Abstract
User engagement is an important metric to improve the connection with the users. Previous studies have analysed the utility of different Twitter parameters that increase user engagement. However, their appropriateness for user engagement has been analysed using a small number of distinct Twitter parameters and on different data sets. In addition, the effects of users’ replies on the tweets of the companies have not been analysed. In the current research, we analyse the tweets of Fortune 100 companies to assess the unified effects of parameters such as the usage of hashtags, images, mentions, emojis, text length and the sentiment of the tweet on user engagement. Using empirical analysis, we first propose an augmented engagement formula to accommodate the sentiment score of the users’ replies. Further, we perform multiple linear regression analysis to assess the effects of different parameters on the revised engagement score. We observe that the tweets with length less than 140 characters have a higher user engagement activity. Results on tweet wise analysis showed that the use of emojis while replying positively affects user engagement. In addition, the sector-wise analysis revealed that the presence of image in tweets has a positive influence on the engagement score for 66% of the sectors, followed by the hashtags, emojis, and the tweet length with 60%, 33%, and 40% sector coverage, respectively. Finally, the study provides significant recommendations that can be employed to increase user engagement.
Student Collaborator
: Tanya Kapur (IIIT-Delhi), Anvit Sachadev (IIIT-Delhi)