A novel approach for modeling positive vectors with inverted Dirichlet-based hidden Markov models

作者：

Highlights：

• The first integration of inverted Dirichlet mixtures in hidden Markov models.

• Efficient modeling of positive vectors for prediction and pattern recognition tasks.

• Adaptation of hidden Markov models to handle positive vectors.

• Improvement of recognition accuracies compared to traditional hidden Markov models.

摘要

•The first integration of inverted Dirichlet mixtures in hidden Markov models.•Efficient modeling of positive vectors for prediction and pattern recognition tasks.•Adaptation of hidden Markov models to handle positive vectors.•Improvement of recognition accuracies compared to traditional hidden Markov models.

论文关键词：Hidden Markov Models,Inverted Dirichlet,Texture categorization,Occupancy detection,Smart buildings,Facial expressions recognition

论文评审过程：Received 13 August 2019, Revised 1 November 2019, Accepted 1 December 2019, Available online 4 December 2019, Version of Record 24 February 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2019.105335