The effect of the TF-IDF algorithm in times series in forecasting word on social media

Forecasting is one of the main topics in data mining or machine learning which forecasting, a group used, has label class target. Thus, many algorithms for solving forecasting problems are categorized as supervised with aim conducting training. In this case, things that were target playing role 'supervisor' who supervise training process achieving certain level accuracy precision. Time series method generally used to forecast based on time and can words social media. study had conducted word twitter 1734 tweets interpreted weighted documents using TF-IDF algorithm frequency often comes out so value getting smaller vice versa. After weight tweets, was performed test results referred 1203 categories Slack 531 verb resulting good accuracy. The division classified into two groups i.e. inactive users active users. obtained processed MAPE calculation 50% 0.1980198%