%T  International Conference on Human-Computer Interaction - RoCHI 2020
%C  Sibiu
%D  22-23 October 2020
%G  ISSN 2501-9422
%E  Teodor Stefanut
%E  Jean Vanderdonckt
%E  Alex Butean
%P  168
%W http://rochi2020.utcluj.ro

%T User Tasks Description: a Retrospective, Recent Contributions and some Research Challenges
%A Philippe Palanque
%P 1-4
%X Describing users tasks has been the focus of research for many years starting with the seminal work from Annett and Duncan in 1967 [2]. Since then, the Human Factors and Human Computer Interaction domains have proposed multiple contributions identifying the elements that have to be gathered and represented in order to describe precisely the relevant aspects of users tasks. This keynote will highlight these fundamental elements of tasks descriptions and will state the current state of the art. A specific view on how to use such descriptions to design and assess automation will be given. Some publicly available tools will also be presented together with their use in various industrial application domains. These applications will be the opportunity to identify remaining research challenges for tasks description and modeling.
%K Task descriptions; task modeling; task-centered design; task-centered evaluation.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Palanque.pdf

%T Steps forward towards developing preschoolers' digital literacy. An experience report
%A Anamaria Moldovan
%A Adriana-Mihaela Guran
%A Grigoreta-Sofia Cojocar
%P 5-10
%X Early childhood education (ECE) encompasses, at its most basic level, all forms of education, both formal and informal, provided to young children up to approximately 8 years of age. Some of the benefits include a diminished risk of social-emotional mental health problems and increased self-sufficiency as children mature and enter adulthood. It is during this period that children go through the most rapid phase of growth and development. Their brains develop faster than at any other point in their lives, so these years are critical. The foundations for their social skills, self-esteem, perception of the world and moral outlook are established during these years, as well as the development of cognitive skills. Lately, digital literacy and its practices are added as a need for future personal and professional development of children and developing digital competence already at kindergarten level can also help to raise awareness on safety issues and build critical thinking among them regarding content and devices that they use. This paper describes a new approach in supporting the development of digital literacy skills of preschoolers aged 5 to 6 in the context of Human-Computer Interaction (HCI) students volunteering for ICT activities in kindergarten along with building edutainment applications. There is also a report of key digital competences established by the kindergarten teachers involved, and monitored for formation during the activities.
%K Education; Digital literacy; Interaction; Human; Preschoolers; Key Competences; ICT activities
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Moldovan.pdf

%T The Lib2Life Platform - Processing, Indexing and Semantic Search for Old Romanian Documents
%A Irina Mitocaru
%A Gabriel Guțu-Robu
%A Melania Nițu
%A Mihai  Dascălu
%A Ștefan Trăușan-Matu
%A Silvia Tomescu
%A Gabriela Florescu
%P 11-18
%X Preserving the cultural heritage of a nation throughout generations is essential in a continuously developing society. This paper introduces the Lib2Life platform powered by advanced Natural Language Processing techniques, focusing on the processing, indexing and semantic search of old documents from the Central University Libraries in Romania. Our platform enables the upload and text pre- processing of scanned documents by librarians, who can afterwards manually correct the extracted content and corresponding metadata. In addition, Lib2Life ensures the exploration of the collection of books using a semantic search engine to retrieve documents fitted to the users’ interests. The platform was evaluated using an usability questionnaire which pinpointed out that Lib2Life is a modern and user-friendly smart search engine for old documents written in the Romanian language. Improvements in terms of server response time and functionality were suggested. The platform proved to be intuitive and easy to use, having the potential to become an analytical system incorporating a rich and diverse collection of books.
%K Library software; automated text extraction; Natural Language Processing.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Mitocaru.pdf

