Students’ Research Works – Spring 2016: Recommenders (PeWe.Rec)

Extended Abstract Template 

Personalized recommendation of TV program

Ondrej Čerman
master study, supervised by Mária Bieliková

Abstract. There are hundreds of TV stations broadcasting thousands of programs every day. Because of this large amounts of data, it is really hard for the viewers to find new good programs to watch. Sure, number of lists (like best rated programs, most viewed programs, etc.) can help some users, but they recommend content only on global preference. The solution is personalized recommendation. Our focus is to propose personalized recommendation for data from TV program mobile application developed in our faculty. Data from application are quite different compared to typical recommendation scenarios (most research are done on data from set-top boxes, video-on-demand servicies or on similar datasets), but we believe that despite of uniqueness of our dataset our recommendation will provide good results.

to the top | to the main | extended abstract
<!– | In Proc. of Spring 2016 PeWe Workshop, pp. 35-36 –>

Support of student’s activity in an e-learning system

Veronika Gondová
bachelor study, supervised by Mária Bieliková

Abstract. Motivation is one of the most important factors, which affect the power of a man. Motivation is a term that is often associated with the concept of games. The main objective of game is to keep attention and deployment a player. Gamification is the concept of applying game mechanics to motivate people in non-game contexts. Gamification currently found its application in the domain of education. In this thesis, we focus on supporting of student’s motivation to activity in a web education system. Motivation of students will be supported via the proposed concept of rooms that play the role of levels in games. The rooms are created by personalized recommendation which support game principle that levels should proceed from simpler to more complex. Student’s activity is rewarded using game principles. Implementation and evaluation is performed in e-learning system ALEF.

to the top | to the main | extended abstract
<!– | In Proc. of Spring 2016 PeWe Workshop, pp. 37-38 –>

Supporting Online Student Communities by Utilization of Questions and Answers Archives

Adrián Huňa
master study, supervised by Ivan Srba

Abstract. Community Question Answering (CQA) sites (e.g. StackOverflow, Yahoo! Answers) allow users to ask complex questions that cannot be understood by search engines, and get detailed answers from other users of the community. This model proved to be effective, and CQA sites currently represent an important part of World Wide Web. Their main purpose is to serve as a place to get answers but they also serve as archives of knowledge.

Recently, Massive Open Online Courses (MOOC) emerged as a new way to access quality education. The courses are of university-level quality, and are offered by world’s leading universities. Students in these online courses are provided with discussion forums, which serve as a place to exchange ideas, ask questions about course material, but also as a way to seek help from other students or course instructors. Because of this, we can see similar patterns in communication between conversations in the forums and knowledge exchange in CQA systems.

In our work we focus on supporting online student communities in MOOCs by utilization of question retrieval for automatic question answering. We are going to incorporate specifics of educational setting into our matching model between new question and archive of resolved questions. The main differences against general CQA systems that we identified are: tendency for periodical question repetition; natural expert users; and direct access to content that questions may be concerned with. If a good match for new question is found, we eliminate time needed for manual question answering, and also reduce the amount of questions course instructors need to check manually.

to the top | to the main | extended abstract
<!– | In Proc. of Spring 2016 PeWe Workshop, pp. 39-40 –>

Recommendation of Solved Questions from Archives in CQA Systems

Viktória Lovasová
master study, supervised by Ivan Srba

Abstract. Community Question Answering (CQA) sites such as Yahoo! Answers or Stack Overflow have become valuable platforms to create, share and seek a massive volume of human knowledge. To prevent information overload of users, we propose a method for personalized recommendation of already solved questions by personalized prediction of questions’ information value which is usually expressed by favouring these questions.

We evaluate the method by means of off-line experiments with dataset from CQA system Android Stack Exchange in two phases. In the first phase, we managed to classify the questions to the positive (user favoured the question) and negative group (user voted for closing the question) with 99% success. In the second part, we generated a newsletter of solved questions that can be interesting for the user where we reached 74% of success, i.e. the favoured question was present in the list of top 10 questions covered by the newsletter.

to the top | to the main | extended abstract
<!– | In Proc. of Spring 2016 PeWe Workshop, pp. 41-42 –>

Recommendation of New Questions in Online Student Communities

Jakub Mačina
master study, supervised by Ivan Srba

Abstract. Community question answering (CQA) systems are commonly used on the open Web and in enterprise environments. With an increasing popularity of Massive Open Online Courses (MOOCs) there is an opportunity for the CQA systems to help students in the online learning communities as well.

