We propose a method for computing user similarity based on a network representing the semantic relationships between the words occurring in the same tweet and the related topics.We use such specially crafted network to define several user profiles to be compared with cosine similarity. We also describe an initial experimental activity to study the effectiveness on a limited dataset.
CITATION STYLE
Mizzaro, S., Pavan, M., & Scagnetto, I. (2015). Content-based similarity of Twitter users. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9022, pp. 507–512). Springer Verlag. https://doi.org/10.1007/978-3-319-16354-3_56
Mendeley helps you to discover research relevant for your work.