Belal Al-Khateeb | University of Anbar (original) (raw)
Papers by Belal Al-Khateeb
المجلة العراقية للحاسبات والمعلوماتية, Sep 13, 2023
Medication supply and storage are essential components of the medical industry and distribution. ... more Medication supply and storage are essential components of the medical industry and distribution. Most medications have a predetermined expiration date. When the demand is met in large quantities that exceed the actual need, this leads to the accumulation of medicines in the stores, and this leads to the expiration of the materials. If demand is too low, this will have an impact on consumer happiness and drug marketing. Therefore, it is necessary to find a way to predict the actual quantity required for the organization's needs to avoid material spoilage and storage problems. A mathematical prediction model is required to assist any management in achieving the required availability of medicines for customers and safe storage of medicines. The research question is to design a system based on deep learning that can predict the amount of drugs required with high efficiency and accuracy based on the chronology of previous years.Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Bidirectional LSTM, and Gated Recurrent Unit (GRU) are used to build prediction models. Those models allow for the optimization of inventory levels, thus reducing costs and potentially increasing sales. Various measures such as mean squared error (MSE), mean absolute squared error (MASE), root mean squared error (RMSE), and others are used to evaluate the prediction models. RNN model achieved the best result with MSE: 0.019 MAE: 0.102, RMSE: 0.0.
At its inception, cuneiform signs were designed to represent sounds, enabling the recording of sp... more At its inception, cuneiform signs were designed to represent sounds, enabling the recording of spoken language. This served to preserve language for posterity, allowing thoughts and concepts to be conveyed in written form. Among the earliest known forms of writing is cuneiform, etched onto clay tablets using a reed stylus with a wedge-shaped tip, hence its name [1]. Presently, 'cuneiform' denotes a writing system characterized by wedge-shaped markings. While numerous transliterated Akkadian texts that can be accessed for free in digitized corpora; however, there are a much smaller number of texts that have been phonologically transcribed. Employed for three millennia, cuneiform script emerged in the late fourth millennium BCE and persisted until the first century CE, producing a diverse array of texts by scribes. Though primarily etched onto clay, some found their place on wax-coated stone, metal, or wooden boards. Over a million cuneiform texts have been uncovered to date in a large Near Eastern region stretching from Anatolia to Iran and northern Iraq to Bahrain and Egypt [2]. This region includes a portion of the Arabian Peninsula and the Persian Gulf. The amalgamation of wedges gives rise to an array of cuneiform signs, given that the script is created by pressing a stylus into fresh clay. Cuneiform was developed in ancient Mesopotamia, the present methodology involves manual reconstruction of missing data [3]. Executed by a select group proficient in Akkadian and cuneiform, this laborintensive process seeks to recover obscured lines in scientific, lexical, literary, and religious texts through comparison with intact tablets and parallel passages on deteriorated ones. Fortunately, the availability of duplicate copies, facilitates this endeavor. Nevertheless, it necessitates profound familiarity with various text types and collections, and the suggested restorations by scholars often invite diverse interpretations [4]. Though the system's brilliance permits effortless replication, scribes grappled with memorizing myriad signs. This discrepancy arises from the labor-intensive nature of transcription, which offers limited aid in academically interpretating cuneiform texts. Numerous findings include literary works like the Epic of Gilgamesh, clay envelopes containing letters that were sealed with clay, and historical narratives. Remarkably, vast libraries, exemplified by the collection of Assyrian king Ashurbanipal (668-627 B.C.E.), have come to light [1]. An astronomical text dated to 75 CE is the most recent example ABSTRACT: The cuneiform script reveals some previously unknown aspects of our past. However, reading ancient clay tablets demands a substantial investment of time and persistent practice over a long period of time. As the fourth millennium came to a close, earlier recording methods gave way to the development of writingthe visual representation of spoken language. The first language to be transcribed in written form in Mesopotamia was Sumerian. Predominantly, the earliest tablets originate from the Uruk site in southern Mesopotamia, possibly marking its birthplace. Digitization cuneiform documents is imperative to boost research focused on the ancient Middle East. A few initiatives embarked upon this endeavor around the year 2000. Nonetheless, the digitization process is time-consuming due to the extensive volume of documents, and a dependable (semi) automatic methodology has yet to be developed. Given the antiquity of cuneiform script, recognizing cuneiform signs using real-world applications via two graph-based algorithms, each with complementary runtime characteristics, remains a manual procedure. Translating cuneiform proves to be a daunting task. Only in relatively recent times has grammar been established scientifically, while lexical challenges remain abundant and far from resolved. Furthermore, the majority of the Sumerian tablets have succumbed to the ravages of time, leaving behind only a handful of ancient depictions. Some of these old images have been preserved in a unique collection or in museums worldwide, allowing specialists to easily apply the sign detector to their cuneiform text studies. In this paper, we will discuss the categorization and analysis of clay tablets using a trained cuneiform model, employing artificial intelligence methodologies. Additionally, we will explore the methods employed, highlighting their strengths and weaknesses. Finally, we will propose potential directions for future research.