%T Emerging Patterns in Romanian Literature and Interactive Visualizations based on the "General Dictionary of Romanian Literature"
%A Irina Toma
%A Laurentiu-Marian Neagu
%A Mihai  Dascălu
%A Ștefan Trăușan-Matu
%A Laurențiu Hanganu
%A Eugen Simion
%P 19-26
%X The General Dictionary of Romanian Literature (DGLR) is a comprehensive work carried out by researchers from the literary institute of the Romanian Academy. DGLR offers detailed information about writers, editors, translators, literary publications, and cultural institutions that contributed to the Romanian national literature. The current work presents interactive web visualizations, based on statistical studies performed on the DGLR corpus, such as: biographical information; geographic information, including countries that part of the most important writers have visited, lived in, or studied in; active literary entities per year; timeline of publications for important writers. The purpose of the visualizations is to provide overviews regarding the Romanian literature to the general audience. In addition, our views offer valuable insights on the writers and their work across time. A survey was conducted on 20 users and most of them had a pleasant experience; recommendations on future developments were also provided.
%K Analytical approach; Quantitative study; DGLR; Romanian writer; ReaderBench framework; Interactive visualizations.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Toma.pdf

%T Violence Detection in Images Using Deep Neural Networks
%A Edwin-Mark Grigore
%P 27-30
%X Software and technology has evolved and expanded so much over the last decades, that is present in everybody’s life in every little aspect, and more and more significantly at children’s disposal. Starting from this reality, it is necessary the identification of the images that contain scenes of violence or emotionally disturbing scenes, images that contain blood or depict human bodies with open wounds, violent fires, or presence of guns and weapons. Machine learning (ML) is capable of extracting features from images and learn to identify the images that depict inappropriate scenes for children, using different techniques. With the recent advances in deep learning, traditional ML methods, such as Support Vector Machines, have been surpassed by deep neural networks that are also employed by our solution for violence detection.
%K Computer vision; violence; violence detection; neural network; deep learning.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Grigore.pdf

%T Automatic detection of cyberbullying on social media platforms
%A Ștefăniță Stan
%A Traian Eugen Rebedea
%P 31-38
%X The presence of cyberbullying on the Internet has grown alarmingly in recent years. Teenagers and children are the most affected by this phenomenon that is often the cause of higher suicide rates and social isolation. The detection and prevention of cyberbullying depends firstly on its correct understanding and secondly on the correct selection of a classification model trained on features that have a high discrimination factor between cyberbullying and non- cyberbullying. In this paper, we aim to create an automatic detection model for cyberbullying posts that is not biased towards a specific social media platform or a certain type of bullying. We describe the method we used for selecting the best features for two different classifiers trained on datasets collected from Twitter and Formspring. Next, we explain how we use the predictions made by these classifiers for labelling a new dataset collected by us from Twitter. The results of the automatic classification of the dataset have been compared to the manual classification of a sample of data from it, resulting in a rate of agreement larger than 50% between automatic detection and human annotation.
%K Cyberbullying detection; Natural language processing; Social media; Online Aggressivity.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Stan.pdf

%T Stroke Detector - An Application that applies the F.A.S.T. Test to Identify Stroke
%A Mihnea-Ioan Rezmeriță
%A Irina-Elena Cercel
%A Adrian Iftene
%P 39-47
%X In Romania, stroke is the second leading cause of death and disability (Neagu, 2019). The speed of reaction in the case of a stroke can make the difference between life and death, between a healthy patient and a patient who remains disabled for life. In this context, we wanted to create a mobile application in Romanian that will help users to do a F.A.S.T. test very quickly. The test identifies possible problems at the face, arms and speech levels. This article shows how we made this application using a client-server architecture and what users think about it.
%K F.A.S.T., Client-server architecture, Stroke, Face detection, Speech recognition.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Rezmerita.pdf

%T Intrinsic motivation and motives for Facebook use - a formative measurement approach
%A Costin Pribeanu
%P 48-53
%X For over a decade, Facebook is part of the everyday life of university students. Its growing popularity stimulated the research aiming to understand the motivations that are driving people to join and use social networking websites. To answer this research question various approaches have been taken: technology acceptance model (TAM), uses and gratification theory, the theory of consumption values, and multidimensional models. The objective of this paper is to analyze the relationship between the intrinsic motivation of using Facebook and the motives for its use. Intrinsic motivation for using Facebook has been conceptualized as perceived enjoyment. The motives for Facebook use have been conceptualized as a formatively-measured construct that impacts two drivers of the intention of use: the perceived enjoyment and the perceived ease of use. The results show that four motives for using Facebook are predicting Facebook's perceived enjoyment: keeping in touch with known people, entertainment, finding useful information and resources, and socialization.
%K Motives for Facebook use, hedonic technologies, MIMIC models, perceived enjoyment, perceived ease of use, formative measurement.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Pribeanu.pdf

