Spring 2014-15

Game-based emotion detection

Date: 16th-17th June 2015
Experiment conductor: Peter Gašpar
Supervisor: Mária Bieliková
Short description: There have already been many possibilities and tools for the detection of emotions. Many of them rely on software solutions that process streamed or recorded video of participants’ face. Another approach is to record their physiological status, especially an eye’s pupil. In our research we are trying to reveal pros and cons of each of these tools in order to offer the most suitable settings for another user-experience experiments. Another goal is to distinguish the accuracy of detected emotion in comparison to the available annotated datasets. We are using an interactive game to observe users’ mood and behaviour in order to analyze it subsequently. In our experiment we use an eye-tracker and a camera capable of recording users’ emotions from their face.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Profiling of user’s Internet expertise

Date: 15th June 2015
Experiment conductor: Peter Krátky
Supervisor: Alenka Kovárová, Jozef Tvarožek, Pavol Návrat, Daniela Chudá
Short description: To determine if a user is novice or experienced in Internet use, generally questionnaires are used. In our research we are going to develop a method to model such user expertise in implicit way that could be used easily in large scale. We propose monitoring mouse and keyboard use as well as gaze and correlating those data with survey data of operational, formal and informational Internet use.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Refining interest of information seekers on the Web by implicit feedback

Date: 29-30th April 2015
Experiment conductor: Jozef Marcin
Supervisor: Pavol Návrat
Short description: In this experiment we want to test our solution for refining interest of information seekers. We use eye-tracker to retrieve seeker´s gaze. When seeker gaze at some element on page, by XPATH we gathered from this element, we can obtain words, that are interesting for seeker´s web search in some way . By TF-IDF and some NLP operations, we can create some kind of vector, which holds interesting words and with this words we can enhance web search and provide more relevant results to seeker. To evaluate our solution, we need a set of seeker´s that will be searching for specific content by specific query and by our solution, we will provide more results to them. Subsequently, seekers will rate the results. Next we compare these results with our rating and evaluate our method by metrics explained in formal description.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Eye Tracked Crowd: Training Dataset Acquisition for TV Show Categorization

Date: 5-6th April 2015
Experiment conductor: Jakub Šimko, Peter Dubec
Supervisor: Mária Bieliková
Short description: In this crowdsourcing and eye-tracking experiment, descriptive categories are acquired for documentary TV shows in sample CSFD dataset using human manual work of multiple participants. Based on the TV show’s name (original and translated) and description (several senteces long), the worker’s task is to assign primary (and optionally secondary) content category (out of 30 possible options) to each given TV show. During working, participants are eyetracked and their clicking behavior logged. The purpose of the experiment is (1) to acquire training dataset for automated TV show categorization method and (2) gather implicit feedback data of an crowdsourcing activity, usable for validating future hypotheses about behavior of crowd workers and designing future method for improving gain from similar crowdsourcing activities.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Detection of Emotion Based on Picture Annotation

Date: 11th April 2015
Experiment conductor: Peter Gašpar
Supervisor: Mária Bieliková
Short description: There have already been many possibilities and tools for the detection of emotions. Many of them rely on software solutions that process streamed or recorded video of participants’ face. Another approach is to record their physiological status, especially an eye’s pupil or a skin conductance. In our research we are trying to reveal pros and cons of each of these tools in order to offer the most suitable settings for another user-experience experiments. Another goal is to distinguish the accuracy of detected emotion in comparison to the available annotated datasets. We use selected set of pictures to evoke an emotion in participant. Afterwards, participant’s task is to provide simple explicit feedback about his/her feeling to make the comparison even more interesting.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Deception Detection

Date: 11th April 2015
Experiment conductor: Metod Rybár
Supervisor: Mária Bieliková
Short description: In the experiment we are going to verify some hypothesis regarding filling out personality questionnaire BigFive. We are going to use implicit user measures, mainly eye-tracking, to detect user deceptive or non-honest behavior. We are looking for indications of such behavior such as change in time responses, order and style of reading the questions and answers or changing in physiological measures like pupil diameter.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Usability of Mobile Operator Websites (dataset acquisition)