Expert Systems with Applications, 2009
... 2. The solved course timetabling problem. The definition of the solved problem given below wa... more ... 2. The solved course timetabling problem. The definition of the solved problem given below was introduced by Aladag and Hocaoglu (2007a). 2.1. Objectives and constraints of the problem. ... TR, Different types of rooms, m rt, Number of rooms of type r, F, Set of pre-assigned lessons ...
Lecture notes in networks and systems, 2022
Xinan Jiaotong Daxue Xuebao, Jun 30, 2019
Particle Swarm Optimization (PSO) is a very common algorithm in swarm intelligence algorithms. PS... more Particle Swarm Optimization (PSO) is a very common algorithm in swarm intelligence algorithms. PSO has been used to solve a lot of problems with one or more goals. Actuality, the multi-objectives improvement issues in all real life are combinatorial in nature. Therefore, PSO has been improved to be able to handle very large number of decision variables and reduce or decrease computational complexity. In this work, a chaos multi objective PSO algorithm is improved for solving discrete (binary) optimization issues with crossover operation. The developed Chaos Discrete Multi Objective PSO (CDMOPSO) algorithm is applied to pavement management problem for flexible pavement to get optimal maintenance and rehabilitation plan. The results shown that there is significant improvement in the solutions satisfying pavement conditions and maintenance cost goals. It is required to a very short time of execution by the improved algorithm to reach a very good solution. Also, comparing the convergence of solutions with the rest of the PSO algorithms, it has found that the suggested algorithm is better.
Journal of Personalized Medicine, Feb 18, 2022
Social Science Research Network, Sep 25, 2019
Analysis of image contents has become one of the important subjects in modern life. In order to r... more Analysis of image contents has become one of the important subjects in modern life. In order to recognize the images with efficient way, several techniques have appeared and periodically enhanced by the developers. Image retrieval becomes one of the main problems that face the computer society inside the revolution of technology. To increase the effectiveness of computing similarities between images, hashing approaches became the focusing of the programmers. Indeed, deep learning in the past few years has been considered the backbone of image analysis using a convolutional neural network (CNN). The paper is providing a survey of the latest work carried out in the field of image retrieval. Several techniques have appeared in this field. However, the most common of these techniques are using neural network-based hash encoding, which can be categorized into three main classes: Supervised, unsupervised, and semi-supervised techniques according to each technique's learning method. The most important related works appeared in the literature are reviewed and constructive comparisons have been done to show the strengths and limitations of various techniques. Keywords: Image Retrieval, Deep Learning, Convolutional Neural Network (CNN), Hashing Techniques. RESUMEN / El análisis del contenido de la imagen se ha convertido en uno de los temas importantes en la vida moderna. Para reconocer las imágenes de manera eficiente, han aparecido varias técnicas que los desarrolladores han mejorado periódicamente. La recuperación de imágenes se convierte en uno de los principales problemas que enfrenta la sociedad informática dentro de la revolución de la tecnología. Para aumentar la efectividad de las similitudes informáticas entre imágenes, los enfoques de hash se convirtieron en el foco de los programadores. De hecho, el aprendizaje profundo en los últimos años se ha considerado la columna vertebral del análisis de imágenes utilizando una red neuronal convolucional (CNN). El documento proporciona una encuesta sobre el último trabajo realizado en el campo de la recuperación de imágenes. Varias técnicas han aparecido en este campo. Sin embargo, la más común de estas técnicas es utilizar la codificación hash basada en redes neuronales, que se puede clasificar en tres clases principales: técnicas supervisadas, no supervisadas y semi-supervisadas de acuerdo con el método de aprendizaje de cada técnica. Se revisan los trabajos relacionados más importantes que aparecen en la literatura y se han realizado comparaciones constructivas para mostrar las fortalezas y limitaciones de varias técnicas. Palabras clave: recuperación de imágenes, aprendizaje profundo, red neuronal convolucional (CNN), técnicas de hash.