%T Targeted Romanian Online News in a Mobile Application Using AI
%A Marius-Cristian Buzea
%A Ștefan Trăușan-Matu
%A Traian Eugen Rebedea
%P 54-60
%X Nowadays, being informed is a very important aspect of our life. Few online news broadcasters, such as bbc.com or cnn.com, send real-time mobile notifications with several interactivity facilities with the application to be informed with the latest news. In this paper, we present an application based on Machine Learning (ML), a basic technique used in Artificial Intelligence (AI), available for mobile devices, which detects irony, the articles’ polarity and the fake news. The mobile application is based on a machine learning system using manually annotated Romanian corpora, and a crawler network designed to automatically extract data from more than 730 public websites. The aim of the application is to collect information regarding user behavior, to change and adapt the content to user’s preferences and requirements in order to process and better understand the user’s necessities. Furthermore, the user is able to search through the news, filter the news, create favorites lists and visualize graphical representations for different time periods. Therefore, the main goal of the application is to develop and improve the way that people receive and access reliable information.
%K Adaptive Mobile News Platform; Artificial Intelligence; Machine Learning; Romanian Corpora; Graphical Representations;
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Buzea.pdf

%T Analysis of the time spent on Facebook by Romanian university students
%A Valentina Iuliana Manea
%A Costin Pribeanu
%P 61-64
%X The use of Facebook by university students is a hot topic of research for more than ten years. However, few studies exist that are questioning the minutes spent daily for various tasks, such as socialization, posting/reading information, entertainment, and creation/update of the personal profile. Also, there are few studies investigating group differences as regards the various aspects of Facebook usage. This paper aims to analyze the differences by gender and year of study as regards the general characteristics of Facebook usage, motives for Facebook use, and time spent on Facebook for various activities. The results show that female students have larger Facebook networks and spend more time daily, especially for reading, posting, and sharing information. After the first year of study, students spend less time on entertainment and more time for editing the personal profile.
%K Social networking websites, Facebook, university students.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Manea.pdf

%T JIST: Java Interaction Separation Toolkit
%A Adrian-Radu Macocian
%A Dorian Gorgan
%P 65-72
%X A GUI toolkit is a library consisting of the elements needed for developing interactive applications. In the 2000s, a lot of effort was devoted to building platforms that enabled the creation of rich internet applications. This let the field of desktop applications underdeveloped. The Java Interaction Separation Toolkit (JIST) was developed with the intention of having a lightweight, cross-platform and support for a declarative UI. By using the WORA aspect of Java most of the desktop platforms are covered. It features XML support for describing user interfaces in a more natural way than the classic wall of text associated with the native code-behind approach.
%K Java; GUI sub-system; Graphical User Interface; Interactive Applications; Markup Language; Functional and Interaction Separation; GUI toolkit.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Macocian.pdf

%T Discover the Wonderful World of Plants with the Help of Smart Devices
%A Cosmin Irimia
%A Mihai Costandache
%A Andrei Mădălin Matei
%A Matei Lipan
%A Ștefan Romanescu
%A Adrian Iftene
%P 73-80
%X “Augmented Reality in Botanical Garden” (ARGB) is the name of the application we developed, which brings visitors closer to the Botanical Garden “Anastasie Fătu” from Iași. Even if they want to search for a plant to know where they can find it or maybe they are interested in knowing the garden and its sections, visitors can just access the app and get all the help they need. More than this, the app provides enhanced features for the people responsible for the maintenance of the garden such as the possibility to generate placards for different plants during different plant exhibitions.
%K Android; iOS; Mobile app; Web app; Maps; Orientation and movement outside buildings.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Irimia.pdf

%T Secure management and integration system for electrical devices
%A Daniel Florin Bîrsoan
%A Teodor Ștefănuț
%P 81-86
%X The main purpose of an Internet of Things(IoT) network is to make people's work and life easier by providing processes and services as close as possible to their needs. Globally, it is stated that the Internet of Things (IoT) must be available everywhere. As the Internet is almost ubiquitous today, this is not an unreasonable requirement. But to create such a network, it would be necessary for all devices, regardless of the date of creation or manufacturer to be able to be inter-connected to a common platform and made accessible securely through the Internet. In the current article we are proposing an architecture that responds to this need for the interconnectivity of devices and facilitates secure communication through its components. Through the installation of dedicated board/boards in the desired space and through the connection of the electrical devices on different pins to them, the “objects” are connected to the internet and the user can control their ON/OFF state remotely. So this purpose the proposed architecture features four main components: (1) on-site boards that control the electrical items and the internet connection; (2) a server that orchestrates the communication between all the other components; (3) a web application for electrical items management; (4) a smartwatch application for electrical items control.
%K System Security; Modular Software System; Architecture in Software Application, Web Application, Smartwatch Application;
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Birsoan.pdf

