Sauki, Mocmd (2024) Analisis Sentimen Terhadap Cafe Di Pekanbaru Dengan Metode Support Vector Machine (SVM). Other thesis, Universitas Islam Riau.

Text Mocmd Sauki.pdf - Submitted Version Restricted to Registered users only Download (5MB) | Request a copy

Abstract

Coffee shops or cafes are one of the most popular for the community, especially among young people, even coffee culture has become a new lifestyle for them. This study aims to analyze the sentiment towards cafes in Pekanbaru using the Support Vector Machine and Logistic Regression methods. Thus, it is expected to understand opinions or views on the community systematically and objectively. This research was carried out through methodological stages such as Collect Data obtained from an online platform, namely Google Maps by using the Beautifulsoup library for the data crawling process with a total of 5217 data. Data preprocessing: cleaning and filtering data to remove unnecessary words, such as removing numbers, hashtags, punctuation marks and spacing words that do not have spaces. Labeling with the Lexicon InSet process which makes words classified as positive and negative. Applying the algorithm used, then this study carried out a data splitting process with several sharing ratios, namely 90:10, 80:20, 70:30 and 60:40. Evaluate the model by performing the Confusion Matrix calculation process, namely accuracy, precision, recall and f-1 score. The results of the evaluation of 2 methods, namely Support Vector Machine and Logistic regression, where svm has good performance with a division ratio of 90:10 with an accuracy result of 93% while Logistic Regression with a division ratio of 90:10 gets an accuracy result of 90%. So of the 2 methods used with a difference of 3%.

Actions (login required)

View Item