Sentiment-analysis-per-texts-in-Albanian-tongue

The purpose of this topic is to analyze the sentiment of the text in Albanian tongue through deep learning. Deep learning techniques have been used and successfully tested to analyze the text in Albanian tongue. The texts that have been analyzed are comments on the posts from the IKSHPK Facebook page during a period of almost 2 weeks in pandemic time when the flow of activity on this page was large enough to collect the material. From the collected comments, a dataset with Albanian language texts has been created, which has been used to train the machine learning model, which analyzes the sentiment and classifies it based on the assessment of the sentiment as positive, negative or neutral. Machine learning was implemented through the use of artificial neural networks, respectively RNN using the practice of another kind of this network that is LSTM. This study was conducted at the Faculty of Electrical and Computer Engineering as a Bachelor thesis.