A multi-source heterogeneous data analytic method for future price fluctuation prediction

Open Webpage

Lei Chai, Hongfeng Xu, Zhiming Luo, Shaozi Li

Published: 01 Jan 2020, Last Modified: 11 Oct 2025Neurocomputing 2020EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•We proposed a forecasting framework with multiple source heterogeneous data.•We constructed a domain-specific relation map to reduce the sparsity issue in news events data.•We built a domain specific dictionary for better quantifying the text-related information.•A Hidden Mixture Gaussian Markov Model is applied for the time-series data.