Title: Sentiment analysis of positive and negative comments, extracted from social networks and web in Albanian language

Authors: Mërgim H. Hoti; Hamdi Hoti; Edisona Kurhasku

Addresses: Faculty of Electrical and Computer Engineering, University 'Hasan PRISHTINA' Prishtinë, Kosovo ' Faculty of Economic, University 'Ukshin HOTI' Prizren, Kosovo ' Faculty of Economic, University 'Hasan PRISHTINA' Prishtinë, Kosovo

Abstract: Nowadays, massive texts that contain a variety of viewpoints, attitudes, and emotions for products and services are generated on the web. Sentiment analysis is one of the fastest-growing research areas in computer science, making it difficult to keep track of all activities in the area, including data we made in social networks, the web as well as communicating with each other, and so on. This paper explains sentiment analysis of positive and negative comments taken from social networks and news sites of low resources languages such as Albanian. These comments are taken from different topics and are analysed with a support vector machine, using Sklearn and other algorithms. As we know, social networks and online communication involve both positive and negative relationships between each other's. This research measures and shows how people are prone to express positive or negative opinions about any important case. Being aware of these problems, the edge sign prediction problem that aims to predict whether an interaction between a pair of nodes will be positive or negative should be considered. Therefore, this study provides theoretical results regarding this problem that motivates natural improvements.

Keywords: sentiment analysis; predictable data; Albanian language.

DOI: 10.1504/IJASS.2024.140018

International Journal of Applied Systemic Studies, 2024 Vol.11 No.2, pp.83 - 96

Accepted: 20 Feb 2024
Published online: 15 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article