Computers & Electrical Engineering, Sep 1, 2022
Due to the COVID-19 epidemic and the curfew caused by it, many people have sought to find an ADPS... more Due to the COVID-19 epidemic and the curfew caused by it, many people have sought to find an ADPS on the internet in the last few years. This hints to a new age of medical treatment, all the more so if the number of internet users continues to expand. As a result, automatic illness prediction online applications have attracted the interest of a large number of researchers worldwide. This work aims to develop and implement an automated illness prediction system based on speech. The system will be designed to forecast the sort of ailment a patient is suffering from based on his voice, but this was not feasible during the trial, therefore the diseases were divided into three categories (painful, light pain and psychological pain), and then the diagnose process were implemented accordingly. The medical dataset named “speech, transcription, and intent” served as the baseline for this study. The smoothness, MFCC, and SCV properties were used in this work, which demonstrated their high representation to human being medical situations. The noise reduction forward-backward filter was used to eliminate noise from wave files captured online in order to account for the high level of noise seen in the deployed dataset. For this study, a hybrid feature selection method was created and built that combined the output of a genetic algorithm (GA) with the inputs of a NN algorithm. Classification was performed using SVM, neural network, and GMM. The greatest results obtained were 94.55% illness classification accuracy in terms of SVM. The results showed that diagnosing illness through speech is a difficult process, especially when diagnosing each type of illness separately, but when grouping the different illness types into groups, depending on the amount of pain and the psychological situation of the patient, the results were much higher.
Advances in intelligent systems and computing, Sep 9, 2018
Several metaheuristic algorithms and improvements to the existing ones have been presented over t... more Several metaheuristic algorithms and improvements to the existing ones have been presented over the years. Most of these algorithms were inspired either by nature or the behavior of certain swarms, such as birds, ants, bees, or even bats. These algorithms have two major components, which are exploration and exploitation. The interaction of these components can have a significant influence on the efficiency of the metaheuristics. Meanwhile, there are basically no guiding principles on how to strike a balance between these two components. This study, therefore, proposes a new multi-swarm-based balancing mechanism for keeping a balancing between the exploration and exploitation attributes of metaheuristics. The new approach is inspired by the phenomenon of the leadership scenario among a group of people (a group of people being governed by a selected leader(s)). These leaders communicate in a meeting room, and the overall best leader makes the final decision. The simulation aspect of the study considered several benchmark functions and compared the performance of the suggested algorithm to that of the standard PSO (SPSO) in terms of efficiency.
International Journal of Online and Biomedical Engineering (iJOE)
Deep learning and its variant techniques have surpassed classical machine algorithms due to their... more Deep learning and its variant techniques have surpassed classical machine algorithms due to their high performance gaining remarkable results and are used in a broad range of applications. However, adopting deep learning models over the cloud introduces privacy and security issues for data owners and model owners, including computational inefficiency, expansion in ciphertext, error accumulation, security and usability trade-offs, and deep learning model attacks. With homomorphic encryption, computations on encrypted data can be performed without disclosing its content. This research examines the basic concepts of homomorphic encryption limitations, benefits, weaknesses, possible applications, and development tools concentrating on neural networks. Additionally, we looked at systems that integrate neural networks with homomorphic encryption in order to maintain privacy. Furthermore, we classify modifications made on neural network models and architectures that make them computable vi...
Journal of Intelligent Systems
The supply and storage of drugs are critical components of the medical industry and distribution.... more The supply and storage of drugs are critical components of the medical industry and distribution. The shelf life of most medications is predetermined. When medicines are supplied in large quantities it is exceeding actual need, and long-term drug storage results. If demand is lower than necessary, this has an impact on consumer happiness and medicine marketing. Therefore, it is necessary to find a way to predict the actual quantity required for the organization’s needs to avoid material spoilage and storage problems. A mathematical prediction model is required to assist any management in achieving the required availability of medicines for customers and safe storage of medicines. Artificial intelligence applications and predictive modeling have used machine learning (ML) and deep learning algorithms to build prediction models. This model allows for the optimization of inventory levels, thus reducing costs and potentially increasing sales. Various measures, such as mean squared error...
