Abstract

Amount of data the exponential growth we have seen in social-media, but it is not widely used to be something of value, one of which is tweets on twitter. This paper describes the result of using engine web crawler with pre-processing text mining method. The use of engine web crawler aims to take tweets via API Twitter as unstructured text data and then represented into web form. The use of pre-processing method aims to filter out tweets on three stages: cleansing, case folding and parsing. Applications designed in this research using the waterfall model and implemented with PHP. Testing method in this research using the black box testing to check whether the design result can already be run in accordance between expectations or not. Result from this research are in the form of applications that can help prepare tweets into data that can be processed according to needs based on keywords and the search date desire by the user. This is done because of several related research shown that data on social media, especially Twitter is currently the destination company or agency to understand public opinion.

Authors and Affiliations