Modeling and forecasting of rainfall reoccurrence changes using Markov Switching in Iran

This paper represents the recurrence (reoccurrence) changes in the rainfall series using Markov Switching models (MSM). The switching employs a dynamic pattern that allows a linear model to be combined with nonlinearity models a discrete structure. The result is the Markov Switching models (MSM) reoccurrence predicting technique. Markov Switching models (MSM) were employed to analyze rainfall reoccurrence with spatiotemporal regime probabilities. In this study, Markov Switching models (MSM) were used based on the simple exogenous probability frame by identifying a first-order Markov process for the regime probabilities. The Markov transition matrix and regime probabilities were used to analyze the rainfall reoccurrence in 167 synoptic and climatology stations. The analysis results show a low distribution from 0.0 to 0.2 (0–20%) per day spatially from selecting stations, probability mean of daily rainfall recurrence is 0.84, and a different distribution based on the second regime was found to be more remarkable to the rainfall variability. The rainfall reoccurrence in daily rainfall was estimated with relatively low variability and strong reoccurrence daily with ranged from 0.851 to 0.995 (85.1–99.5%) per day based on the spatial distribution. The variability analysis of rainfall in the intermediate and long variability and irregular variability patterns would be helpful for the rainfall variability for environmental planning.

A Comparative Study of the Impact of Data Augmentation in Machine Learning Based Classification Accuracy

Traumatic Brain Injury is the primary cause of death and disability all over the world. Monitoring the intracranial pressure (ICP) and classifying it for hypertension signals is of crucial importance. This thesis explores the possibility of a better classification of the ICP signal and detection of hypertensive signal prior to the actual occurrence of the hypertensive episodes. This study differ from other approaches astime series is converted into images by Gramian angular field and Markov transition matrix and augmented with data. Due to unbalanced data, the effect of smote extended nearest neighbour algorithm for balancing the data is examined. We use various machine learning algorithms to classify the ICP signals. The results obtained shoe that Ada boost performance is the best among compared algorithms. F1 score of the Ada boost is 0.95 on original dataset, and 0.9967 on balanced and augmented dataset. Quadratic Discriminant Analysis F1 score is 1 when data is augmented and balanced.

A Comparative Study of the Impact of Data Augmentation in Machine Learning Based Classification Accuracy

Traumatic Brain Injury is the primary cause of death and disability all over the world. Monitoring the intracranial pressure (ICP) and classifying it for hypertension signals is of crucial importance. This thesis explores the possibility of a better classification of the ICP signal and detection of hypertensive signal prior to the actual occurrence of the hypertensive episodes. This study differ from other approaches astime series is converted into images by Gramian angular field and Markov transition matrix and augmented with data. Due to unbalanced data, the effect of smote extended nearest neighbour algorithm for balancing the data is examined. We use various machine learning algorithms to classify the ICP signals. The results obtained shoe that Ada boost performance is the best among compared algorithms. F1 score of the Ada boost is 0.95 on original dataset, and 0.9967 on balanced and augmented dataset. Quadratic Discriminant Analysis F1 score is 1 when data is augmented and balanced.

Predicting Land Use Changes within the Tazekka Park and Its Borders via a Cellular Automata-Markov Modeling of Satellite Images

This article describes a new approach of land cover study to predicting and combatting deforestation based on satellite imagery as environmental statistics. Specifically, a stochastic mathematical cellular automata-Markov model was used to predict land-use changes in the Tazekka Park and its borders in TAZA province in Morocco. The model was used mainly to create thematic forecast maps. Through the proposed approach, we derived data and statistics covering the period 2000 to 2020 and then constructed a predictive map for the year 2040 using ArcGIS 10.4. The evaluation of our model’s effectiveness was confirmed by calculating the Markov transition matrix in the derivation of the final map. These results can improve the management of forest areas and serve as a reference in addressing the direct effects of forests on the environment.

Effective Network Level Optimization for Bridges Maintenance

The objective of this paper is to present an effective new methodology to optimize the maintenance costs of bridges stock. Optimization takes place at the network level and not in a project level (bridge by bridge). The dynamics of passage between bridges condition state (from 1 to 5) is achieved by the Markov chains probabilistic method. The Markov transition matrix is determined either by ratios of total areas and areas degraded annually, or by the resolution of an optimization problem. In the latter case, the nonlinear optimization algorithm SQP (Sequanciel Quadratic Programming) is developed. A bridge maintenance matrix is introduced in the calculation of the repair cost. The originality of our approach is to parameterize this matrix by introducing the different optimization variables of the problem. Finally, the cost function to be optimized annually is calculated and optimized by a genetic algorithm. This cost function represents the cost of maintaining the entire asset.

Using Markov Chains to Determine Expected Propagation Time for Probabilistic Zero Forcing

Zero forcing is a coloring game played on a graph where each vertex is initially colored blue or white and the goal is to color all the vertices blue by repeated use of a (deterministic) color change rule starting with as few blue vertices as possible. Probabilistic zero forcing yields a discrete dynamical system governed by a Markov chain. Since in a connected graph any one vertex can eventually color the entire graph blue using probabilistic zero forcing, the expected time to do this is studied. Given a Markov transition matrix for a probabilistic zero forcing process, an exact formula is established for expected propagation time. Markov chains are applied to determine bounds on expected propagation time for various families of graphs.

Steganalysis of Adaptive Multi-Rate Speech Based on Extreme Gradient Boosting

Steganalysis of adaptive multi-rate (AMR) speech is a hot topic for controlling cybercrimes grounded in steganography in related speech streams. In this paper, we first present a novel AMR steganalysis model, which utilizes extreme gradient boosting (XGBoost) as the classifier, instead of support vector machines (SVM) adopted in the previous schemes. Compared with the SVM-based model, this new model can facilitate the excavation of potential information from the high-dimensional features and can avoid overfitting. Moreover, to further strengthen the preceding features based on the statistical characteristics of pulse pairs, we present the convergence feature based on the Markov chain to reflect the global characterization of pulse pairs, which is essentially the final state of the Markov transition matrix. Combining the convergence feature with the preceding features, we propose an XGBoost-based steganalysis scheme for AMR speech streams. Finally, we conducted a series of experiments to assess our presented scheme and compared it with previous schemes. The experimental results demonstrate that the proposed scheme is feasible, and can provide better performance in terms of detecting the existing steganography methods based on AMR speech streams.

Empirical Convergence Rate of a Markov Transition Matrix

The convergence rate of a Markov transition matrix is governed by the second largest eigenvalue, where the first largest eigenvalue is unity, under general regularity conditions. Garren and Smith (2000) constructed confidence intervals on this second largest eigenvalue, based on asymptotic normality theory, and performed simulations, which were somewhat limited in scope due to the reduced computing power of that time period. Herein we focus on simulating coverage intervals, using the advanced computing power of our current time period. Thus, we compare our simulated coverage intervals to the theoretical confidence intervals from Garren and Smith (2000).