Tabular data are abundant. Most machine learning/data science applications require input data to be in tabular format. Yet, most learning algorithms are not explainable by default, e.g., neural networks. Predictions of OWL Class Expression learners are inherently explainable as an expression is a first-order logical expression.
In this thesis, the student will focus on techniques that map a tabular data into RDF Knowledge Graph. Thereafter, a classification problem can be tackled by an OWL expression learner. The student will closely work on owlapy and ontolearn.
In case you have further questions, feel free to contact Caglar Demir.