Skip to main content

Shohel Sayeed

Followers

113

Following

85

Co-authors

3

Mentions

1

Public Views

Dr. Shohel has been a member of Multimedia University since 2001, and now he serves as a Professor of the Faculty of Information Science and Technology. Dr. Shohel’s core research interest is in the area of Biometrics, information security, image and signal processing, pattern recognition and classification. Till date, he has published over 40 research papers in international peer-reviewed journals and international conference proceedings as a result of his research work. His research works have been accepted by journals such as IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), International Journal of Pattern Recognition and Artificial Intelligence (IJPRI), Discrete Dynamics in Nature and Society (DDNS). These papers have been cited in various international journals and conferences a number of times. Several of his findings have been presented in a number of well recognized IEEE conferences such as ICSP2006, ICIAS2007, ITSIM2008, CSECS2009 and ITSIM2010. He has been appointed technical paper reviewer for Journal of Pattern Recognition Letters, IEEE Transaction on Neural Netwroks, IEEE Transactions on Automation Science and Engineering, Journal of Computer Methods and Programs in Biomedicine and International Journal of Computer Theory and Engineering. He has also been invited to review technical papers for several international conferences. In recognition of his professional contribution, he has obtained recognition as a member of IEEE Computer Society, IEEE Communication Society and International Association of Computer Science and Information Technology (IACSIT).

less

University of California, Santa Barbara

University of British Columbia

University College Cork

Suryadiputra Liawatimena

BINUS University

Ucla

Sharif University of Technology

Roshan Chitrakar

Nepal College of Information Technology

Graduate Center of the City University of New York

Juraj Marušiak

Slovak Academy of Sciences

Maria V. Stanyukovich

Russian academy of sciences, Peter the Great Museum of Antropology and Ethnography

Interests

Uploads

Papers by Shohel Sayeed

Heat stroke prediction: a perspective from the internet of things and machine learning approach

International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering, Jun 1, 2024

Heat stroke prediction: a perspective from the internet of things and machine learning approach

by International Journal of Electrical and Computer Engineering (IJECE) and Shohel Sayeed

International Journal of Electrical and Computer Engineering (IJECE), 2024

With the increasing occurrence of heat-related illnesses due to rising temperatures worldwide, th... more With the increasing occurrence of heat-related illnesses due to rising temperatures worldwide, there is a need for effective detection and prediction systems to mitigate the risks. Heat stroke, a life-threatening condition occurs when the body's temperature exceeds 104 °F (40 °C). It can happen due to prolonged exposure to temperatures. When the body struggles to cool itself down adequately. The internet of things (IoT) and machine learning (ML) are two advancing technologies that have the potential to revolutionize industries and enhance our lives in numerous ways. Currently, monitoring devices are primarily used to diagnose when individuals suffering from heatstroke are at the location. This paper delves into the exploration of utilizing the IoT and ML algorithms to predict heat strokes. It reviews existing studies in this field, focusing on how IoT has been deployed and the application of machine learning techniques. The research aims to define the integration of IoT devices and ML algorithms that has a great potential to detect and predict heat-related illnesses such as heat stroke at an early stage.

Design new hybrid system using differential phase shift keying (DPSK) modulation technique for optical access network

This research work focuses to design Hybrid OCDMA over WDM system with advance modulation techniq... more This research work focuses to design Hybrid OCDMA over WDM system with advance modulation techniques DPSK which will improve the performance for long distance fiber length. Hybrid system merges with different types of techniques like OCDMA and WDM. Modified double weight (MDW) code is proposed as the signature address code. Multiuser interference and increase the bit-error-rate might be condense satisfactory by this Hybrid System. OptiSystem 12.0 simulation software to be employs for comparing two different Bit Rate 622 Mb/s and 1 Gb/s. In this paper, spectral direct detection technique has been used. Simulation results illustrate that hybrid-DPSK system yield better performance and enhance the network competence.

Glove-Based Approach to Online Signature Verification

IEEE Transactions on Pattern Analysis and Machine Intelligence, Jun 1, 2008

Utilizing the multiple degrees of freedom offered by the data glove for each finger and the hand,... more Utilizing the multiple degrees of freedom offered by the data glove for each finger and the hand, a novel online signature verification system using the Singular Value Decomposition (SVD) numerical tool for signature classification and verification is presented. The proposed technique is based on the Singular Value Decomposition in finding r singular vectors sensing the maximal energy of glove data matrix A A A A, called principal subspace, so the effective dimensionality of A A A A can be reduced. Having modeled the data glove signature through its r-principal subspace, signature authentication is performed by finding the angles between the different subspaces. A demonstration of the data glove is presented as an effective highbandwidth data entry device for signature verification. This SVD-based signature verification technique is tested and its performance is shown to be able to produce Equal Error Rate (EER) of less than 2.37 percent.

