|
Expiry date recognition using deep neural networks Vlad Florea, Traian Rebedea |
1-17 |
|
Interactive language learning - How to explore complex environments using natural language? Tatiana-Andreea Petrache, Traian Rebedea, Ştefan Trăuşan-Matu |
18-31 |
|
Mobile application to improve reading habits using Virtual Reality Estefany Chavez-Helaconde, Israel Pancca-Mamani, Julio Vera-Sancho, Betsy Cisneros-Chavez, Wilber Valdez-Aguilar |
32-47 |
|
NLP based Deep Learning Approach for Plagiarism Detection Răzvan Roșu, Alexandru Ștefan Stoica, Paul Stefan Popescu, Marian Cristian Mihăescu |
48-60 |
1 University Politehnica of Bucharest
313 Splaiul Independentei, Bucharest, Romania
2 Open Gov SRL
95 Blvd. Alexandru Ioan Cuza, Bucharest, Romania
Abstract. This paper proposes a deep learning solution for optical character recognition, specifically tuned to detect expiration dates that are printed on the packaging of food items. This method can be used to reduce food waste, having a significant impact on the design of smart refrigerators and can prove especially useful for persons with vision difficulties, by combining it with a speech synthesis engine. The main problem in designing an efficient solution for expiry date recognition is the lack of a large enough dataset to train deep neural networks. To tackle this issue, we propose to use an additional dataset composed of synthetically generated images. Both the synthetic and real image datasets are detailed in the paper and we show that the proposed method offers a 9.4% accuracy improvement over using real images alone.
Keywords: Expiry date recognition, synthetic ocr dataset, deep learning, computer vision, vision impaired.
Cite this paper as:
Florea, V., Rebedea, T. Expiry date recognition using deep neural networks.
International Journal of User-System Interaction 13(1),
1-17, 2020.
1 University Politehnica of Bucharest
313 Splaiul Independentei, Bucharest, Romania
2 Open Gov SRL
95 Blvd. Alexandru Ioan Cuza, Bucharest, Romania
3 Institutul de Cercetări în Inteligenţa Artificială
Calea 13 Septembrie nr. 13, Bucureşti
4 Academy of Romanian Scientists
Splaiul Indpendentei 54, Bucharest, Romania
Abstract. Implicit knowledge about the physical world we live in is gained almost effortlessly through interaction with the environment. In the same manner, this knowledge cannot be simply inferred from language, as humans normally avoid stating what is trivially implied or observed in the world. This paper is about a novel perspective into progressing artificial intelligence toward understanding the true language meaning through interaction with complex environments. The arising field of text-based games seems to hold the key for such an endeavour. Text-based games placed in a reinforcement learning formalism have the potential of being a strategic path into advancing real-world natural language applications - the human world itself is one of partial understanding through communication and acting on the world using language. We present a comparative study highlighting the importance of having a unified approach in the area of learning agents to play families of text-based games, with the scope of establishing a benchmark that will enable the community to advance the state of the art. To this end, we will look at the corpora and the first two winner solutions from the competition launched by Microsoft Research - FirstTextWorld Problems. The games from the proposed corpora share the same objective, cooking a meal after collecting ingredients from a modern house environment, having the layout and the recipes change from one game to another.
Keywords: Reinforcement learning, natural language processing, text-based games, partial understanding, language meaning.
Cite this paper as:
Petrache, T.-A., Rebedea, T., Trăuşan-Matu, S. Interactive language learning - How to explore complex environments using natural language?.
International Journal of User-System Interaction 13(1),
18-31, 2020.
Universidad Nacional de San AgustÃn de Arequipa
Calle Santa Catalina Nro. 117, Cercado - Arequipa, Peru
Abstract. Education is changing rapidly, so emerging technologies are being used to im-prove this process. One of these technologies is the Virtual Reality (VR), whose field of action is increasingly broad, so it has been incorporating new methods in teaching having a great positive impact in recent years, but the applicability in the area of communication is minimal. Seeing the challenges faced by the Ministry of Education in the development of reading skills and strengthening the capabilities of students in our country. In the this research a new alternative is proposed, to improve the beginnings of the habit of reading in students of second grade of Elementary School having to use of a mobile application with VR named Diverticuentos that generates scenes of the readings in 360o besides being con-nected to a databases like firebase that allows us to see the progress of each stu-dent verifying that it is possible to integrate this new technology in the sessions and to generate a beginning of habit of reading of the students.
Keywords: Virtual reality, Reading habits, M-Learning, Unity.
Cite this paper as:
Chavez-Helaconde, E., Pancca-Mamani, I., Vera-Sancho, J., Cisneros-Chavez, B., Valdez-Aguilar, W. Mobile application to improve reading habits using Virtual Reality.
International Journal of User-System Interaction 13(1),
32-47, 2020.
University of Craiova, Department of Computers and Information Technology
Blvd. Decebal nr. 107, RO-200440, Craiova, Romania
Abstract. Plagiarism detection represents an application domain for the NLP research area, which has not been investigated too much by researchers in the context of lately developed attention mechanism and sentence transformers. In this paper, we present a plagiarism detection approach which uses state-of-the-art deep learning techniques in order to provide more accurate results than classical plagiarism detection techniques. This approach goes beyond classical word searching and matching, which is time-consuming and can be easily cheated because it uses attention mechanisms and aims for text encoding and contextualization. In order to get proper insight regarding the system, we investigate three approaches in order to be sure that the results are relevant and well-validated. The experimental results show that the systems that use BERT pre-trained model offers the best results and outperforms GloVe and RoBERTa.
Keywords: BERT, RoBERTa, sentence transformers, plagiarism, NLP.
Cite this paper as:
RoÈ™u, R., Stoica, A. È., Popescu, P. S., Mihăescu, M. C. NLP based Deep Learning Approach for Plagiarism Detection.
International Journal of User-System Interaction 13(1),
48-60, 2020.