Machine Learning Approach To Forecast the Word in Social Media
Rajanala Vijaya Prakash · 2022
Forecasting is one of machine learning's most significant features. In machine learning, forecasting issues are classified as supervised learning algorithms. Label or target data process the given text to achieve a specific degree of accuracy or precision. The objective of time series analysis is to build models that best capture or explain the data to figure out what is driving a time series' fundamental causes. In this paper, words are forecasted on Twitter data with 1830 tweets. These tweets are interpreted as weighted documents using the term frequency and inverse document frequency (TF-IDF) algorithm. After obtaining the tweet's word frequency value, forecasting is done using the test data, and the findings accurately referred to 1303 slack word categories and 541 verb tweets as training data. Inactive users and active users were split into two categories when it came to word forecasting. The data were then processed using a MAPE calculation procedure that was set at 50% for inactive users and 20% for active users.