Online Shopping Preferences Visualization System Using Web Content Mining

Recently, social media sites have been attracting people to start and operate online business. It has been gaining more attention especially during this COVID-19 pandemic, where most activities are conducted via online platform. For instance, Instagram is one of the social media platform where people share and post their pictures and videos on their account, but some people take this opportunity to promote their business and shop online via features provided by Instagram. Based on observation, there is no specific function that visualizes the information of Instagram's business accounts to help people in gaining information and making decisions during shopping process. Thus, a system that specifically visualizes Instagram's extracted data was developed using Web Content Mining technique with the assistance of hashtags as data-points and Phantombuster API. Tableau software integrated with JavaScript library is used as a visualization tool to display the users shopping activities. The result of the extracted data was visualized in the form of Bubble Chart, Bar Chart, and Multiple Bar Chart that were placed in different dashboards. The results from the obtained functional testing and users' feedbacks have indicated 64.3% of the respondents agreed that the proposed system helps users in making the right choice during online shopping. However, this system has some limitation that could be further enhanced such as updating the information automatically in real time without using any manpower and to increase the interactivity between system and user instead of using static visualization graphs.