| dc.description.abstract | This study aims to analyze Twitter user behavior patterns in an effort to increase user awareness, as well as identify parameters that are significant to these patterns. User awareness is measured using a metric attribute provided by the Twitter API, namely impression_count. The data used in this study is taken from tweets that have keywords related to food and beverages. The analysis of user activeness is done by looking at the pattern of tweeting based on time, while the analysis of user awareness is done by using the Random Forest method to determine the parameters that have the most influence on impression_count. The results show that, when viewed from the time of creation, the most tweet activity occurs at 16:00, 15:00, and 10:00 according to the time zone of the detected location. When viewed from the day of creation, the peak activity occurred on weekends, namely Saturday and Sunday. In the aspect of user awareness measured through impression_count, the most significant parameters include the number of likes, retweets, citations, replies, number of followers, and number of tweets made. Accounts with many followers and a high frequency of tweets tend to have greater user awareness. By understanding these factors, strategies can be designed to more effectively increase user engagement and awareness on the Twitter platform. | en_US |