Expert Systems
Voice pathology diagnosis requires extracting significant features from voice signals, and classi... more Voice pathology diagnosis requires extracting significant features from voice signals, and classical machine learning models can overfit to the training data, which can cause difficult issues and pose challenges. The study aimed to develop a reliable and efficient system for identifying voice pathologies utilizing the long short‐term memory (LSTM) method. The study combined unique feature sets such as the mel frequency cepstral coefficients (MFCCs), zero crossing rate (ZCR), and mel spectrograms, which have not been used together in previous works. Voice pathology identification improved the accuracy rate using the LSTM approach on the Saarbruecken voice database (SVD) samples. The best results achieved by the proposed system showed an accuracy rate of 99.3% for /u/ vowel samples in neutral pitch, 99.2% for /a/ vowel samples in high pitch, 99% for /i/ vowel samples in neutral pitch, and 99.2% for sentence samples. The experimental results were evaluated utilizing accuracy, precision...
Proceedings of the IEEE-EMBS Special Topic Conference on Molecular, Cellular and Tissue Engineering
An area of significant potential in Micro-Electro-Mechanical Systems (MEMS)-related technology is... more An area of significant potential in Micro-Electro-Mechanical Systems (MEMS)-related technology is the construction of microelectrode arrays that can be used to preferentially kill nearby biological cells while leaving cells more distant from the array intact. This paper details some of the basic considerations related to this research area, and goes on to explore electromagnetic field strength and its spatial variation around a simple simulated microelectrode array near a biological medium with σ and ε characteristics similar to that of human blood. Simulations were conducted at maximum field strengths of | 1.1 | volts at a frequency of 100 kHz and 1 MHz.
Journal of Intelligent Systems
Problem Recognizing written languages using symbols written in cuneiform is a tough endeavor due ... more Problem Recognizing written languages using symbols written in cuneiform is a tough endeavor due to the lack of information and the challenge of the process of tokenization. The Cuneiform Language Identification (CLI) dataset attempts to understand seven cuneiform languages and dialects, including Sumerian and six dialects of the Akkadian language: Old Babylonian, Middle Babylonian Peripheral, Standard Babylonian, Neo-Babylonian, Late Babylonian, and Neo-Assyrian. However, this dataset suffers from the problem of imbalanced categories. Aim Therefore, this article aims to build a system capable of distinguishing between several cuneiform languages and solving the problem of unbalanced categories in the CLI dataset. Methods Oversampling technique was used to balance the dataset, and the performance of machine learning algorithms such as Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), and deep learning such as deep neural networks (DNNs)...
Advances in Electrical and Electronic Engineering
Sensors
Recently, transfer learning approaches appeared to reduce the need for many classified medical im... more Recently, transfer learning approaches appeared to reduce the need for many classified medical images. However, these approaches still contain some limitations due to the mismatch of the domain between the source domain and the target domain. Therefore, this study aims to propose a novel approach, called Dual Transfer Learning (DTL), based on the convergence of patterns between the source and target domains. The proposed approach is applied to four pre-trained models (VGG16, Xception, ResNet50, MobileNetV2) using two datasets: ISIC2020 skin cancer images and ICIAR2018 breast cancer images, by fine-tuning the last layers on a sufficient number of unclassified images of the same disease and on a small number of classified images of the target task, in addition to using data augmentation techniques to balance classes and to increase the number of samples. According to the obtained results, it has been experimentally proven that the proposed approach has improved the performance of all ...
IAES International Journal of Artificial Intelligence (IJ-AI)
The importance of the multimedia information retrieval (MIR) is highlighted by the extensive amou... more The importance of the multimedia information retrieval (MIR) is highlighted by the extensive amount of the information on the internet. Image, audio, video, and text are all examples of the characteristics of the raw multimedia data. It is greatly challenging to represent a concept of human perception and how the machine-level language can grasp it (semantic gap of MIR). However, this paper aims to improve the information retrieval model that retrieves data from multimedia. This can be implemented by leveraging the use of variety of algorithms that go through training and testing to extract the model. One of these algorithms extracts text information based on the query language's nature as the vector space model (VSM) and the latent semantic index (LSI) were used. The other technique uses curvelet decomposition and statistic parameters like mean, standard deviation, and signal energy to recover these properties. Additionally, a discrete wavelet transforms (DWT) and signal charac...
