A city's land transportation infrastructure has always been critical. People are increasingly turning to social media to share their opinions on civic projects like land transit. Government agencies are having difficulty recognising concerns that arise as a result of people criticising land transportation infrastructure on social media. Sentiment analysis, a crucial task in Natural Language Processing, can be used to analyse these text-based social media messages (NLP). Sentiment analysis is the process of determining a statement's or sentence's sentiment. This study employed a social networking website to develop a model of sentiments on land transportation infrastructure in Region XI (Davao Region), Philippines, and then used a data set to verify the model's accuracy. There are a total of 1,200 text data sets, separated into two categories: test dataset (25 percent) and training dataset (75 percent) (75 percent). Machine learning text classifiers such as Support Vector Machines (SVM), Random Forest (RF), and Multinomial Nave Bayesian (MNB) were used to do sentiment analysis on text data sets. The f1-score is used to calculate the performance of each classification model by generating a confusion metric with precision and recall calculations. The accuracy rating was also calculated. The outcomes of the testing of the three machine learning classifiers were also compared. Based on the findings of the experiment, SVM has the highest accuracy, with 76.12 percent and a f1-score of 71.98 percent. The findings of this study will be used to inform and promote policy development and land transportation infrastructure development in the Davao Region.
Author(s) Details
Mark Van M. Buladaco
Institute of Computing, Davao del Norte State College, Philippines
Jumar S. Buladaco
Institute of Computing, Davao del Norte State College, Philippines
Laarni M. Cantero
Institute of Computing, Davao del Norte State College, Philippines
No comments:
Post a Comment