<abstract xmlns="http://www.w3.org/1999/xhtml">

The provision of services by companies in specific domains requires continuous commitment and a constructive approach to meet customer requirements, however, it is often challenging to determine the level of customer satisfaction without their feedback. Therefore, this paper attempts to provide a solution to this problem by using comments from social networks and evaluating their sentiment using the VADER model.
In order to accomplish the aim of our research, a lexicon has been built with more than 9500 adjectives and verbs from the Albanian language based on VADER which is just in the initial form and the sentiments are evaluated as positive, neutral, or negative. The lexicon was constructed for the Albanian language and two companies of the Republic of Kosovo were researched as case studies.
Furthermore, the sentiment estimation, using the VADER model, in case of our datasets, we obtained a high accuracy, approximately between 89% and 95%. This level of accuracy has been primarily attributed to the application of all preprocessing steps within the dataset, which significantly enhances the model’s performance.
</abstract>

The provision of services by companies in specific domains requires continuous commitment and a constructive approach to meet customer requirements, however, it is often challenging to determine the level of customer satisfaction without their feedback. Therefore, this paper attempts to provide a solution to this problem by using comments from social networks and evaluating their sentiment using the VADER model.
In order to accomplish the aim of our research, a lexicon has been built with more than 9500 adjectives and verbs from the Albanian language based on VADER which is just in the initial form and the sentiments are evaluated as positive, neutral, or negative. The lexicon was constructed for the Albanian language and two companies of the Republic of Kosovo were researched as case studies.
Furthermore, the sentiment estimation, using the VADER model, in case of our datasets, we obtained a high accuracy, approximately between 89% and 95%. This level of accuracy has been primarily attributed to the application of all preprocessing steps within the dataset, which significantly enhances the model’s performance.

Sentiment Analysis Using the Vader Model for Assessing Company Services Based on Posts on Social Media

SEEU Review

This work is licensed under the Creative Commons Attribution 4.0 International License.

The provision of services by companies in specific domains requires continuous commitment and a constructive approach to meet customer requirements, however,...

Sentiment Analysis Using the Vader Model for Assessing Company Services Based on Posts on Social Media

Published Online: Dec 29, 2023

Page range: 19 - 33

DOI: https://doi.org/10.2478/seeur-2023-0043

Keywords
Sentiment, Vader model, social network, posts, sentiment evaluation

© 2023 Mërgim H. Hoti et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Sentiment Analysis Using the Vader Model for Assessing Company Services Based on Posts on Social Media

Published Online: Dec 29, 2023

Page range: 19 - 33

DOI: https://doi.org/10.2478/seeur-2023-0043

KeywordsSentiment, Vader model, social network, posts, sentiment evaluation

© 2023 Mërgim H. Hoti et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Keywords
Sentiment, Vader model, social network, posts, sentiment evaluation