Recommendation system of authorities and content based on Twitter for language therapy through data mining techniques

Diego Quisi-Peralta, Vladimir Robles-Bykbaev, Martin Lopez-Nores, Liliana Chaglla-Rodriguez, Diego Chiluisa-Castillo

Research output: Contribution to conferencePaper

1 Scopus citations


According to latest estimations of the World Health Organization (WHO), approximately 15% of persons have some disability. From this group, a significant percentage of persons present different kinds of communication and language disorders. Additionally, it is important mentioning that language supports other essential processes in children development, such as learning, skills to interact with his/her peers, and establishing relationships. On the other hand, nowadays exist several digital platforms like blogs, twitter, and in general, social networks, where experts and practitioners contribute with contents related to speech and language therapy. For these reasons, in this paper, we present a recommender system to support the content filtering, identify experts or interest groups related to communication disorders. Our recommender uses data mining techniques to perform the contents filtering, and a clustering approach to classifying the wich twitter contents are related to speech-language therapy area. In order to validate our proposal, we have conducted several tests with the support of a team of experts. The achieved results show that the recommendations of the system are useful, coherent, and understandable.

Original languageEnglish
StatePublished - 20 Feb 2019
Event2018 IEEE Biennial Congress of Argentina, ARGENCON 2018 - San Miguel de Tucuman, Argentina
Duration: 6 Jun 20188 Jun 2018


Conference2018 IEEE Biennial Congress of Argentina, ARGENCON 2018
Abbreviated titleARGENCON 2018
CitySan Miguel de Tucuman


  • Language therapy
  • artificial intelligence
  • natural language processing
  • recommendation system
  • twitter


Dive into the research topics of 'Recommendation system of authorities and content based on Twitter for language therapy through data mining techniques'. Together they form a unique fingerprint.

Cite this