Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorSalamó Llorente, Maria-
dc.contributor.authorRodríguez Queraltó, Jordi-
dc.descriptionTreballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2015, Director: Maria Salamó Llorenetca
dc.description.abstractIntelligent Tutoring Systems are systems based on Artificial Intelligence that can analyze models of knowledge to get to understand the learner’s state. Its function is trying to help him or her through emulating the role of a human tutor. In this project we’ll implement the Pedagogical Module of an Intelligent Tutoring System. Specifically, we’ll implement a prediction system that tries to estimate the grades students will have at the end of the academic year. For this we’ll incorporate methods of Machine Learning. Machine Learning is a discipline from the field of Mathematics and Computer Science that studies algorithms to predict future outcomes and act accordingly, and, if applicable, implement an automated behaviour that tries to aim for the best possible result. Specifically, the algorithms we’ll use are called classifiers, a kind of Supervised Learning algorithm. We’ll implement a library with different kinds of classifiers, test their functionality using Ten-Fold Cross-Validation and simulate its usage with different datasets that come from the UCI repository and from 4 subjects that are taught in grades at the University of Barcelona, to analyze their accuracy and efficiency. Furthermore, this library will need to be usable by teachers. Using it, they will be able to input their history from previous years and their current course to get an estimate of that year’s results. By doing this, they will be able to detect a deviation from previous courses and work on getting it back on track. The implementation will have the options to either go for the most theoretically precise classifier or use the pre-set option for a fast execution. The most precise option will be determined by testing every single classifier using the train set, while the pre-set classifier will be chosen based on this project’s tests and the algorithm’s consistency. These functions will work through a user interface, which will be used by teachers from different fields of the University of Barcelona. This means the interface will need to be user-friendly and easy to
dc.format.extent50 p.-
dc.rightsmemòria: cc-by-nc-sa (c) Jordi Rodríguez Queraltó, 2015-
dc.rightscodi: GPL (c) Jordi Rodríguez Queraltó, 2015-
dc.subject.classificationIntel·ligència artificialcat
dc.subject.classificationEnsenyament assistit per ordinadorcat
dc.subject.classificationTreballs de fi de graucat
dc.subject.classificationTutoria (Ensenyament)ca
dc.subject.otherArtificial intelligenceeng
dc.subject.otherComputer-assisted instructioneng
dc.subject.otherComputer softwareeng
dc.subject.otherBachelor's theseseng
dc.subject.otherTutoring (Teaching)en
dc.titleMòdul pedagògic en un sistema tutor intel·ligent per a predir l'evolució de l'alumnatca
Appears in Collections:Programari - Treballs de l'alumnat
Treballs Finals de Grau (TFG) - Enginyeria Informàtica

Files in This Item:
File Description SizeFormat 
memoria.pdfMemòria1.13 MBAdobe PDFView/Open
codi_font.zipCodi font8.23 MBzipView/Open

This item is licensed under a Creative Commons License Creative Commons