Glove-based Approach To Dynamic Signature Verification

The purpose of this research is to develop a novel, accurate and efficient on-line signature veri... more

Low Latency, Area and Optimal Power Hybrid Lightweight Cryptography Authentication Scheme for Internet of Things Applications

Wireless Personal Communications, May 4, 2022

Modeling Consumers’ Usage Intention of Augmented Reality in Online Buying Context: Empirical Setting with Measurement Development

Journal of Global Marketing

Comprehensive Review on IoT: Indoor Objects Location Tracking

Solid State Technology, Feb 29, 2020

Internet of Things (IoT) is improving the quality of modern life. IoT-based objects tracking is a... more Internet of Things (IoT) is improving the quality of modern life. IoT-based objects tracking is a crying need for a smart indoor environment. There are several technologies to identify the location of particular objects. However, most of the technologies used to identify object locations at outdoor; GPS is the most renowned outdoor location tracking technology that directly depends on satellites. Furthermore, it is unable to determine accurate location of indoor objects. Several indoor location tracking systems have been implemented based on different techniques. Unfortunately, these technologies performance are not up to the mark, because of limitations and dependency. In this situation, research on indoor object tracking with IoT is a hot topic. This review provides a valued direction from existing techniques, for an accurate IoT-based indoor location tracking. The outcomes of the review might be beneficial to selecting appropriate directions for reliable indoor location tracking

Poverty Prediction Using Machine Learning Approach

Xinan Jiaotong Daxue Xuebao, Feb 28, 2022

The incidence of poverty is not a taboo topic. In fact, it happens in every country worldwide whe... more The incidence of poverty is not a taboo topic. In fact, it happens in every country worldwide where the policymakers and governments struggle to reduce their country’s poverty rate. However, the existing ways of finding the right targeted impoverished group to provide economic aid are often flawed because of multiple issues such as data transparency and incorrect, redundant, or imbalanced data. Even though there are solutions available for the problems, some of them are expensive and require a lot of human intervention efforts. Thus, there is an alternative solution to use machine learning approach to forecasting and providing cost efficiency and effective solution to address poverty issues. This research utilizes machine learning model which comprises several pre-processing techniques for data cleaning, feature engineering and then followed by three different regression methods for predictive analytics. The proposed machine learning model is evaluated using Costa Rican household poverty dataset and the experimental results indicate that out of three different regression models used, Random Forest Regressor has the best result in terms of R2 and RMSE score which are 0.9462 well-fitted and 0.2591 error rate, respectively. Moreover, we analyzed the importance of individual features for poverty prediction using SHAP value based on summary plot analysis.

Plant identification based on leaf shape and texture pattern using Local Graph Structure (LGS)

Australian journal of basic and applied sciences, 2013

Plant identification is the process of matching a specimen plant to a known plant group. Most of ... more Plant identification is the process of matching a specimen plant to a known plant group. Most of the people identify plants visually (morphologically). The natural key systems use morphological characteristics that can be compared with known databases to find the plants' genus. Characteristics observed to include general character, structures of stems, roots, leaves and flowers. People who are not intimately familiar with particular species require assistance in their identification. In this context, we proposed an approach called local graph structure to identify and classify the plant species based on their leaf shape and texture pattern. The experimental results have shown a significant level of accuracy of the plant species identification.

