Modeling Emerging, Evolving and Fading Topics Using Dynamic Soft Orthogonal NMF with Sparse Representation

Dynamic topic models (DTM) are of great use to analyze the evolution of unobserved topics of a text collection over time. Recent years have witnessed the explosive growth of streaming text data emerging from online media, which creates an unprecedented need for DTMs for timely event analysis. While there have been some matrix factorization methods in the literature for dynamic topic modeling, further study is still in great need to model emerging, evolving and fading topics in a more natural and effective way. In light of this, we first propose a matrix factorization model called SONMFSR (Soft Orthogonal NMF with Sparse Representation), which makes full use of soft orthogonal and sparsity constraints for static topic modeling. Furthermore, by introducing the constraints of emerging, evolving and fading topics to SONMFSR, we easily obtain a novel DTM called SONMFSR_d for dynamic event analysis. Extensive experiments on two public corpora demonstrate the superiority of SONMFSR_d to some state-of-the-art DTMs in both topic detection and tracking. In particular, SONMFSR_d shows great potential in real-world applications, where popular topics in Two Sessions 2015 are captured and traced dynamically for possible insights.