Date: 15th-30th April 2015
Experiment conductor: Vladimír Ľalík
Supervisor: Jakub Šimko
Short description: In this experiment, we tracked the behavior of users during typical tasks performed on mobile operator websites, such as seeking of a nearest kiosk, acquiring information on phone account quota or ordering of a new phone. The experiment was conducted as a part of early phase of diploma thesis solving. The thesis is focused on automated user interface usability evaluation. The experiment had a primary purpose of collecting a dataset for further analysis.
Link to a formal description of the UX experiment: experiment description (in Slovak)

Evaluation of Game Learnability

Date: 20-24th April 2015
Experiment conductor: Peter Demčák
Supervisor: Jakub Šimko
Short description: Learning case mining (LCM) is a method aimed at support of game learnability evaluation in running playtests of games. It makes use on gaze tracking as a valuable and unobtrusive source of feedback for the playtests. The LCM method is based on finding the similarity of player behavior with learning cases, which are models of player behavior expected and formulated by game designers in advance, based on the game mechanics and dynamics they plan to playtest for. The purpose of this experiment is to apply our method to two games alongside a video review of the play sessions, to determine whether the LCM method can truly collect accurate information about the player’s learning of game mechanics/dynamics in the course of a playtest.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Implicit Feedback-based Estimation of Student’s Knowledge

Date: 21th April 2015
Experiment conductor: Veronika Štrbáková
Supervisor: Mária Bieliková
Short description: The characteristics of the learning object such as the Automated Readability Index metric and the LIX formula metric along with the observed student’s actions such as signals from eye camera, the display time of the learning object, the reading time of the learning object, mouse clicks, scrolling and pressing different keys can help us to estimate the level of user’s knowledge. By comparison of the evaluation of the acquired knowledge based on the student results from the pretest and post-test from the Functional and logical programming course with the time of the student’s active work spent on the learning objects we can find the dependence between the time of the active work with the learning object and the student’s knowledge. Based on this we estimate the level of his knowledge on concepts related to given learning objects.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Usability of Complex Visualization Software for Medical Purposes

Date: 30th April 2015
Experiment conductor: Peter Šutarík
Supervisor: Wanda Benešová (thesis), Jakub Šimko (lab)
Short description: The goal of the experiment is to evaluate interaction with medical software, especially regarding complex visualization of medical data. The study is conducted in two stages: first, the interaction is evaluated with lay user with no medical experience (in order to setup main experiment). The follow-up experiment will be conducted with medical personnel.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Player’s attention during puzzle solving

Date: 21th April 2015
Experiment conductor: Ján Stibila
Supervisor: Jozef Tvarožek
Short description: Purpose of this experiment is to verify hypothesis regarding how interesting, but unimportant at the time, elements of puzzle can divert players attention. For this we are going to measure players focus on specific elements of game board during game-play of the game 2048 and verifying this hypothesis by calculating how much players focus their attention on biggest number on the board, especially at the time, when there is nothing player can do with this number. We will also compare differences in focus between better players and players with worse results in the game to determine if there is correlation between focus on unimportant parts of game and game results.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Management of microtasks – user study

Date: 12th May 2015
Experiment conductor: Dušan Cymorek, Peter Gašpar, Slavomír Šárik
Supervisor: Michal Kompan
Short description: Every person’s behaviour usually differents as a consequence of personal, educational and other individual characteristics. This may be very crucial especially when using web applications which have become available to large crowds. In addition, their content is supposed to change dramatically and therefore it may be nearly impossible to identify possible problems. Our team has developed a new system for management of microtasks. We propose a new web portal to handle most of the aspects of the task management from creation to evaluation of microtasks. The main goal of our experiment is to reveal aspects of User Experience (UX) while using our web portal with possible daily users – students and researchers. Gathered results will be used to improve overall design and user interface of our product in order to become more enjoyable and usable.
Link to a formal description of the UX experiment: experiment description (In Slovak)

Visualization of large data volumes

Date: 11th May 2015
Experiment conductor: Andrej Štajer
Supervisor: Viera Rozinajová (thesis), Jakub Šimko (lab)
Short description: Visualization is one of the main parts of Visual analytics, hence it is not easy task to evaluate quality and usability. In our experiment we focus on user retrieving information and analyze usage of displayed visualizations. Visual analytics uses knowledge of visualization techniques and automated data analysis. Based on our energy domain dataset we examined different visualization techniques with well explaining ability.
Link to a formal description of the UX experiment: experiment description (In Slovak)