المجلة العراقية للحاسبات والمعلوماتية, Sep 13, 2023
Medication supply and storage are essential components of the medical industry and distribution. ... more Medication supply and storage are essential components of the medical industry and distribution. Most medications have a predetermined expiration date. When the demand is met in large quantities that exceed the actual need, this leads to the accumulation of medicines in the stores, and this leads to the expiration of the materials. If demand is too low, this will have an impact on consumer happiness and drug marketing. Therefore, it is necessary to find a way to predict the actual quantity required for the organization's needs to avoid material spoilage and storage problems. A mathematical prediction model is required to assist any management in achieving the required availability of medicines for customers and safe storage of medicines. The research question is to design a system based on deep learning that can predict the amount of drugs required with high efficiency and accuracy based on the chronology of previous years.Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Bidirectional LSTM, and Gated Recurrent Unit (GRU) are used to build prediction models. Those models allow for the optimization of inventory levels, thus reducing costs and potentially increasing sales. Various measures such as mean squared error (MSE), mean absolute squared error (MASE), root mean squared error (RMSE), and others are used to evaluate the prediction models. RNN model achieved the best result with MSE: 0.019 MAE: 0.102, RMSE: 0.0.
At its inception, cuneiform signs were designed to represent sounds, enabling the recording of sp... more At its inception, cuneiform signs were designed to represent sounds, enabling the recording of spoken language. This served to preserve language for posterity, allowing thoughts and concepts to be conveyed in written form. Among the earliest known forms of writing is cuneiform, etched onto clay tablets using a reed stylus with a wedge-shaped tip, hence its name [1]. Presently, 'cuneiform' denotes a writing system characterized by wedge-shaped markings. While numerous transliterated Akkadian texts that can be accessed for free in digitized corpora; however, there are a much smaller number of texts that have been phonologically transcribed. Employed for three millennia, cuneiform script emerged in the late fourth millennium BCE and persisted until the first century CE, producing a diverse array of texts by scribes. Though primarily etched onto clay, some found their place on wax-coated stone, metal, or wooden boards. Over a million cuneiform texts have been uncovered to date in a large Near Eastern region stretching from Anatolia to Iran and northern Iraq to Bahrain and Egypt [2]. This region includes a portion of the Arabian Peninsula and the Persian Gulf. The amalgamation of wedges gives rise to an array of cuneiform signs, given that the script is created by pressing a stylus into fresh clay. Cuneiform was developed in ancient Mesopotamia, the present methodology involves manual reconstruction of missing data [3]. Executed by a select group proficient in Akkadian and cuneiform, this laborintensive process seeks to recover obscured lines in scientific, lexical, literary, and religious texts through comparison with intact tablets and parallel passages on deteriorated ones. Fortunately, the availability of duplicate copies, facilitates this endeavor. Nevertheless, it necessitates profound familiarity with various text types and collections, and the suggested restorations by scholars often invite diverse interpretations [4]. Though the system's brilliance permits effortless replication, scribes grappled with memorizing myriad signs. This discrepancy arises from the labor-intensive nature of transcription, which offers limited aid in academically interpretating cuneiform texts. Numerous findings include literary works like the Epic of Gilgamesh, clay envelopes containing letters that were sealed with clay, and historical narratives. Remarkably, vast libraries, exemplified by the collection of Assyrian king Ashurbanipal (668-627 B.C.E.), have come to light [1]. An astronomical text dated to 75 CE is the most recent example ABSTRACT: The cuneiform script reveals some previously unknown aspects of our past. However, reading ancient clay tablets demands a substantial investment of time and persistent practice over a long period of time. As the fourth millennium came to a close, earlier recording methods gave way to the development of writingthe visual representation of spoken language. The first language to be transcribed in written form in Mesopotamia was Sumerian. Predominantly, the earliest tablets originate from the Uruk site in southern Mesopotamia, possibly marking its birthplace. Digitization cuneiform documents is imperative to boost research focused on the ancient Middle East. A few initiatives embarked upon this endeavor around the year 2000. Nonetheless, the digitization process is time-consuming due to the extensive volume of documents, and a dependable (semi) automatic methodology has yet to be developed. Given the antiquity of cuneiform script, recognizing cuneiform signs using real-world applications via two graph-based algorithms, each with complementary runtime characteristics, remains a manual procedure. Translating cuneiform proves to be a daunting task. Only in relatively recent times has grammar been established scientifically, while lexical challenges remain abundant and far from resolved. Furthermore, the majority of the Sumerian tablets have succumbed to the ravages of time, leaving behind only a handful of ancient depictions. Some of these old images have been preserved in a unique collection or in museums worldwide, allowing specialists to easily apply the sign detector to their cuneiform text studies. In this paper, we will discuss the categorization and analysis of clay tablets using a trained cuneiform model, employing artificial intelligence methodologies. Additionally, we will explore the methods employed, highlighting their strengths and weaknesses. Finally, we will propose potential directions for future research.
