Sentiment Analysis Using MnB

Oct 13, 2025 by recoveryshake.com 106 views

Analysis of Sentiment Based on Aspects of Garuda Indonesia Airlines Through Social Media Twitter Using the Latent Latent Algorithm of Allocation and Multinomial Naïve Bayes

Introduction

The aviation industry has experienced significant growth in recent years, with air transportation becoming the preferred mode of travel for many people in Indonesia. Among the various airlines operating in the country, Garuda Indonesia is one of the most popular and considered options. However, with the increasing number of airlines and passengers, the quality of service and facilities has become a major concern for many travelers. Social media platforms, particularly Twitter, have become a popular medium for passengers to express their opinions and experiences about airlines. However, the analysis of these opinions has been limited, and the aspects discussed by the community have not been thoroughly explored. Therefore, this study aims to analyze the sentiment of tweets related to Garuda Indonesia using the Latent Dirichlet Allocation (LDA) and Multinomial Naïve Bayes (MNB) algorithms.

Research Methodology

The data used in this study consisted of 1750 tweets related to Garuda Indonesia. The analysis process began with data grouping using the LDA algorithm, which successfully identified three main aspects that are often discussed, namely facilities, services, and schedules. After that, the data was divided into two sets: 1400 tweets for training data and 350 tweets for testing data, with a ratio of 80:20.

Before the analysis process was carried out, the data underwent a preprocessing stage, which included cleaning data, changing letters to small (case folding), normalization, tokenisation, filtering, and stemming. This process aimed to prepare the data to make it easier for the algorithm to be used.

Sentiment Analysis Using MnB

After the data was processed, the Multinomial Naïve Bayes (MNB) algorithm was applied to classify the sentiments of the analyzed tweet. MNB is one of the effective methods in text analysis and is very suitable for dealing with categorical data such as sentiment. In this study, the tweet that had been classified would be grouped into three categories of sentiment: positive, negative, and neutral.

The system evaluation results were carried out using a confusion matrix, which showed an accuracy of 80%. This means that the model used was quite reliable in identifying sentiments from tweets related to Garuda Indonesia.

Aspect-Based Sentiment Analysis

The aspect-based sentiment analysis of tweets about Garuda Indonesia revealed that the community had a variety of opinions about three main aspects: facilities, services, and schedules. The analysis showed that the majority of tweets were related to services, followed by facilities and schedules. This finding suggests that passengers are more concerned about the quality of services provided by Garuda Indonesia.

Conclusion

Through an aspect-based sentiment analysis of tweets about Garuda Indonesia, it was found that the community had a variety of opinions about three main aspects: facilities, services, and schedules. With this understanding, the airline can take strategic steps to improve services and meet customer expectations.

By using the LDA and MNB algorithms, this research provides a clearer picture of the public perception of Garuda Indonesia through social media. This finding can be a foundation for better policy development and improve the quality of airline services. Therefore, analysis like this is very important for airlines to understand and respond to the needs and desires of its passengers.

Implications of the Study

The findings of this study have several implications for Garuda Indonesia and the aviation industry as a whole. Firstly, the study highlights the importance of understanding customer needs and expectations through social media analysis. Secondly, the study suggests that airlines should focus on improving their services to meet customer expectations. Finally, the study provides a framework for airlines to analyze customer sentiment and make data-driven decisions.

Limitations of the Study

This study has several limitations. Firstly, the study only analyzed tweets related to Garuda Indonesia and did not consider other airlines. Secondly, the study only used two algorithms, LDA and MNB, and did not explore other sentiment analysis techniques. Finally, the study only analyzed a limited number of tweets and did not consider other social media platforms.

Future Research Directions

This study provides several avenues for future research. Firstly, future studies can explore other airlines and social media platforms to gain a more comprehensive understanding of customer sentiment. Secondly, future studies can use other sentiment analysis techniques, such as deep learning algorithms, to improve the accuracy of sentiment analysis. Finally, future studies can explore the impact of sentiment analysis on airline policy development and service improvement.

Recommendations for Airlines

Based on the findings of this study, airlines can take several steps to improve their services and meet customer expectations. Firstly, airlines should focus on improving their services to meet customer expectations. Secondly, airlines should use social media analysis to understand customer needs and expectations. Finally, airlines should make data-driven decisions based on sentiment analysis to improve their services.

Conclusion

In conclusion, this study provides a comprehensive analysis of sentiment based on aspects of Garuda Indonesia Airlines through social media Twitter using the Latent Dirichlet Allocation (LDA) and Multinomial Naïve Bayes (MNB) algorithms. The study highlights the importance of understanding customer needs and expectations through social media analysis and suggests that airlines should focus on improving their services to meet customer expectations. The study provides a framework for airlines to analyze customer sentiment and make data-driven decisions.
Q&A: Sentiment Analysis of Garuda Indonesia Airlines Through Social Media Twitter

Introduction

In our previous article, we discussed the analysis of sentiment based on aspects of Garuda Indonesia Airlines through social media Twitter using the Latent Dirichlet Allocation (LDA) and Multinomial Naïve Bayes (MNB) algorithms. In this article, we will answer some of the frequently asked questions (FAQs) related to the study.

Q: What is sentiment analysis, and why is it important for airlines?

A: Sentiment analysis is a technique used to determine the emotional tone or attitude of a piece of text, such as a tweet. It is important for airlines because it helps them understand customer needs and expectations, which can inform their service improvement strategies.

Q: What are the benefits of using social media analysis for airlines?

A: Social media analysis provides airlines with real-time feedback from customers, which can help them identify areas for improvement and make data-driven decisions. It also allows airlines to track customer sentiment over time, which can help them identify trends and patterns.

Q: What are the limitations of the study?

A: The study only analyzed tweets related to Garuda Indonesia and did not consider other airlines. It also only used two algorithms, LDA and MNB, and did not explore other sentiment analysis techniques. Finally, the study only analyzed a limited number of tweets and did not consider other social media platforms.

Q: How can airlines use the findings of this study to improve their services?

A: Airlines can use the findings of this study to identify areas for improvement and make data-driven decisions. For example, if the study found that customers are unhappy with the airline's services, the airline can use this information to develop strategies to improve those services.

Q: What are the implications of the study for the aviation industry as a whole?

A: The study highlights the importance of understanding customer needs and expectations through social media analysis. It also suggests that airlines should focus on improving their services to meet customer expectations.

Q: Can you explain the LDA and MNB algorithms used in the study?

A: The LDA algorithm is a type of topic modeling algorithm that is used to identify underlying themes or topics in a dataset. The MNB algorithm is a type of machine learning algorithm that is used to classify text as positive, negative, or neutral.

Q: How can airlines use sentiment analysis to track customer sentiment over time?

A: Airlines can use sentiment analysis to track customer sentiment over time by analyzing a large dataset of tweets and identifying trends and patterns. They can also use this information to develop strategies to improve their services and meet customer expectations.

Q: What are some potential challenges or limitations of using sentiment analysis for airlines?

A: Some potential challenges or limitations of using sentiment analysis for airlines include the need for large datasets, the complexity of the algorithms used, and the potential for bias in the data.

Q: How can airlines ensure that their sentiment analysis is accurate and reliable?

A: Airlines can ensure that their sentiment analysis is accurate and reliable by using high-quality data, developing robust algorithms, and testing their results against a variety of metrics.

Conclusion

In conclusion, this Q&A article provides answers to some of the frequently asked questions related to the sentiment analysis of Garuda Indonesia Airlines through social media Twitter. We hope that this article has provided valuable insights and information for airlines and other stakeholders in the aviation industry.