Discovery of Shifting Patterns in Sequence Classification
This addresses the challenge of shifting patterns in sequential data for applications such as environmental monitoring and emotion recognition, representing an incremental improvement by combining multi-instance learning with LSTM modeling.
The paper tackles the problem of multi-variate sequence classification where discriminative patterns shift over time, degrading traditional methods, and proposes a novel method that improves classification performance and detects these shifting patterns in real-world applications like cropland mapping and affective state recognition.
In this paper, we investigate the multi-variate sequence classification problem from a multi-instance learning perspective. Real-world sequential data commonly show discriminative patterns only at specific time periods. For instance, we can identify a cropland during its growing season, but it looks similar to a barren land after harvest or before planting. Besides, even within the same class, the discriminative patterns can appear in different periods of sequential data. Due to such property, these discriminative patterns are also referred to as shifting patterns. The shifting patterns in sequential data severely degrade the performance of traditional classification methods without sufficient training data. We propose a novel sequence classification method by automatically mining shifting patterns from multi-variate sequence. The method employs a multi-instance learning approach to detect shifting patterns while also modeling temporal relationships within each multi-instance bag by an LSTM model to further improve the classification performance. We extensively evaluate our method on two real-world applications - cropland mapping and affective state recognition. The experiments demonstrate the superiority of our proposed method in sequence classification performance and in detecting discriminative shifting patterns.