However, existing CQA systems have difficulties with an increasing proportion of questions that remain unanswered and with attracting helpers to address the posted problems. In the online student communities, it is even a bigger problem, as it can lead to students drop outs. Because of the specifics of the educational domain, our aim is to propose a new approach for a recommendation of new questions (question routing), which will be specifically designed for CQA systems employed in educational settings.

Our main objective is to provide a better educational-specific online recommendation in comparison to the general recommendation approaches. Many of the existing methods tend to overload only few experts by recommending most of the questions to them. We are going to focus more on answerer preferences by routing questions based on student expertise, interests and motivation. Furthermore, our method will route questions to the whole community by taking into account question difficulty, QA related and non-QA related data (e.g. course or assignment grades) of students. We plan to use so called knowledge gap phenomenon (i.e. expert users tend to ask more difficult questions while the opposite is true for less experienced users) for question difficulty estimation.

to the top | to the main | extended abstract
<!– | In Proc. of Spring 2016 PeWe Workshop, pp. 43-44 –>

Personal Computer Assistant for Supporting University Study

Matúš Salát
bachelor study, supervised by Jozef Tvarožek

Abstract. Abstract Learning at university can be troublesome for first-year students. Different teaching methods than what they are used to from high school, and the quantity of learning may force them to drop out. Students need to organize their deadlines, when they should start with learning for some midterm test or when they have deadlines for their project.

The many sources of information often make students misunderstand what teachers want from them. This leads often to forgetting the obligations and the follow-up failure. Students currently only seldom share useful tricks and hints on how to plan their time effectively for other classmates. Good time planning is the key for perspective and balanced learning.

In our work we want to create prediction model for preparation duration of events and personal recommendation for priority of student events. We collect different kind of inputs from student using web based application for evidence of semester events. After that we search for inputs that influence student preparation duration the most. These inputs are used in prediction of preparation duration. After evaluation of this model we will create recommender for recommendation of events with higher priority. Recommender will use student event rank and subject rank with other specific inputs that should be considered too.

to the top | to the main | extended abstract
<!– | In Proc. of Spring 2016 PeWe Workshop, pp. 45-46 –>

The Next Step of CQA Systems’ Utilization in Educational Domain: MOOCs

Ivan Srba
doctoral study, supervised by Maria Bielikova

Abstract. In our previous work, we proposed a concept of educational and organizational Community Question Answering systems (CQA) system which takes organizational specifics (e.g. closed community) as well as educational specifics (e.g. presence of a teacher) into consideration. In order to verify this concept, we implemented CQA system Askalot (demo of Askalot is available at https://askalot.fiit.stuba.sk/demo. Askalot is currently deployed at our faculty for the third year and it is used by a community which consists of about 1000 students and teachers.

We recognized that the potential of CQA systems in educational domain is significantly wider – especially, they can be used as a supplementary tool in MOOCs (Massive Open Online Courses). We started a cooperation with Harvard University in order to transform original design of Askalot into a plugin that can be used in MOOC system edX. The purpose of Askalot is to replace, the standard unstructured discussion forum with a more effective help seeking tool. At the present, this transformation of Askalot is already done and we plan to proceed with the next phase – deployment Askalot at several selected courses.

to the top | to the main | extended abstract
<!– | In Proc. of Spring 2016 PeWe Workshop, pp. 47-48 –>

Presentation of personalized recommendations via web

Martin Svrček
master study, supervised by Michal Kompan

Abstract. Nowadays, personalized recommendations are widely used and popular. There are a lot of systems in various fields, which use recommendations for different purposes. However, one of the basic problems is the distrust of users of recommendation systems. They consider them as intrusion of their privacy. Therefore, it is important to make recommendations transparent and understandable to users.

Our main goal is to propose several methods for presenting the results of the recommendations. On one hand we proposed approach to explain recommendations to end user. We created method of explaining recommendations without the need for knowledge of recommendation technique. This will allow us to explain each item differently by using information about user and his preferences. Therefore, this method represents personalized explanation of recommendation items, which also does not depend on the recommendation technique. In this context we focus on different approaches to obtaining data about users. These information allow us to provide suitable explanations. The evaluation of proposed approach is planned in the news domain in order to obtain statistically significant results using of implicit and/or explicit feedback.

to the top | to the main | extended abstract
<!– | In Proc. of Spring 2016 PeWe Workshop, pp. 49-50 –>