Daoui Cherki

• PhD
• Chair at Université Sultan Moulay Slimane

In the era of ChatGPT, user feedback is essential for understanding this advanced technology and navigating the evolving field of artificial intelligence. To gain insights into the risks, limitations, and areas for improvement, our study focuses on analyzing Twitter interactions. We address the unique challenges posed by Twitters characteristicsbre...

Efficiently managing networks is crucial for handling the increasing data flow in IoT devices. IoT gateways act as intermediaries between devices and cloud servers, managing data transfer. Effective bandwidth management is key to ensuring smooth data movement through these gateways. However, current methods struggle to adapt to different needs, esp...

This work is part of a research project carried out during the COVID-19 pandemic, involving the design and realization of an autonomous mobile hospital robot. Many real-world robotic tasks suffer from the critical characteristics: Noisy sensing, imperfect control, and environment changes. The Markov decision process MDP and its variants provide a m...

The Internet of Things (IoT) is evolving, driven by the increasing demand for bandwidth. A key focus is on minimizing communication delays. This paper introduces a new solution called Predictive Dynamic Bandwidth Allocation (PDBA), using adaptive predictive algorithms in the IoT context. The approach involves predicting resource needs and network c...

This work is part of a research project during the COVID-19 pandemic that aims to design and develop a mobile autonomous robot for hospitals. In practice, implementing a navigation program directly on a physical robot is both expensive and hazardous. The solution is to perform a simulation using ROS (Robot Operating System), which offers several ad...

Research background Markov Decision Processes ( MDPs ) are a powerful framework for modeling many real-world problems with finite-horizons that maximize the reward given a sequence of actions. Although many problems such as investment and financial market problems where the value of a reward decreases exponentially with time, require the introducti...

The increasing number of situations in which wireless communication networks are damaged due to natural disasters has motivated researchers to utilize unmanned aerial vehicles (UAVs) to deliver fast and efficacious backup communication in post-disaster scenarios. UAVs are a logical option for public safety cellular networks due to their key feature...

Conventional algorithms for solving Markov decision processes (MDPs) become intractable for a large finite state and action spaces. Several studies have been devoted to this issue, but most of them only treat infinite-horizon MDPs. This paper is one of the first works to deal with non-stationary finite-horizon MDPs by proposing a new decomposition...

Artificial intelligence (AI) is showing its success in various types of applications. Motivated by this trend, automatic trading has taken a keen interest in applying of artificial intelligence methods to predict the future price of a financial asset to overcome trading challenges including asset price fluctuations and dynamics, Investors must ther...

Alzheimer’s disease is recognized as a progressive loss of memory and often leads to a total loss of autonomy, which renders it difficult to tolerate. Hippocampus is among the first cerebral regions to be infected in AD. An accurate diagnosis at an early stage of AD is crucial for the intervention process. The low contrast of supporting tissues and...

In this work, we segment some fuzzy color and grey level images using the fuzzy model of Markov: Fuzzy Hidden Markov Chain, we use three algorithms EM, SEM and ICE for estimating the parameters to this model. We compare these estimators under some criteria of evaluation such as: segmentation quality, running time and convergence and complexity. The...

This paper presents an application of textured color images segmentation using hidden Markov chain model. We propose two comparative studies, The first one is between EM(Expectation-Maximization) algorithm,SEM (Stochastic Expectation-Maximization) algorithm and ICE (Iterative Conditional Estimation) algorithm, these estimators are used to estimate...

Hidden Markov Models have been extensively used in various fields, especially in speech recognition, biology, image and signal processing and digital communication. They are well known by their effectivenss in modeling the correlations between adjacent symbols, domains or events, but they often suffer from high dimensionality problems. In this work...

This paper presents four stationaries models of Markov used in image segmentation, such as: Hidden Markov Chain with Independent Noise, Hidden Markov Chain, Pairwise Markov Chain and Pairwise Markov Chain with Independent Noise. We carry out a comparative study between these models, this comparaison is in level of segmentation quality. From the val...

We consider Stochastic Shortest Path (SSP) Markov Decision Processes (MDPs) with dead ends and energy constraint. The objective is to find an optimal policy that maximizes the probability of reaching the target and minimizes the expected cost if the energy is sufficient. Firstly, we present a new Transformed SSP MDP that guarantees the convergence...

