Scholarlink

In the contemporary digital landscape, forming and expressing opinions have become easy and streamlined. The current study tries to evaluate crowd cluster formations from tweets. The philosophy of micro-blogging has enabled Twitter users to express opinions and highlight them by linking them to hashtags. This study presents a comprehensive methodology for analyzing Twitter data obtained from five prominent hashtags. Leveraging three sentence embedding techniques which are Universal Sentence Encoders, distilBERT, and Sentence Transformer, this research investigates the formation of clusters within the dataset, employing the K-means clustering algorithm. The resulting clusters are rigorously assessed using an array of metrics, and the clusters formed are visualized. The methodology for evaluation used in this work is not seen to have been explained in detail elsewhere.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Framework to Evaluate Twitter Tweets Using K-Means Clustering

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Vineeth Menon Munnoorpillil,
  • J. V. Bibal Benifa

摘要

In the contemporary digital landscape, forming and expressing opinions have become easy and streamlined. The current study tries to evaluate crowd cluster formations from tweets. The philosophy of micro-blogging has enabled Twitter users to express opinions and highlight them by linking them to hashtags. This study presents a comprehensive methodology for analyzing Twitter data obtained from five prominent hashtags. Leveraging three sentence embedding techniques which are Universal Sentence Encoders, distilBERT, and Sentence Transformer, this research investigates the formation of clusters within the dataset, employing the K-means clustering algorithm. The resulting clusters are rigorously assessed using an array of metrics, and the clusters formed are visualized. The methodology for evaluation used in this work is not seen to have been explained in detail elsewhere.