No abstract

Land transport infrastructure has been a vital part of a city. People nowadays use social media to post their sentiments towards developments in a city such as land transport. Government agencies have difficulty in identifying issues that arises from the people using social media towards land transport infrastructure. These social media posts in the form of a text can be analyzed using sentiments analysis, which is a significant task of Natural Language Processing (NLP). This research experimented on creating a model of sentiments on land transport infrastructure in Region XI (Davao Region), the Philippines from the social media website, and test a data set on the accuracy of the model. There are a total of 1,200 text data sets, and it's divided into two: test dataset is 25%, and the training dataset is 75%. The machine learning text classifiers used are Support Vector Machines (SVM), Random Forest (RF) and Multinomial Naïve Bayesian (MNB) to process sentiments analysis of the text data sets. The performance of each classification model is estimated by generating confusion metric with the calculation of precision and recall, the f1-score. The accuracy rating was also computed. A comparison was also conducted based on the results of experiments of the three machine learning classifiers. Based on the resulting experiment, SVM has the highest accuracy, with 76.12% and a f1-score of 71.98%. This research will be utilized as support and notes for policy-making and development for land transport infrastructure in the Davao Region.

No abstract

No abstract

No abstract