Temporal Feature Selection on Networked Time Series
This addresses the limitation of existing shapelet selection methods that assume i.i.d. time series, enabling more accurate analysis of social networked data like tweets, though it is incremental as it builds on prior feature selection approaches.
The paper tackles the problem of learning discriminative features from networked time series data by proposing a Network Regularized Least Squares (NetRLS) model that incorporates linked information among time series, and it outperforms the state-of-the-art LTS method on real-world Twitter and DBLP data.
This paper formulates the problem of learning discriminative features (\textit{i.e.,} segments) from networked time series data considering the linked information among time series. For example, social network users are considered to be social sensors that continuously generate social signals (tweets) represented as a time series. The discriminative segments are often referred to as \emph{shapelets} in a time series. Extracting shapelets for time series classification has been widely studied. However, existing works on shapelet selection assume that the time series are independent and identically distributed (i.i.d.). This assumption restricts their applications to social networked time series analysis, since a user's actions can be correlated to his/her social affiliations. In this paper we propose a new Network Regularized Least Squares (NetRLS) feature selection model that combines typical time series data and user network data for analysis. Experiments on real-world networked time series Twitter and DBLP data demonstrate the performance of the proposed method. NetRLS performs better than LTS, the state-of-the-art time series feature selection approach, on real-world data.