Expert Systems with Applications, 2009
... 2. The solved course timetabling problem. The definition of the solved problem given below wa... more ... 2. The solved course timetabling problem. The definition of the solved problem given below was introduced by Aladag and Hocaoglu (2007a). 2.1. Objectives and constraints of the problem. ... TR, Different types of rooms, m rt, Number of rooms of type r, F, Set of pre-assigned lessons ...
Lecture notes in networks and systems, 2022
Xinan Jiaotong Daxue Xuebao, Jun 30, 2019
Particle Swarm Optimization (PSO) is a very common algorithm in swarm intelligence algorithms. PS... more Particle Swarm Optimization (PSO) is a very common algorithm in swarm intelligence algorithms. PSO has been used to solve a lot of problems with one or more goals. Actuality, the multi-objectives improvement issues in all real life are combinatorial in nature. Therefore, PSO has been improved to be able to handle very large number of decision variables and reduce or decrease computational complexity. In this work, a chaos multi objective PSO algorithm is improved for solving discrete (binary) optimization issues with crossover operation. The developed Chaos Discrete Multi Objective PSO (CDMOPSO) algorithm is applied to pavement management problem for flexible pavement to get optimal maintenance and rehabilitation plan. The results shown that there is significant improvement in the solutions satisfying pavement conditions and maintenance cost goals. It is required to a very short time of execution by the improved algorithm to reach a very good solution. Also, comparing the convergence of solutions with the rest of the PSO algorithms, it has found that the suggested algorithm is better.
Journal of Personalized Medicine, Feb 18, 2022
Social Science Research Network, Sep 25, 2019
Analysis of image contents has become one of the important subjects in modern life. In order to r... more Analysis of image contents has become one of the important subjects in modern life. In order to recognize the images with efficient way, several techniques have appeared and periodically enhanced by the developers. Image retrieval becomes one of the main problems that face the computer society inside the revolution of technology. To increase the effectiveness of computing similarities between images, hashing approaches became the focusing of the programmers. Indeed, deep learning in the past few years has been considered the backbone of image analysis using a convolutional neural network (CNN). The paper is providing a survey of the latest work carried out in the field of image retrieval. Several techniques have appeared in this field. However, the most common of these techniques are using neural network-based hash encoding, which can be categorized into three main classes: Supervised, unsupervised, and semi-supervised techniques according to each technique's learning method. The most important related works appeared in the literature are reviewed and constructive comparisons have been done to show the strengths and limitations of various techniques. Keywords: Image Retrieval, Deep Learning, Convolutional Neural Network (CNN), Hashing Techniques. RESUMEN / El análisis del contenido de la imagen se ha convertido en uno de los temas importantes en la vida moderna. Para reconocer las imágenes de manera eficiente, han aparecido varias técnicas que los desarrolladores han mejorado periódicamente. La recuperación de imágenes se convierte en uno de los principales problemas que enfrenta la sociedad informática dentro de la revolución de la tecnología. Para aumentar la efectividad de las similitudes informáticas entre imágenes, los enfoques de hash se convirtieron en el foco de los programadores. De hecho, el aprendizaje profundo en los últimos años se ha considerado la columna vertebral del análisis de imágenes utilizando una red neuronal convolucional (CNN). El documento proporciona una encuesta sobre el último trabajo realizado en el campo de la recuperación de imágenes. Varias técnicas han aparecido en este campo. Sin embargo, la más común de estas técnicas es utilizar la codificación hash basada en redes neuronales, que se puede clasificar en tres clases principales: técnicas supervisadas, no supervisadas y semi-supervisadas de acuerdo con el método de aprendizaje de cada técnica. Se revisan los trabajos relacionados más importantes que aparecen en la literatura y se han realizado comparaciones constructivas para mostrar las fortalezas y limitaciones de varias técnicas. Palabras clave: recuperación de imágenes, aprendizaje profundo, red neuronal convolucional (CNN), técnicas de hash.
