<![CDATA[The initial problem in entrepreneurship is that it is usually difficult for entrepreneurs to understand the demands and needs of potential consumers. Market uncertainty and changes in demand make entrepreneurs unable to predict market demand and needs. The purpose of this study is to conduct a robust evaluation of the collection of information obtained from microblogging data to estimate market variables. By designing a predictive system for entrepreneurial market interest through the use of the Twitter microblogging platform using the nave Bayes method, this system can generate market prediction models automatically through the use of popular microblogging data. The nave Bayes method is used to classify data for certain market interests. This prediction result is expected to be an effective way to improve performance and competitive advantage in entrepreneurship. The dataset is taken from one of the microblogging platforms, namely the twitter application with five keywords from the clothing business branch including pants, jackets, mukena, tunic and kebaya. The dataset taken is divided into two, training data and test data. The training data is used to make prediction models, while the test data are used as evaluation data for the design of the model. After testing, the accuracy of the system generated by the model is 87%. The model made using the nave Bayes method is categorized as very good at making predictions. The results of prediction accuracy can be seen in the form of popular news as a reference in providing products according to the demands and needs of potential consumers]]>
Copyrights © 2022