%T RASA Conversational Agent in Romanian for Predefined Microworlds
%A Bianca Nenciu
%A Dragoș Georgian Corlatescu
%A Mihai  Dascălu
%P 87-94
%X Technology is becoming omnipresent in our lives due to its accessibility and ease of use. Conversational agents facilitate interactions in natural language and are frequently employed to perform repetitive tasks in a specific context. We introduce a conversational agent for Romanian built on top of the open-source RASA framework, capable to communicate in predefined microworlds. Two scenarios were considered, namely: a smart home assistant which interprets commands to IoT devices, and an interactive info- point for our university focusing on providing guidance to students. Several enhancements were considered, including an NLP pre-processing pipeline from spaCy and a knowledge graph implemented using Grakn for conceptualizing the information accessible to the agent. Our agent can quickly classify intents and extract entities with high accuracy for a given microworld (F1-score of 97% for the first microworld and 93% for the second). A survey on 10 users showed high satisfaction in terms of the usefulness and the succinctness of the provided information.
%K Conversational agent; Natural Language Understanding; Romanian language; Microworlds.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Nenciu.pdf

%T Conversational Agent in Romanian for Storing User Information in a Knowledge Graph
%A Gabriel Boroghina
%A Dragoș Georgian Corlatescu
%A Mihai  Dascălu
%P 95-102
%X Information surrounds us and keeping track of relevant details can be challenging. Although there are multiple applications to take notes, organize ideas, or set reminders, existing solutions are semantic-agnostic and rely on the user to manually search for desired information by keywords. We propose a novel method to help people store and retrieve such details with ease in Romanian language. Our conversational agent built on top of the RASA framework is capable to extract relevant information from the user’s utterances, store them in a persistent knowledge graph, and ultimately, access them when requested. A set of specific intents regarding locations, timestamps, and properties were created and learned by the agent using manually built examples. In addition, an interaction logic based on a knowledge graph was added to enable the storage and retrieval of information, based on the identified semantic components from the input sentences. The performed tests showed a good accuracy for intent detection, and promising results for the sentence parser. Our conversational agent is accessible as a web application which can process text or speech inputs, and responds with a textual representation of the user’s memorized facts.
%K Natural Language Processing; Conversational agent; Knowledge representation.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Boroghina.pdf

%T Seeking an Empathy-abled Conversational Agent
%A Andreea Grosuleac
%A Ștefania Budulan
%A Traian Eugen Rebedea
%P 103-107
%X A fairly novel area of research, at the conjunction between Artificial Intelligence (AI) and Human-Computer Interaction (HCI), resides in developing conversational agents as more users prefer this type of interaction to conventional interfaces. In this paper, we present an open-domain empathic chatbot, encompassing two of the biggest challenges of dialog systems: understanding emotions and offering appropriate responses. Although these tasks are trivial for a human, it is difficult to create a system that can recognize others’ feelings in a discussion. The proposed model is developed based on the Generative Pre-Trained Transformer and it uses two datasets, PersonaChat and Empathetic Dialogues to achieve an empathic chatbot with a cordial personality. The measured performance - 18.20 perplexity, 6.56 BLEU score, and 6.56 accuracy - comes close to the state-of-the-art models, while offering a further refined dialogue persona.
%K Empathic dialogue; Open-domain conversational agents; Empathetic chatbot; Conversational interfaces.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Grosuleac.pdf