Computers & Electrical Engineering, Sep 1, 2022
Due to the COVID-19 epidemic and the curfew caused by it, many people have sought to find an ADPS... more Due to the COVID-19 epidemic and the curfew caused by it, many people have sought to find an ADPS on the internet in the last few years. This hints to a new age of medical treatment, all the more so if the number of internet users continues to expand. As a result, automatic illness prediction online applications have attracted the interest of a large number of researchers worldwide. This work aims to develop and implement an automated illness prediction system based on speech. The system will be designed to forecast the sort of ailment a patient is suffering from based on his voice, but this was not feasible during the trial, therefore the diseases were divided into three categories (painful, light pain and psychological pain), and then the diagnose process were implemented accordingly. The medical dataset named “speech, transcription, and intent” served as the baseline for this study. The smoothness, MFCC, and SCV properties were used in this work, which demonstrated their high representation to human being medical situations. The noise reduction forward-backward filter was used to eliminate noise from wave files captured online in order to account for the high level of noise seen in the deployed dataset. For this study, a hybrid feature selection method was created and built that combined the output of a genetic algorithm (GA) with the inputs of a NN algorithm. Classification was performed using SVM, neural network, and GMM. The greatest results obtained were 94.55% illness classification accuracy in terms of SVM. The results showed that diagnosing illness through speech is a difficult process, especially when diagnosing each type of illness separately, but when grouping the different illness types into groups, depending on the amount of pain and the psychological situation of the patient, the results were much higher.
Advances in intelligent systems and computing, Sep 9, 2018
Several metaheuristic algorithms and improvements to the existing ones have been presented over t... more Several metaheuristic algorithms and improvements to the existing ones have been presented over the years. Most of these algorithms were inspired either by nature or the behavior of certain swarms, such as birds, ants, bees, or even bats. These algorithms have two major components, which are exploration and exploitation. The interaction of these components can have a significant influence on the efficiency of the metaheuristics. Meanwhile, there are basically no guiding principles on how to strike a balance between these two components. This study, therefore, proposes a new multi-swarm-based balancing mechanism for keeping a balancing between the exploration and exploitation attributes of metaheuristics. The new approach is inspired by the phenomenon of the leadership scenario among a group of people (a group of people being governed by a selected leader(s)). These leaders communicate in a meeting room, and the overall best leader makes the final decision. The simulation aspect of the study considered several benchmark functions and compared the performance of the suggested algorithm to that of the standard PSO (SPSO) in terms of efficiency.
International Journal of Online and Biomedical Engineering (iJOE)
Deep learning and its variant techniques have surpassed classical machine algorithms due to their... more Deep learning and its variant techniques have surpassed classical machine algorithms due to their high performance gaining remarkable results and are used in a broad range of applications. However, adopting deep learning models over the cloud introduces privacy and security issues for data owners and model owners, including computational inefficiency, expansion in ciphertext, error accumulation, security and usability trade-offs, and deep learning model attacks. With homomorphic encryption, computations on encrypted data can be performed without disclosing its content. This research examines the basic concepts of homomorphic encryption limitations, benefits, weaknesses, possible applications, and development tools concentrating on neural networks. Additionally, we looked at systems that integrate neural networks with homomorphic encryption in order to maintain privacy. Furthermore, we classify modifications made on neural network models and architectures that make them computable vi...
Journal of Intelligent Systems
The supply and storage of drugs are critical components of the medical industry and distribution.... more The supply and storage of drugs are critical components of the medical industry and distribution. The shelf life of most medications is predetermined. When medicines are supplied in large quantities it is exceeding actual need, and long-term drug storage results. If demand is lower than necessary, this has an impact on consumer happiness and medicine marketing. Therefore, it is necessary to find a way to predict the actual quantity required for the organization’s needs to avoid material spoilage and storage problems. A mathematical prediction model is required to assist any management in achieving the required availability of medicines for customers and safe storage of medicines. Artificial intelligence applications and predictive modeling have used machine learning (ML) and deep learning algorithms to build prediction models. This model allows for the optimization of inventory levels, thus reducing costs and potentially increasing sales. Various measures, such as mean squared error...
