facebook

linkedin

User Modeling and User-Adapted Interaction (UMUAI) provides an interdisciplinary forum for the dissemination of new research results on interactive computer systems that can be adapted or adapt themselves to their current users, and on the role of user models in the adaptation process.

UMUAI has been published since 1991 by Kluwer Academic Publishers (now merged with Springer Verlag).

UMUAI homepage with description of the scope of the journal and instructions for authors.

Springer UMUAI page with online access to the papers.

Latest Results for User Modeling and User-Adapted Interaction

09 May 2021

The latest content available from Springer
  • Personalized task difficulty adaptation based on reinforcement learning

    Abstract

    Traditionally, the task difficulty level is often determined by domain experts based on some hand-crafted rules. However, with the adoption of Massive Open Online Courses (MOOCs), it has become harder to manually personalize task difficulty as the system designers are faced with a very large question bank and a user base of individuals with diverse backgrounds and ability levels. This research focuses on developing a data-driven method to adaptively adjust difficulty levels in order to maintain a target user performance level over a series of tasks whose difficulty level is highly variable among different individuals. Specifically, the issue of difficulty adaptation was formulated as a reinforcement learning problem. To ensure responsiveness of the interactive systems, a novel bootstrapped policy gradient (BPG) framework was developed, which can incorporate prior knowledge of difficulty ranking into policy gradient to enhance sample efficiency. To obtain high-quality prior information on difficulty ranking, a clustering-based approach was proposed which can learn a personalized difficulty ranking to capture users’ individual differences. To evaluate the effectiveness of the difficulty adaptation method, we focused on a visual memory training problem with a large question bank and a diverse user base. Specifically, the proposed algorithms were combined and applied to a real-world application consisting of an online visual-spatial memory recall game and were shown to outperform the traditional rule-based adaptation approach in adapting to the slow players while achieving comparable performance in adapting to the fast players.

  • The role of transparency in multi-stakeholder educational recommendations

    Abstract

    Recommender systems have been successfully applied to alleviate the problem of information overload and assist users’ decision makings. Multi-stakeholder recommender systems produce the item recommendations to the end user by considering the perspective of multiple stakeholders. Existing research on multi-stakeholder recommendations relies on the offline evaluations only, and the online studies are still under investigation. This paper made the first attempt to examine the multi-stakeholder recommendations through online studies. On the one hand, we use online user studies to compare different recommendation models. On the other hand, we develop novel user interfaces to enhance the transparency and exploit the role of transparency in multi-stakeholder recommender systems. We collect our own dataset in educational learning and use it as case study to perform the online studies in multi-stakeholder recommendations. We observe that the explanation of the key parameters in the recommendation models can enhance the transparency, which further affects the decision making of different stakeholders.

  • A serious game to extract Hofstede’s cultural dimensions at the individual level

    Abstract

    Cultural dimensions are an important aspect of a user model and useful for many applications such as adapting user interface, managing marketing campaigns, customer relationship management, and human resource management. Traditionally, these dimensions are measured through CVSCALE, which is a reliable and standard scale provided to measure Hofstede’s cultural dimensions at the individual level. The problems with the questionnaire-based data collection are low response rates, lack of willingness to complete, poor engagement of participants, and concern about the quality of data collected. To solve these problems, we present a serious video game called “Treasure Island” to measure the cultural dimensions of the player at the individual level (currently only for the Persian language). We have developed and validated a Persian-language version of CVSCALE applied to design the game. We developed the very first general game development process to build serious games for gathering the same data as a standard psychometric questionnaire. Treasure Island has been evaluated by statistical analysis of the data collected from a sample of 285 participants that have played the game and completed the questionnaire. The results indicate that Treasure Island is effective to measure the individual cultural dimensions. Moreover, the efficiency of the game has been tested in terms of task load imposed on the user during playing the game and completing the questionnaire. The results demonstrate that the game imposes less task load on the user, consequently, improves user satisfaction and engagement. Since individual cultural dimensions can be considered an important facet of a user model for certain applications, the proposed serious game can be applied for user modeling and personalization purposes.

  • BARGAIN: behavioral affective rule-based games adaptation interface–towards emotionally intelligent games: application on a virtual reality environment for socio-moral development

    Abstract

    This paper presents a framework for adapting game elements to the player’s affective state and the integration of the framework in a virtual reality environment for moral development. These game elements include gestural and facial expressions of avatars during dialogues with the player, background music, the score, game mechanics, aesthetics and learning. The framework BARGAIN (Behavioral Affective Rule-based Games Adaptation Interface) is an authoring tool for affective game design providing a visual interface based on finite state machine (FSM) technique to represent the affective rules as state transitions graph dependent on the player emotional state assessed using facial expression recognition system based on electroencephalography (EEG) data. We conducted a user study (n = 29) examining the effects of the resulting affective virtual reality game on players’ experience using the Game experience Questionnaire (GEQ) (IJsselsteijn et al. in The game experience questionnaire, Technische Universiteit Eindhoven, Eindhoven, 2013). The results show significant correlation between the GEQ dimensions and the player's facial expressions during his interaction with the Non-Player Characters (NPCs) within the VR game. These findings highlight that adapting games to user's emotions enhance the players’ experience.

  • Meta-User2Vec model for addressing the user and item cold-start problem in recommender systems

    Abstract

    The cold-start scenario is a critical problem for recommendation systems, especially in dynamically changing domains such as online news services. In this research, we aim at addressing the cold-start situation by adapting an unsupervised neural User2Vec method to represent new users and articles in a multidimensional space. Toward this goal, we propose an extension of the Doc2Vec model that is capable of representing users with unknown history by building embeddings of their metadata labels along with item representations. We evaluate our proposed approach with respect to different parameter configurations on three real-world recommendation datasets with different characteristics. Our results show that this approach may be applied as an efficient alternative to the factorization machine-based method when the user and item metadata are used and hence can be applied in the cold-start scenario for both new users and new items. Additionally, as our solution represents the user and item labels in the same vector space, we can analyze the spatial relations among these labels to reveal latent interest features of the audience groups as well as possible data biases and disparities.