%T Analysis of convergence and divergence in chat conversations
%A Liviu-Andrei Niță
%A Ștefan Trăușan-Matu
%A Traian Eugen Rebedea
%P 108-114
%X A discussion between several participants is often accompanied by an exchange of information between speakers and, in the case of collaborative learning, a polyphonic interanimation is desired. They have, according to the polyphonic model, different points of view that are convergent or divergent by which one of the participants approves or disapproves of another person that is participating in the discussion. This paper presents an approach for identifying convergent or divergent utterances using neural networks and other machine learning methods in order to help people analyze a dialogue conducted in an online environment. Especially in collaborative chats used in education, this solution allows professors to identify how the participants in the discussion exchange information and how the debate evolved in time.
%K Dialogue, chat, neural networks, machine learning, convergent, divergent, polyphony.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Nita.pdf

%T Controlling a programming environment through a voice based virtual assistant
%A Sonia Grigor
%A Constantin Ioan Nandra
%A Dorian Gorgan
%P 115-122
%X The use of smart personal assistants is intended to provide a solution that employs the voice interaction model in order to help improve accessibility for everyday tasks. This model is more than suitable for simple tasks, such as internet searches or device-controlling commands. In this paper we explore the possibility of using this interaction model for completing more complex, composite, context-dependent tasks. Particularly, we look into the potential benefits of using custom spoken commands to help novice users develop insight into the workings of a computer program. Throughout this paper, we present a solution based on an existing, customizable voice assistant that is meant to both help users grasp the structure of a program and improve accessibility for programming activities. The latter is achieved by providing a framework for a voice- based programming environment, offering features like code fragment insertion, navigation, error detection, handling and program running, while also providing voice and text-based feedback for the executed commands.
%K Voice-based interaction; Programming assistant; Amazon Web Services; Echo Dot; Alexa.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Grigor.pdf

%T Increasing Diversity with Deep Reinforcement Learning for Chatbots
%A Cristian Pavel
%A Ștefania Budulan
%A Traian Eugen Rebedea
%P 123-128
%X Dialogue generation for open-domain conversations is a difficult and open problem that, so far, has not been able to approach human-level performance. Recently, a popular solution is to apply a sequence-to-sequence architecture, similar to the machine translation problem. These models try to map the input - given as the previous utterances, to the output - the next utterance. Unfortunately, they usually tend to repeat sentences, often preferring dull responses, that end the conversation abruptly. Therefore, Reinforcement Learning techniques have been combined with the standard sequence-to-sequence models in order to avoid their shortcomings. Our model applies a Policy Gradient method that maximizes the expected reward of generating the next utterance given a history of previous utterances. The results show an improvement in diversity up to 0.16 - almost 10x higher than the model without RL, while keeping the responses relevant to the input message.
%K Dialogue generation; Reinforcement learning; Conversational agents; Sequence-to-sequence model.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Pavel.pdf

%T Design features of a VR software system for personnel training in aviation
%A Andrei Bulai
%A Diana Andronache
%A Dorin-Mircea Popovici
%P 129-136
%X This paper presents the steps that must be taken to develop a Simulator type software using VR technology and the presentation of our own software system for the training aviation personnel. We present the state of the art of VR and describe the developed application through the chosen solution, meeting all the criteria of a learning and testing environment for the aviation personnel, simulating the targeting operations of a helicopter and adjacent elements. In fact, we make a description of both the data structure and the technical design. Logical design and system architecture, use cases and conceptual classes are also presented. The technical part describes in detail the implementation stages of the application.
%K Virtual reality; Virtual Simulation; Software development; Triggers and Gestures; User Interface for virtual reality; User Experience for virtual reality; 3D animation and modeling
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Bulai.pdf

%T Usability Testing of Mobile Augmented Reality Applications for Cultural Heritage - A Systematic Literature Review
%A Diana Tiriteu
%A Silviu Vert
%P 137-144
%X This paper presents an overview of the existing literature to date regarding the usability testing of Mobile Augmented Reality applications developed for cultural heritage. A Systematic Literature Review was conducted with this purpose. Four databases were interrogated with specific keywords. Our of the 88 found research papers, only 31 met the eligibility criteria to be included in this survey. Four major usability testing methods were identified: questionnaires, focus groups, user testing and interviews. The questionnaire method was used in almost all of the studies, and only a few of them combined more usability testing methods. Nearly all papers targeted the outdoors use centered on location-based augmented reality. The perceived ease of use, usefulness and enjoyment were confirmed by a majority of the papers. The ease of use was also correlated to the ease of learning, as well as with the interest of using the application. Cultural heritage is one of the domains that can benefit greatly from the latest improvements in mobile augmented reality technologies.
%K Mobile augmented reality; usability; user experience; questionnaire; focus group; user testing; interview; cultural heritage.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Tiriteu.pdf

