Efficient Hashtag Recommender System For Digital Marketing

The advent of social media and digital marketing has revolutionized advertising by enabling advertisers to monitor rivals and provide more visible campaign outcomes. Machine learning (ML) can estimate future developments and aid decision-making, transforming strategic decision-making for businesses. Twitter, a powerful networking platform, has evolved into a social media platform. This paper proposes an on-demand hashtag recommender framework that helps users discover relevant hashtags. Using TF_IDF (Frequency - inverse document frequently), LSA (latent Semantic Analysis), and LDA(Linear Discriminant Analysis) techniques, the research proposes an efficient hashtag recommender system to help make precise business strategy decisions. The Passive-aggressive classifier is the best classifier used, and the performance evaluation demonstrates its effectiveness, improving the Jaccard Similarity of identifying critical tweets by up to 64.31% for the Aggressive classifier and 63.7% for Linear SVC.