Concept drift detection on social network data using cross-recurrence quantification analysis.

This paper presents our efforts to detect Concept Drifts (changes in data generation processes), using the Cross-Recurrence Quantification Analysis, on time series produced by social network systems. Experiments were performed on the TSViz project (http://www.tsviz.com.br), which collects online tweets associated with predefined hashtags and processes them to generate different time series: one to measure the amount of information contained in textual short messages and another to quantify the positiveness and negativeness of users' sentiments, etc. In that context, this work proposed and evaluated a Concept Drift approach to point out when generating processes change along time, indicating the detection of relevant textual changes in terms of the amount of information and sentiments. As a main contribution, results show that our approach indicates when the most important social events happen, which were confirmed by official news.