<![CDATA[The trend regarding food and beverages in recent years has been widely discussed and has significantly impacted the activities of Small and Medium-sized Enterprises (SMEs) that sell food and beverages. Therefore, SMEs engaged in food and beverage sales should pay attention to public opinions regarding food and beverage trends, as well as observe the patterns of emerging food and beverage trends that are currently trending. To comprehend public sentiments concerning food and beverage trends, this study implements the Support Vector Machine (SVM) algorithm to assess its capacity in analyzing positive and negative sentiments within comments related to food and beverage trends on Twitter. The dataset employed consists of tweets created from the year 2018 to 2021, related to food and beverage trends. This dataset underwent several stages of processing, including preprocessing, data division into training and testing sets, and the application of weights using the TF-IDF method. Subsequently, the data is processed using the SVM algorithm. This study employs three ratios for data division: 90:10, 80:20, and 70:30 for training and testing. The most significant accuracy results for each dataset are as follows: 91.19% for "Es Kepal Milo" with a 90:10 ratio, 91.78% for "Baso Aci" with a 90:10 ratio, 87.98% for "Dalgona" with a 90:10 ratio, and 92.34% for "Corndog" with a 90:10 ratio. The implementation of the SVM algorithm yields high accuracy values, indicating that SVM is suitable for sentiment analysis of food and beverage trends. To understand the patterns of food and beverage trends, it can be inferred from the analysis conducted on food trends from 2018 to 2021 that the rise in food and beverage trends in Indonesia lasts for only one (1) to three (3) months before experiencing a sustained and drastic decline. Hence, SME stakeholders engaged in food and beverage sales need to prepare for such conditions.]]>
Copyrights © 2023
