EUVIMEDEUVIMED entdecken
Crossref

Opinion Mining in Albanian Language: A Neural Network Model

Nelda Kote, Marenglen Biba

Albanian Journal of Natural and Technical Sciences · 30.04.2023 · 28 (1) · 53-62

DOI: 10.67537/ajnts2801005 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Albanian Journal of Natural and Technical Sciences, online erschienen am 30.04.2023. Verfasst von Nelda Kote und 1 weiteren Autor:innen. Die Arbeit nennt 21 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Opinion Mining (OM) has become an interesting research area applicable in different fields, such as social media posts, product or service reviews, online blogs, etc. OM aims to analyze and gain information from online media opinions. The most popular techniques used for opinion analysis are machine learning. The present paper aims to evaluate the performance of an artificial neural network model to classify the text opinion based on the polarity of the expressed sentiment using a two-classes annotation schema: positive and negative. The dataset of the opinions used is in the Albanian language. A CNN model is implemented and trained using the Keras framework and Tensorflow as a backend. The proposed classification model for opinions in the Albanian languages has an accuracy of 73%.

Zitieren

APA

Kote, N., & Biba, M. (2023). Opinion Mining in Albanian Language: A Neural Network Model. Albanian Journal of Natural and Technical Sciences, 28(1), 53-62. https://doi.org/10.67537/ajnts2801005

Vancouver

Kote N, Biba M. Opinion Mining in Albanian Language: A Neural Network Model. Albanian Journal of Natural and Technical Sciences. 2023;28(1):53-62. doi:10.67537/ajnts2801005

BibTeX
@article{Kote2023,
  author = {Kote, Nelda and Biba, Marenglen},
  title = {Opinion Mining in Albanian Language: A Neural Network Model},
  journal = {Albanian Journal of Natural and Technical Sciences},
  year = {2023},
  volume = {28},
  number = {1},
  pages = {53--62},
  doi = {10.67537/ajnts2801005}
}
Kostenlos starten →