Context‐based classification via mixture of hidden Markov model experts with applications in landmine detection

Abstract

In many applications data classification may be hindered by the existence of multiple contexts that produce an input sample. To alleviate the problems associated with multiple contexts, context‐based classification is a process that uses different classifiers depending on a measure of the context. Context‐based classifiers offer the promise of increasing performance by allowing classifiers to become experts at classifying input samples of certain types, rather than trying to force single classifiers to perform well on all possible inputs. This study introduces a novel mixture of experts (ME) model, the mixture of hidden Markov model experts, for context‐based classification of samples that are variable length sequences; and derives the update equations for a single probabilistic model that to learn the experts and a gate that connects the experts. The model has a similar high‐level structure to the ME model but has the novelty that the gates and the experts are HMMs and the input data are sequences. Experimental results are presented on three datasets including one for landmine detection. Detailed analysis of the model is provided; which, over multiple runs and cross‐validation experiments, show superior results over the compared algorithms.

Investigating Transfer Learning Performances of Deep Learning Models for Classification of GPR B-Scan Images

“…However, due to the challenging nature of processing such data and its difficult interpretation, automated approaches were needed and developed over time [1]. Although, the number of studies using conventional machine learning techniques were decreased in the past 3-4 years, Support Vector Machines [2], Artificial Neural Networks (ANN) [3,4], boosting algorithms [5,6], Hidden Markov Models [7] were utilized for detection and classification tasks on GPR images. With the improvement in computation power of GPUs, deep learning techniques which proved their superiority on image classification started to replace them.…”

Treatment for Patients with Severe Heart Failure after Coronary Artery Bypass Grafting Based on the application Markov Model

“…The treatment and nursing care for patients with severe heart failure is an important factor that affects the hemodynamic characteristics and requires in-depth research [15,16] . The research on the numerical simulation based on the application Markov model can provide a more reasonable data reference for the clinical diagnosis of the coronary restenosis [17] . In this paper, on the basis of the CT image data of the patients, a 3D application Markov model after the coronary artery bypass grafting is reconstructed.…”

Practical and theoretical aspects of mixture‐of‐experts modeling: An overview

“…MoE models have been broadly applied to numerous areas of business, science, and technology for the tasks of classification, clustering, and regression. A sample of recent applications that were not covered by Yuksel et al (2012) and Masoudnia and Ebrahimpour (2014) includes: modeling neural connectivity (Bock & Fine, 2014), fusion and segmentation of images (Camplani, del Blanco, Salgado, Jaureguizar, & Garcia, 2014), segmentation of spectral images (Cohen & Le Pennec, 2014), phone activity recognition (Lee & Cho, 2014), climatic change modeling (Nguyen & McLachlan, 2014), parallel mapping of threads in dynamic runtime environments (Emani & O'Boyle, 2015), cardiac stress monitoring via heart sounds (Herzig, Bickel, Eitan, & Intrator, 2015), aerodynamic performance predictions (Liem, Mader, & Martins, 2015), functional magnetic resonance image analysis (Shoenmakers, Guclu, van Gerven, & Heskes, 2015), heterogeneity modeling in neural connectivity data (Eavani et al, 2016), reinforcement learning (He, Boyd-Graber, Kwon, & Daume III, 2016), landmine detection (Yuksel & Gader, 2016), and attention deficit hyperactivity disorder diagnosis (ADHD) (Yaghoobi Karimu & Azadi, 2017).…”