Invariant Factorization Of Time-Series
This addresses time-series classification for domains needing improved accuracy, though it appears incremental as it builds on existing decomposition and feature extraction ideas.
The study tackled time-series classification by developing a method that decomposes datasets into latent patterns and membership weights, projecting them into a new feature representation. It outperformed 6 state-of-the-art baselines on 43 real-life datasets with statistically significant accuracy gains.
Time-series classification is an important domain of machine learning and a plethora of methods have been developed for the task. In comparison to existing approaches, this study presents a novel method which decomposes a time-series dataset into latent patterns and membership weights of local segments to those patterns. The process is formalized as a constrained objective function and a tailored stochastic coordinate descent optimization is applied. The time-series are projected to a new feature representation consisting of the sums of the membership weights, which captures frequencies of local patterns. Features from various sliding window sizes are concatenated in order to encapsulate the interaction of patterns from different sizes. Finally, a large-scale experimental comparison against 6 state of the art baselines and 43 real life datasets is conducted. The proposed method outperforms all the baselines with statistically significant margins in terms of prediction accuracy.