid-level diversification of the recommendations. Our results have implications on how to personalize recommendation lists (i.e., the amount of diversity that should be provided) depending on users’ personality.

Understanding the role of latent feature diversification on choice difficulty and satisfaction

Martijn C. Willemsen, Mark P. Graus, Bart P. Knijnenburg
Journal Paper User Modeling and User-Adapted Interaction 26(4), Pages 347-389

Abstract

People like variety and often prefer to choose from large item sets. However, large sets can cause a phenomenon called “choice overload”: they are more difficult to choose from, and as a result decision makers are less satisfied with their choices. It has been argued that choice overload occurs because large sets contain more similar items. To overcome this effect, the present paper proposes that increasing the diversity of item sets might make them more attractive and satisfactory, without making them much more difficult to choose from. To this purpose, by using structural equation model methodology, we study diversification based on the latent features of a matrix factorization recommender model. Study 1 diversifies a set of recommended items while controlling for the overall quality of the set, and tests it in two online user experiments with a movie recommender system. Study 1a tests the effectiveness of the latent feature diversification, and shows that diversification increases the perceived diversity and attractiveness of the item set, while at the same time reducing the perceived difficulty of choosing from the set. Study 1b subsequently shows that diversification can increase users’ satisfaction with the chosen option, especially when they are choosing from small, diverse item sets. Study 2 extends these results by testing our diversification algorithm against traditional Top-N recommendations, and finds that diverse, small item sets are just as satisfying and less effortful to choose from than Top-N recommendations. Our results suggest that, at least for the movie domain, diverse small sets may be the best thing one could offer a user of a recommender system.

Understanding real-life website adaptations by investigating the relations between user behavior and user experience

Mark P. Graus, Martijn C. Willemsen, Kevin Swelsen
Conference PapersProceedings of User Modeling, Adaptation and Personalization (UMAP) 2015, 217-220

Abstract

We study how a website adaptation based on segment predictions from click streams affects visitor behavior and user experience. Through statistical analysis we investigate how the adaptation changed actual behavior. Through structural equation modeling of subjective experience we answer why the change in behavior occurred. The study shows the value of using survey data for constructing and evaluating predictive models. It additionally shows how a website adaptation influences user experience and how this in turn influences visitor behavior.

Improving the User Experience during Cold Start through Choice-Based Preference Elicitation

Mark P. Graus, Martijn C. Willemsen
Conference PapersProceedings of the ninth ACM conference on Recommender systems, 273-276

Abstract

We studied an alternative choice-based interface for preference elicitation during the cold start phase and compared it directly with a standard rating-based interface. In this alternative interface users started from a diverse set covering all movies and iteratively narrowed down through a matrix factorization latent feature space to smaller sets of items based on their choices. The results show that compared to a rating-based interface, the choice-based interface requires less effort and results in more satisfying recommendations, showing that it might be a promising candidate for alleviating the cold start problem of new users.

Remembering the stars?: effect of time on preference retrieval from memory

Dirk Bollen, Mark P. Graus, Martijn C. Willemsen
Conference PapersProceedings of the sixth ACM conference on Recommender systems, 217-220

Abstract

Many recommendation systems rely on explicit ratings provided by their users. Often these ratings are provided long after consuming the item, relying heavily on people's representation of the quality of the item in memory. This paper investigates a psychological process, the "positivity effect", that influences the retrieval of quality judgments from our memory by which pleasant items are being processed and recalled from memory more effectively than unpleasant items. In an offline study on the MovieLens data we used the time between release date and rating date as a proxy for the time between consumption and rating. Ratings for movies tend to increase over time, consistent with the positivity effect. A subsequent online user study used a direct measure of time between rating and consumption, by asking users to rate movies (recently aired on television) and to explicitly report how long ago they watched these movies. In contrast to the offline study we find that ratings tend to decline over time showing reduced accuracy in ratings for items experienced long ago. We discuss the impact these rating dynamics might have on recommender algorithms, especially in cases where a new user has to submit his preferences to a system.

Understanding choice overload in recommender systems

Dirk Bollen, Bart P. Knijnenburg, Martijn C. Willemsen, Mark P. Graus
Conference PapersProceedings of the fourth ACM conference on Recommender systems, 63-70

Abstract

Even though people are attracted by large, high quality recommendation sets, psychological research on choice overload shows that choosing an item from recommendation sets containing many attractive items can be a very difficult task. A web-based user experiment using a matrix factorization algorithm applied to the MovieLens dataset was used to investigate the effect of recommendation set size (5 or 20 items) and set quality (low or high) on perceived variety, recommendation set attractiveness, choice ...

The virtual midas touch: helping behavior after a mediated social touch

Antal Haans, Wijnand IJsselsteijn, Mark P. Graus, Juho A. Salminen
Conference PapersCHI '08 Extended Abstracts on Human Factors in Computing Systems, 3507-3512

Abstract

Many recommendation systems rely on explicit ratings provided by their users. Often these ratings are provided long after consuming the item, relying heavily on people's representation of the quality of the item in memory. This paper investigates a psychological process, the "positivity effect", that influences the retrieval of quality judgments from our memory by which pleasant items are being processed and recalled from memory more effectively than unpleasant items. In an offline study on the MovieLens data we used the time between release date and rating date as a proxy for the time between consumption and rating. Ratings for movies tend to increase over time, consistent with the positivity effect. A subsequent online user study used a direct measure of time between rating and consumption, by asking users to rate movies (recently aired on television) and to explicitly report how long ago they watched these movies. In contrast to the offline study we find that ratings tend to decline over time showing reduced accuracy in ratings for items experienced long ago. We discuss the impact these rating dynamics might have on recommender algorithms, especially in cases where a new user has to submit his preferences to a system.

Contact & Meet Me

I'm always open to meet up and discuss anything related to my research. Depending on the day of the week, I can be found either in my office in Eindhoven University, on the High Tech Campus or in JADS in 's-Hertogenbosch.

  •    office: +31 (0) 40 247 8420
  •    m.p.graus@tue.nl
  •    mark.graus@philips.com
  •    m.p.graus@gmail.com
  •    marrk_nl
  •    @newmarrk
  •    linkedin.com/in/markgraus

On Eindhoven University of Technology Campus

Mondays, Tuesdays and Fridays I can be found in my office at Eindhoven University of Technology. My office number is IPO 0.20.

At Philips High Tech Campus

On Wednesdays I work in HTC31 on the High Tech Campus in Eindhoven.

At Jheronimus Academy of Data Science

On Tuesdays I work at JADS in 's-Hertogenbosch.