The rapid development of the new industrial estate in Majalengka requires the support of a good image from the local community. This study aims to assess public sentiment towards factories in West Java as a basis for formulating strategies to improve the image of industry in Majalengka. The Support Vector Machine (SVM) method was applied to classify the sentiment of YouTube comments relating to the factory, after a process of data collection, text preprocessing, and lexicon-based sentiment labelling. The main findings indicated that people's sentiments were distributed in proportions of approximately 30% negative, 35% neutral, and 35% positive. The accuracy of the SVM model was recorded at 78.48%, while the confusion matrix indicated adequate classification performance of the sentiments. matrix indicates adequate classification performance for negative, neutral, and positive sentiments. negative, neutral, and positive sentiments. This finding indicates that there is a significant negative sentiment towards the significant negative sentiment towards the industry, although it is not dominant. In conclusion, sentiment analysis conducted through social media can serve as a basis for formulating public relations and Corporate Social Responsibility (CSR) strategies. Corporate Social Responsibility (CSR) strategies in an effort to improve the industry's image. The industry and the government are advised to improve clear communication and responsive CSR programmes, with the aim of reducing negative sentiment and strengthening public trust. strengthen public trust.