Sentiment Analysis of Shoe Product Reviews on Indonesian E-Commerce Platform Using Lexicon Based and Support Vector Machine

The rapid development of e-commerce has encouraged people, especially young people, to switch from offline shopping to online platforms such as Shopee that offer fashion products, including shoes, at affordable prices and a wide selection. This phenomenon creates great opportunities for sellers, but also poses challenges related to analyzing product quality contained in customer reviews. The large number of scattered and unstructured reviews makes it difficult for potential buyers to accurately assess products. Therefore, this study aims to analyze the sentiment of 10,323 shoe product reviews on Shopee using the Support Vector Machine (SVM) algorithm and the Lexicon-Based method. SVM was chosen because of its advantage in achieving high accuracy in text classification, with accuracy results reaching 92.62%. The Lexicon-Based method is used to detect specific sentiment words, which provides deeper insight into consumer opinions on shoe products. The analysis results show that shoe product reviews are dominated by positive sentiments, reflecting a high level of customer satisfaction. The findings not only provide guidance for sellers in designing more effective marketing strategies, but also help potential buyers in making better decisions based on objective sentiment analysis. In addition, this study contributes to the literature related to sentiment analysis with SVM in the e-commerce domain, especially for fashion shoes. Thus, the combined use of SVM and Lexicon-Based methods shows great potential in providing valuable insights into consumer preferences as well as increasing customer confidence in choosing shoe products in the e-commerce.