Presented in Latin American Conference on Learning Objects and Technologies
Electronic books (e-books) are an alternative to traditional books and have some advantages over it. For example, they present dictionary function, allow searching for words in texts, are ecologically correct, among others. Moreover, research indicates e-books can improve the learning process. The learning process is also improved with the use of enhanced e-books, which includes multimedia resources (e.g., images, videos, sounds) to stimulate adapted and interactive learning. Similarly, multisen-sorial media (MulSeMedia) research presents the possibility of expanding the senses with additional media components, like haptic or gustatory devices, which can enhance the learner quality of experience. In this scenario, this work proposes a new model to combine enhanced e-books and MulSeMedia to improve the learning process. This model is named Multisensorial Electronic Books, or Multisensorial Books, for short. A prototype with auditory, haptic and olfactory effects is also created. This initial prototype showed promising results, indicating the present study may open avenues for further researches.
Congratulations to Rian Tavares, undergraduate student in Computer Science at CEFET/RJ. He has an accepted paper in the Brazilian Workshop on Social Network Analysis and Mining (BraSNAM) 2017. The article’s title is “Classificação de filmes: uma abordagem utilizando o LIWC”. Its main goal is to analyze texts from movies subtitles to classify them into two categories: bad or excellent. The full article will be online shortly.
This article aims to present an approach to classify movies based on their subtitles and information extracted from social networks. The methodology developed uses LIWC program, which contains a dictionary of words that allows extracting linguistic, psychological and social characteristics of texts. Preliminary results were very satisfactory, indicating promising directions for this study.
Congratulations to Rafael Guimarães and Wladimir Pereira, postgraduate students in Computer Science at CEFET/RJ. They have an accepted paper in the Brazilian Workshop on Social Network Analysis and Mining (BraSNAM) 2017. The article’s title is “Inferência de idade utilizando o LIWC: identificando potenciais predadores sexuais”. Its main goal is to analyze Brazilian Portuguese texts from social networks to detect potential sexual predators. The full article will be online shortly.
Social predators use the Internet to exploit children or adolescents with abusive or sexual purposes. More and more these predators use social networks to access their victims, often providing fake profiles pretending to be adolescents. In this scenario, this study aims to analyze texts in Brazilian Portuguese to infer the age of the users. For this purpose, we used a tool named LIWC in its Brazilian Portuguese version. As a case of study, a Brazilian social network was used to perform experiments. This study focused on the analysis of texts written by adolescents and men between 25 and 45 years old, which represent the great majority of sexual predators. Results were relevant and open lacks for further studies.