The isolated handwritten character recognition with multiple styles is a challenging research problem. In this paper, we propose a novel method of features extraction for character recognition based on the mathematical morphology and histogram techniques into vertical, horizontal, diagonal and anti-diagonal directions, knowing that the features...

This paper is about the design of control sequences for discrete event systems (DESs) modeled with bounded partially controlled timed Petri nets (PC-TPNs) including a set of temporal specifications that correspond to minimal firing durations. Petri nets are well-known mathematical and graphical models that are widely used to describe distributed DE...

The standard Value Iteration (VI) algorithm, referred to as Value Iteration Pre-Jacobi (PJ-VI) algorithm, is the simplest Value Iteration scheme, and the well-known algorithm for solving Markov Decision Processes (MDPs). In the literature, several versions of VI algorithm were developed in order to reduce the number of iterations: the VI Jacobi (VI...

p>Image segmentation is a fundamental operation in image processing, which consists to di-vide an image in the homogeneous region for helping a human to analyse image, to diagnose a disease and take the decision. In this work, we present a comparative study between two iterative estimator algorithms such as EM (Expectation-Maximization) and ICE (It...

A review of literature shows that there is a variety of works studying coverage path planning in several autonomous robotic applications. In this work, we propose a new approach using Markov Decision Process to plan an optimum path to reach the general goal of exploring an unknown environment containing buried mines. This approach, called Goals to...

Many hierarchical techniques to solve large Markov decision processes (MDPs) are based on the partition of the state space into strongly connected components (SCCs) that can be classified into some levels. In each level, smaller problems named restricted MDPs are solved, and then these partial solutions are combined to obtain the global solution. I...

A review of literature shows that there are a variety of adaptive filters. In this research study, we propose a new type of adaptive filter that increases the diversification used to compensate the channel distortion effect in the MC-CDMA transmission. First, we show expressions of the impulse responses of the filter in the case of a perfect channe...

As part of a detailed study on blind identification of Gaussian channels, the main purpose was to propose an algorithm based on cumulants and fuzzy number approach involved throughout the whole process of identification. Our objective was to compare the new design of the algorithm to the old one using the higher order cumulants, namely Alg1, Algat...

In this work we will create a system of recognition of Moroccan plate-style vehicle license plates using the invariant moments in features extraction and the support vectors machines such as Moroccan plate is slightly treated by researchers in pattern recognition field. In this context we propose a data set for Moroccan plates composed of 100 image...

In this paper, we present two comparisons in isolated printed Tifinagh characters recognition, in fact the first comparison is between four hybrid methods exploited in features extraction which are the retinal coding combined with the Hu then with Legendre then with Zernike invariant moments, finally with these tree moments at the same time; in con...

In this paper, we present a comparison between the different variations of virtual retina (grid size) in features extraction with the support vectors machines classifier for isolated handwritten Eastern Arabic numerals recognition. For this purpose we have used for pre-processing each numeral image the median filter, the thresholding, normalization...

In this paper, we present a comparison between the different variations of virtual retina (grid size) in features extraction with the support vectors machines classifier for isolated handwritten Eastern Arabic numerals recognition. For this purpose we have used for pre-processing each numeral image the median filter, the thresholding, normalization...

In this work, we present an automatic speech classification system for the Tamazight phonemes. We based on the spectrum presentation of the speech signal to model these phonemes. We have used an oral database of Tamazight phonemes. To test the system’s performances, we calculate the classification rate. The obtained results are satisfactory in comp...

Supplementary data

The territorial organization of Morocco during administratives division of 2009 is based on 16 regions. In this work we will create a system of recognition of handwritten words (names of regions) using the Amazigh language is an official language by the Moroccan Royal Institute of Amazigh Culture (IRCAM) (2003a) [1] such as this language is slightl...

As many real applications need a large amount of states, the classical methods are intractable for solving large Markov Decision Processes. The decomposition technique basing on the topology of each state in the associated graph and the parallelization technique are very useful methods to cope with this problem. In this paper, the authors propose a...

