Sentiments Analysis on Public Land Transport Infrastructure in Davao Region using Machine Learning Algorithms: A Recent Study

Land transport infrastructure has always been an important aspect of a city. People increasingly use social media to express their feelings about city projects such as land transportation. Government organisations are having trouble detecting concerns that develop as a result of people utilising social media to criticise land transportation infrastructure. These text-based social media posts can be examined using sentiment analysis, which is a key task in Natural Language Processing (NLP). Sentiment analysis is a process of the deriving sentiment of a particular statement or sentence. This study used a social networking website to create a model of sentiments on land transportation infrastructure in Region XI (Davao Region), the Philippines, and tested the model's correctness using a data set. There are 1,200 text data sets in total, divided into two categories: test dataset (25 percent) and training dataset (75 percent). To perform sentiments analysis of text data sets, machine learning text classifiers such as Support Vector Machines (SVM), Random Forest (RF), and Multinomial Nave Bayesian (MNB) were utilised. The performance of each classification model is estimated by generating a confusion metric with precision and recall calculations, known as the f1-score. The accuracy rating was calculated as well. A comparison was also made based on the results of the three machine learning classifiers' tests. SVM has the highest accuracy, with 76.12 percent and a f1-score of 71.98 percent, based on the results of the experiment. This research will be used to inform and support policy-making and development of land transportation infrastructure in the Davao Region.