Bacharelado em Sistemas de Informação (UAST)
URI permanente desta comunidadehttps://arandu.ufrpe.br/handle/123456789/2935
Siglas das Coleções:
APP - Artigo Publicado em Periódico
TAE - Trabalho Apresentado em Evento
TCC - Trabalho de Conclusão de Curso
Navegar
3 resultados
Resultados da Pesquisa
Item Aplicação de técnicas de mineração dedados educacionais para previsão do desempenho de estudantes a partir de dados do ENADE(2021-03-04) Lima, Thacyo Eslley Burgo de; Silva, Paulo Mello da; http://lattes.cnpq.br/8976586263200932; http://lattes.cnpq.br/7863661692692517With the increasing growth in data, it is necessary to apply techniques to extract information and knowledge from the set of data that are available, with that comes data mining. This work aims to predict the performance of students in Information System courses from data from the National Student Performance Exam (Enade) held in the year 2017. For this, machine learning algorithms were used to discover knowledge and assist in decision making. Four algorithms were used for comparison that obtained accuracy greater than 60%, showing that it is feasible to make the forecastItem Ferramenta adaptável para auxílio no processo de coleta de dados em pesquisas estruturadas(2019) Nascimento, Thiago Lucas Freire; Albuquerque, Hidelberg Oliveira; Albuquerque, Héldon José Oliveira; http://lattes.cnpq.br/3563291777221100; http://lattes.cnpq.br/0955255065576989; http://lattes.cnpq.br/1424806662151417Science is born of the unceasing search for knowledge starting from the problematization and questioning of historical events about the world and the reality in which we find ourselves. The existence of science depends on the development of studies which may or may not be associated with a hypothesis that needs to be explained and proven. In order to carry out such studies, it is necessary to obtain a large amount of data that may represent a given reality. Subsequently, specific methods and techniques are applied to this data for extracting information that aids in the discovery of results, revealing answers to help with the understanding of a proposed problem or raising new questions. However, doing this task in the traditional way, i.e. manually is considered as an ineffective and uncomfortable task due to several factors, one of them being the occurrence of errors due to the high human interaction in the activities. The process of computerized data collection consists of reducing the human efforts that result from the activities operated manually. This work aimed to develop a computational solution to assist researchers in the process of collecting, storing and processing data in different research areas and settings. Therefore, we conducted a literature review of the available technologies and methods that could be used for developing the solution as a means of meeting the requirements demanded by the proposed domain. This study proposed the development of a tool that allows the researcher to better organize and manage his research. Furthermore, it offers a greater convenience to the researcher by enabling the collection of data through a mobile application, performing this activity in the research setting without the need of internet connection, in a centralized, integrated and secure way. On that account, a Web application was designed allowing the researcher to make the necessary registrations to carry out a research by following the progress of the process and viewing the results through descriptive statistical metrics. The combined use of mobile and Web structures permits users to access information and perform activities anywhere, anytime. For the activity of elicitation and analysis of requirements, several meetings were held with the research group formed by an advisor and five students that took the role of stakeholders in that context. The information about the process and its procedures was obtained through interviews with targeted questions and the prototyping technique were used during meetings with the stakeholders. The purpose of the evaluation was to assess the interface in search of possible problems that might compromise the user interaction experience with the tool and propose feasible improvements. For that reason, a questionnaire was applied with questions related to the type of evaluation with the team of researchers who participated actively in the development process and thoroughly used the tool in the research setting. Based on the evaluation results, it was possible to identify that the tool has some elements that need to be repaired. Another aspect revealed by the evaluation was the level of user familiarity with the tool, considering that there was no difficulty in answering the evaluation questionnaires. In conclusion, the developed tool helps researchers to conduct different types of studies by simplifying and reducing the efforts made through the process done in the traditional manner, that is, manually.Item Classificação automática de discursos de ódio em textos do twitter(2019) Nascimento, Robson Murilo Ferreira do; Souza, Ellen Polliana Ramos; http://lattes.cnpq.br/6593918610781356; http://lattes.cnpq.br/8962852253787699Hate Speech can be defined as any communication that denigrates a person by their ethnicity, race, religion, sexual orientation, nationality or other characteristics. This behavior is becoming increasingly common in social networks, where many people confuse freedom of expression with intolerance. Young people are the main users affected since they represent a portion which might be easier to be influenced by the ideology propagated by haters, which in turn spread violence, racism, xenophobia, religious intolerance, etc. The tool that might help to handle this issue is Text Mining, which is capable of capture patterns or trends of texts in natural language. This task can be defined as a method of extracting relevant information in unstructured databases or semi-structured. Given that the Twitter is one of the most used social networks in Brazil, this work aims to implement and evaluate supervised machine learning techniques in order to automatically identify hate speech in tweets. With that in mind, we build a corpus with data collected from Twitter and part of it is manually annotated and subsequently preprocessed so we can obtain the ground truth collection used for training and evaluation of the supervised models. Finally, we conduct a comparison between machine learning algorithms, namely the SVM, Naive-Bayes and Logistic Regression. Later, we identify the best model under the described domain.