In this paper we have achieved a comparison between the performances of different types of distances used in the k nearest neighbors classifier for recognition of isolated handwritten Arabic numerals extracted from Mnist database, more precisely these distances are those of Euclid, of Manhathan, of Minkowski, and of Tchebychev, to do this, we have...

In this paper, we present two comparisons in isolated handwritten Roman numerals recognition, in fact the first comparison is between zoning methods exploited in features extraction which are the square zones or the triangular zones used in extraction characteristic; in contrast the second comparison is performed in order to deduce what is the most...

In this paper we present a comparison between two methods of learning-classification, the first is the K-Nearest Neighbors (KNN) and the second is the Support Vectors Machines (SVM), these both methods are supervised and used for the recognition of handwritten Latin numerals that are extracted from the MNIST standard database. The recognition proce...

This paper deals with a recognition system of handwritten Arabic numerals extracted to the MNIST standard database (Arabic numerals). This system is composed of three main phases: the pre-processing of numerals followed by the extraction of primitives with the zoning method in order to convert each image into a vector number, which is nothing other...

In this work we will create a system of recognition of Moroccan plate-style vehicle license plates using the invariant moments in features extraction and the support vectors machines such as Moroccan plate is slightly treated by researchers in pattern recognition field. In this context we propose a data set for Moroccan plates composed of 100 image...

In this paper we present a comparison between two supervised classifiers, the first one is a statistic which is the K-Nearest Neighbors (KNN) while the second is a neuronal which is the multi-layer perceptron MLP in the recognition of cursive handwritten Arabic numerals. The recognition process is organized as follows: in the pre-processing of nume...

In this research, we present two comparative studies; the first one is between two methods of features extraction which are the mathematical morphology, the zoning and the hybridization of these two methods. The second comparative study is between both supervised methods used in learning-classification which are the Multi-Layer Perceptron (MLP) and...

The user behavior on a website triggers a sequence of queries that have a result which is the display of certain pages. The Information about these queries (including the names of the resources requested and responses from the Web server) are stored in a text file called a log file. Analysis of server log file can provide significant and useful...

This paper deals with a recognition system of character for handwritten Tifinagh Text. Here in this work a neural network (the multi-layer perceptron MLP) and Hidden Markov Models (HMM) are proposed for handwritten characters identification. The features of Tifinagh characters are abstracted by mathematical morphology. Acquisition, scanning, thinni...

E-learning is considered as one of the areas in which the Semantic Web can make a real improvement whatsoever in finding information, or reusing of educational resources or even personalized learning paths. This paper aimsto develop an educational ontology that will be used to annotate learning materials and pedagogical documents.

This paper deals with an optical character recognition (OCR) system of handwritten digit, with the use of neural networks (MLP multilayer perceptron). And a method of extraction of characteristics based on the digit form, this method is tested on the MNIST handwritten isolated digit database (60000 images in learning and 10000 images in test). This...

In this work, we present a system for automatic speech recognition on the Moroccan dialect. We used the hidden Markov model to model the phonetic units corresponding to words taken from the training base. The results obtained are very encouraging given the size of the training set and the number of people taken to the registration. To demonstrate t...

As classical methods are intractable for solving Markov decision processes (MDPs) requiring a large state space, decomposition and aggregation techniques are very useful to cope with large problems. These techniques are in general a special case of the classic Divide-and-Conquer framework to split a large, unwieldy problem into smaller components a...

We consider limiting average Markov decision processes (MDP) with finite state and action spaces. We propose some algorithms to determine optimal strategies for deterministic and general MDPs. These algorithms are based on graph theory and the construction of levels in some aggregated MDP.

We consider a discrete time finite Markov decision process (MDP) with the discounted and weighted reward optimality criteria. In [1] the authors considered some decomposition of limiting average MDPs. In this paper, we use an analogous approach for discounted and weighted MDPs. Then, we construct some hierarchical decomposition algorithms for both...

. We consider discrete time Markov Decision Process (MDP) with finite state and action spaces under average reward optimality criterion. The decomposition theory, in Ross and Varadarajan [11], leads to a natural partition of the state space into strongly communicating classes and a set of states that are transient under all stationary strategies. T...