<![CDATA[The growth of online economic transaction is experiencing a significant increase recently. One of the massive transactions carried out is through social media platforms. To reach more potential customers, several social media platforms offer paid ad serving services. In utilizing this service, business decision makers often need a decision support system that is currently rarely examined. This research focuses on building a decision support system on how business decision makers can carry out efficient paid advertising campaigns. Two machine learning algorithms are tested and compared in performance to get a robust algorithm to classify the types of posts that are able to reach more potential customers and have more interaction. The result shows that Random Forest is able to achieve an accuracy up to 75% which is better than Support vector machines which only reach 66% accuracy. In addition, Paid ads were found to be less relevant in reaching more potential customers and increase the number of interactions. To provide a guidance in implementing an efficient paid advertising campaign in Facebook, a guidance or decision support system is compiled based on the results of an independent variable weighting.Keyword: social media advertisement, random forest, support vector machine, data mining ]]>
Copyrights © 2020
