|
Mobile Sentiment Analysis by Deep Learning Image Processing Cristiana-Adriana Scăunașu, Paul Stefan Popescu, Marian Cristian Mihăescu |
121-135 |
|
Ethics in Artificial Intelligence Ştefan Trăuşan-Matu |
136-148 |
|
Continuum – a virtual musical instrument Diana Andronache, Andrei Bulai, Dorin-Mircea Popovici |
149-168 |
|
Specific group differences in assessing the usefulness of mobile teaching Elena Ancuța Santi, Gabriel Gorghiu, Costin Pribeanu |
169-181 |
University of Craiova, Department of Computers and Information Technology
Blvd. Decebal nr. 107, RO-200440, Craiova, Romania
Abstract. Once with the growth of free data sources which makes available more data for training machine learning and deep learning algorithms, new challenges tend to appear. Usage of deep learning algorithms for solving image classification problems became one of the most popular approaches because of its flexibility and generally good results. Sentiment analysis is a problem which was previously addressed using several approaches and for a different type of data which makes it a challenging problem. This paper presents a sentiment analysis solution for images gathered on mobile devices. Our approach uses deep learning for model training and part of the system validation and a mobile (android) application for model usage and real-time image classification. We considered the possibility to perform parameters tuning in order to obtain better performance, and we present a comparison between trained models. The experiments conducted with the trained models revealed excellent performance on the mobile device with excellent accuracy real-life tests.
Keywords: Deep learning, sentiment analysis, android application
Cite this paper as:
Scăunașu, C.-A., Popescu, P. S., Mihăescu, M. C. Mobile Sentiment Analysis by Deep Learning Image Processing.
International Journal of User-System Interaction 13(3),
121-135, 2020.
1 University Politehnica of Bucharest
313 Splaiul Independentei, Bucharest, Romania
2 Institutul de Cercetări în Inteligenţa Artificială
Calea 13 Septembrie nr. 13, Bucureşti
3 Academy of Romanian Scientists
Splaiul Indpendentei 54, Bucharest, Romania
Abstract. Artificial intelligence (AI) is becoming nowadays a daily presence in our lives, in several ways, many times not noticeable even for informed people. In addition to the evident benefits of AI, there were unfortunately reported cases of people that had problems due to biased decisions provided by AI applications. Therefore, enquires about ethical issues and trustworthiness of AI are very important to be considered. In this sense, the paper presents an analysis of the ethical aspects that should be taken into account by the developers of AI applications. Regulations for an ethical and trustworthy AI, which were proposed by the European Union are introduced. Ways of developing ethical AI applications are discussed.
Keywords: ethics; artificial intelligence; machine learning; deep learning; neural networks; human-AI interaction
Cite this paper as:
Trăuşan-Matu, S. Ethics in Artificial Intelligence.
International Journal of User-System Interaction 13(3),
136-148, 2020.
Ovidius University Constanta
124 Mamaia Blvd., Constanta, Romania
Abstract. The aim of this paper is to present a possible development in the area of virtual musical instruments. Our solution consists of creating a version of a Laser Harp, a virtual musical instrument that uses laser beams to replace the strings of a real harp. The best current version of Laser Harp uses a Head-Mounted Display (HMD) with controllers that must be held by the user as they play the instrument. This is a common occurrence when it comes to virtual musical instruments. This research paper presents a way to allow users to play the virtual harp with their bare hands, without the need for controllers. In what follows, we will present the steps that were taken in building such an instrument that we named Continuum and the results that we achieved during the process.
Keywords: Virtual Reality, Leap Motion, virtual musical instrument, Laser Harp, audio system.
Cite this paper as:
Andronache, D., Bulai, A., Popovici, D.-M. Continuum – a virtual musical instrument.
International Journal of User-System Interaction 13(3),
149-168, 2020.
1 Valahia University Targoviste
13 Aleea Sinaia, 130004, Targoviste, Romania
2 Academy of Romanian Scientists
Splaiul Indpendentei 54, Bucharest, Romania
Abstract. Introducing mobile teaching in school represents a challenge for teachers that have to adapt their teaching practice to benefit from the technological changes. Although mobile devices are widely used by teachers and students in everyday life, there are several barriers against the adoption of this technology for teaching purposes. There are also differences in the perception of mobile teaching usefulness that depend on various factors, such as age, gender, discipline, and qualification level. This paper aims to analyze the teachers’ perception as regards the benefits of and barriers against mobile teaching, from a domain-specific perspective. The differences are analyzed concerning the discipline they teach, the educational level of the class (form), and the teachers’ qualification (didactical degree). The analysis highlights the differences along three dimensions: learning motivation, learning usefulness, and teaching usefulness. Overall, the results show a positive attitude towards mobile teaching and learning.
Keywords: Mobile devices, mobile learning, mobile teaching, group differences.
Cite this paper as:
Santi, E. A., Gorghiu, G., Pribeanu, C. Specific group differences in assessing the usefulness of mobile teaching.
International Journal of User-System Interaction 13(3),
169-181, 2020.