Identification and Recommendation of Authorities on Different Topics Based on Twitter

Roque Daniel Contreras Chacon, Jack Fernando Bravo Torres, Jennifer Andrea Yepez Alulema, Diego Andres Cuji Dutan, Paul Esteban Vintimilla Tapia

Research output: Contribution to journalArticle


The social platform Twitter is one of the most famous world microblogs. Monthly, it hosts hundreds of millions of visitors who publisha large amount of data that is consumed by others. Some of these publications help to others to generate knowledge. Users who generate content in a particular area, helping people to increase their knowledge, are classified as authorities in that field. This content could be exploited by students to improve their academic performance. In this paper, we present a system to identify and recommend authorities in a particular area based on machine learning, supervised and unsupervised techniques. To verify accuracy and our model, we developed several tests within a particular topic. The results show that the system can identify authorities with an average accuracy of [78.82 ±2.51] %, representing a high degree of confidence.
Translated title of the contributionIdentificación y Recomendación de Autoridades en Diferentes Temas Basado en Twitter
Original languageEnglish (US)
Pages (from-to)1-10
Number of pages10
JournalInternational Journal of Pure and Applied Mathematics
Issue number114
StatePublished - 1 Jan 2017


  • Big data
  • Recommender systems
  • Social networks
  • Twitter

CACES Knowledge Areas

  • 116A Computer Science


Dive into the research topics of 'Identification and Recommendation of Authorities on Different Topics Based on Twitter'. Together they form a unique fingerprint.

Cite this