Expert Systems
Voice pathology diagnosis requires extracting significant features from voice signals, and classi... more Voice pathology diagnosis requires extracting significant features from voice signals, and classical machine learning models can overfit to the training data, which can cause difficult issues and pose challenges. The study aimed to develop a reliable and efficient system for identifying voice pathologies utilizing the long short‐term memory (LSTM) method. The study combined unique feature sets such as the mel frequency cepstral coefficients (MFCCs), zero crossing rate (ZCR), and mel spectrograms, which have not been used together in previous works. Voice pathology identification improved the accuracy rate using the LSTM approach on the Saarbruecken voice database (SVD) samples. The best results achieved by the proposed system showed an accuracy rate of 99.3% for /u/ vowel samples in neutral pitch, 99.2% for /a/ vowel samples in high pitch, 99% for /i/ vowel samples in neutral pitch, and 99.2% for sentence samples. The experimental results were evaluated utilizing accuracy, precision...
Proceedings of the IEEE-EMBS Special Topic Conference on Molecular, Cellular and Tissue Engineering
An area of significant potential in Micro-Electro-Mechanical Systems (MEMS)-related technology is... more An area of significant potential in Micro-Electro-Mechanical Systems (MEMS)-related technology is the construction of microelectrode arrays that can be used to preferentially kill nearby biological cells while leaving cells more distant from the array intact. This paper details some of the basic considerations related to this research area, and goes on to explore electromagnetic field strength and its spatial variation around a simple simulated microelectrode array near a biological medium with σ and ε characteristics similar to that of human blood. Simulations were conducted at maximum field strengths of | 1.1 | volts at a frequency of 100 kHz and 1 MHz.
Journal of Intelligent Systems
Problem Recognizing written languages using symbols written in cuneiform is a tough endeavor due ... more Problem Recognizing written languages using symbols written in cuneiform is a tough endeavor due to the lack of information and the challenge of the process of tokenization. The Cuneiform Language Identification (CLI) dataset attempts to understand seven cuneiform languages and dialects, including Sumerian and six dialects of the Akkadian language: Old Babylonian, Middle Babylonian Peripheral, Standard Babylonian, Neo-Babylonian, Late Babylonian, and Neo-Assyrian. However, this dataset suffers from the problem of imbalanced categories. Aim Therefore, this article aims to build a system capable of distinguishing between several cuneiform languages and solving the problem of unbalanced categories in the CLI dataset. Methods Oversampling technique was used to balance the dataset, and the performance of machine learning algorithms such as Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), and deep learning such as deep neural networks (DNNs)...
Advances in Electrical and Electronic Engineering
Sensors
Recently, transfer learning approaches appeared to reduce the need for many classified medical im... more Recently, transfer learning approaches appeared to reduce the need for many classified medical images. However, these approaches still contain some limitations due to the mismatch of the domain between the source domain and the target domain. Therefore, this study aims to propose a novel approach, called Dual Transfer Learning (DTL), based on the convergence of patterns between the source and target domains. The proposed approach is applied to four pre-trained models (VGG16, Xception, ResNet50, MobileNetV2) using two datasets: ISIC2020 skin cancer images and ICIAR2018 breast cancer images, by fine-tuning the last layers on a sufficient number of unclassified images of the same disease and on a small number of classified images of the target task, in addition to using data augmentation techniques to balance classes and to increase the number of samples. According to the obtained results, it has been experimentally proven that the proposed approach has improved the performance of all ...
IAES International Journal of Artificial Intelligence (IJ-AI)
The importance of the multimedia information retrieval (MIR) is highlighted by the extensive amou... more The importance of the multimedia information retrieval (MIR) is highlighted by the extensive amount of the information on the internet. Image, audio, video, and text are all examples of the characteristics of the raw multimedia data. It is greatly challenging to represent a concept of human perception and how the machine-level language can grasp it (semantic gap of MIR). However, this paper aims to improve the information retrieval model that retrieves data from multimedia. This can be implemented by leveraging the use of variety of algorithms that go through training and testing to extract the model. One of these algorithms extracts text information based on the query language's nature as the vector space model (VSM) and the latent semantic index (LSI) were used. The other technique uses curvelet decomposition and statistic parameters like mean, standard deviation, and signal energy to recover these properties. Additionally, a discrete wavelet transforms (DWT) and signal charac...