%T Interactive Assembly Simulation in an Immersive Virtual Environment
%A Cătălin Moldovan
%A Adrian Sabou
%P 145-152
%X This paper describes the development of a Virtual Reality (VR) simulation application for educational purposes in assembly production. People are able to retain more information by simulating real experiences. Nowadays, modern technology can be used to achieve better results in a virtual reality learning environment, being available to everyone. It is predicted to be more common and low-priced in the future. Interactive user gestures and immersive VR technologies are used to develop remote solutions for engineering students by matching the product components in the proper order and location. This research provides instructional assembly methods and natural experience in interacting with elements, both in a dynamic space or in a sitting position.
%K Virtual Reality; Leap Motion; Assembly Simulation; OpenGL; Educational Application.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Moldovan-2.pdf

%T Exploring the main factors driving a satisfactory use of the Moodle platform
%A Elena Ancuța Santi
%A Gabriel Gorghiu
%A Costin Pribeanu
%P 153-156
%X E-learning platforms are largely used in educational institutions around the world, in hybrid, mixed or blended learning formats. The power of those platforms has been proved extensively in the period of the global pandemic being used as main frameworks for offering and achieving educational services. This work is an exploratory study aiming to analyze the contribution of the perceived ease of use, usefulness, and enjoyment to the satisfaction of using a Moodle e-learning platform. To do this, a structural modeling approach has been taken. The model testing results show that perceived enjoyment was the most important factor which suggests that, apart from being usable and useful, a learning platform should be also attractive and interesting.
%K e-learning, Moodle, TAM, usability, perceived enjoyment, user satisfaction.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Santi.pdf

%T Student Testing Activity Dataset from Data Structures Course
%A Paul Ștefan Popescu
%A Marian Cristian Mihăescu
%A Oana Maria Teodorescu
%A Mihai Mocanu
%P 157-164
%X E-learning platforms became more and more popular not only for distance learning but also for learning in full-time education. As this popularity grows, we can use the data extracted from them to complement the professor's work and make predictions regarding students’ performance. In this paper, we present a dataset extracted from our e-Learning platform, which is based on the logs collected from testing activity. The focus of this paper is to present the dataset; the experiments presented in the paper are meant to explore the dataset along with its capabilities. The dataset consists of attributes relevant to the testing activity and provides labels which consist of average test grade and final exam grade. Our focus when building the dataset was to keep only the attributes relevant for the learning activity and to provide means to analyse and predict the student's final grade or failure. The paper presents the structure of the dataset, the methodology of collecting the data and experiments using several popular algorithms. The experimental results reveal that the actions performed by the users correlate with the results of the tests and the exam failure can be predicted with a pretty good accuracy using the default set of tuning parameters for our algorithms. As feature work, we can extend the set of experiments with other algorithms, and we can also use parameter tuning for each algorithm for a slight increase in performance.
%K Machine Learning, Educational Dataset, e-Learning, User Modelling.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Popescu.pdf

%T Exploring age, gender and area differences of teachers as regards mobile teaching
%A Gabriel Gorghiu
%A Elena Ancuța Santi
%A Costin Pribeanu
%P 165-168
%X Introducing mobile teaching and learning in schools represents a real challenge in nowadays education. Mobile devices are able to support a diversity of tasks such as communication, interaction, collaboration, information, and resource sharing. Although mobile devices are widely used for e-learning, but also very present in non-formal and informal learning, there are many barriers against the adoption of mobile technology by teachers, for teaching purposes. This paper analyzes the perceptions of Romanian teachers as regards the use of mobile technology in the teaching and learning process. The purpose of the paper is to explore the influence of gender differences, age, and area (urban vs. rural) on the perceptions of mobile teaching and learning. The results illustrate an important influence of such variables when considering the integration of mobile technology in education, as well as of a variable that has to be taken into consideration seriously: the teacher’s attitude concerning the use of technology.
%K Mobile devices, mobile learning, mobile teaching, group differences.
%W http://rochi.utcluj.ro/articole/8/RoCHI2020-Gorghiu.pdf