Smartic: A smart tool for Big Data analytics and IoT

F1000Research, Jan 7, 2022

The Internet of Things (IoT) is leading the physical and digital world of technology to converge.... more The Internet of Things (IoT) is leading the physical and digital world of technology to converge. Real-time and massive scale connections produce a large amount of versatile data, where Big Data comes into the picture. Big Data refers to large, diverse sets of information with dimensions that go beyond the capabilities of widely used database management systems, or standard data processing software tools to manage within a given limit. Almost every big dataset is dirty and may contain missing data, mistyping, inaccuracies, and many more issues that impact Big Data analytics performances. One of the biggest challenges in Big Data analytics is to discover and repair dirty data; failure to do this can lead to inaccurate analytics results and unpredictable conclusions. We experimented with different missing value imputation techniques and compared machine learning (ML) model performances with different imputation methods. We propose a hybrid model for missing value imputation combining ML and sample-based statistical techniques. Furthermore, we continued with the best missing value inputted dataset, chosen based on ML model performance for feature engineering and hyperparameter tuning. We used k-means clustering and principal component analysis. Accuracy, the evaluated outcome, improved dramatically and proved that the XGBoost model gives very high accuracy at around 0.125 root mean squared logarithmic error (RMSLE). To overcome overfitting, we used K-fold cross-validation.

A Framework of Modified Adaptive Fuzzy Inference Engine (MAFIE) and Its Application

This paper introduces a complete framework of Modified Adaptive Fuzzy Inference Engine (MAFIE) an... more This paper introduces a complete framework of Modified Adaptive Fuzzy Inference Engine (MAFIE) and its application. The fuzzy with hybridization schemes has become of research interest in versatile applications over the past decade. The fuzzy hybridizations models are quite popular among practitioners or researchers in various advanced promising fields to help solve problems with a small number of inputs. However, there are limitations faced by all popular fuzzy systems when they are applied to systems with a large number of inputs. A modified apriori algorithm technique is utilized to reduce a minimal set of decision rules based on input-output dataset. A TSK type fuzzy inference system is constructed by the automatic generation of membership functions and fuzzy rules by the hybrid fuzzy clustering (Fuzzy C-Means and Subtractive Clustering) and apriori algorithms techniques, respectively. The generated adaptive fuzzy inference engine is adjusted by the least-square estimator and a conjugate gradient descent algorithm towards better performance with a minimal set of fuzzy rules. The proposed MAFIE is able to reduce the number of fuzzy rules which increases exponentially when large input dimensions are involved. The performance of the proposed MAFIE is compared with other existing models when applied to pattern classification schemes using Fisher's Iris and Wisconsin Breast Cancer benchmark datasets. The results are shown to be very competitive and MAFIE is ready for high dimension practical applications.

Classification Algorithms in Human Activity Recognition using Smartphones

Zenodo (CERN European Organization for Nuclear Research), Aug 27, 2012

Rapid advancement in computing technology brings computers and humans to be seamlessly integrated... more Rapid advancement in computing technology brings computers and humans to be seamlessly integrated in future. The emergence of smartphone has driven computing era towards ubiquitous and pervasive computing. Recognizing human activity has garnered a lot of interest and has raised significant researches' concerns in identifying contextual information useful to human activity recognition. Not only unobtrusive to users in daily life, smartphone has embedded built-in sensors that capable to sense contextual information of its users supported with wide range capability of network connections. In this paper, we will discuss the classification algorithms used in smartphone-based human activity. Existing technologies pertaining to smartphone-based researches in human activity recognition will be highlighted and discussed. Our paper will also present our findings and opinions to formulate improvement ideas in current researches' trends. Understanding research trends will enable researchers to have clearer research direction and common vision on latest smartphone-based human activity recognition area.

Deep supervised hashing for gait retrieval

F1000Research, Jun 22, 2022

Background: Gait recognition is perceived as the most promising biometric approach for future dec... more Background: Gait recognition is perceived as the most promising biometric approach for future decades especially because of its efficient applicability in surveillance systems. Due to recent growth in the use of gait biometrics across surveillance systems, the ability to rapidly search for the required data has become an emerging need. Therefore, we addressed the gait retrieval problem, which retrieves people with gaits similar to a query subject from a large-scale dataset. Methods: This paper presents the deep gait retrieval hashing (DGRH) model to address the gait retrieval problem for large-scale datasets. Our proposed method is based on a supervised hashing method with a deep convolutional network. We use the ability of the convolutional neural network (CNN) to capture the semantic gait features for feature representation and learn the compact hash codes with the compatible hash function. Therefore, our DGRH model combines gait feature learning with binary hash codes. In addition, the learning loss is designed with a classification loss function that learns to preserve similarity and a quantization loss function that controls the quality of the hash codes Results: The proposed method was evaluated against the CASIA-B, OUISIR-LP, and OUISIR-MVLP benchmark datasets and received the promising result for gait retrieval tasks. Conclusions: The end-to-end deep supervised hashing model is able to learn discriminative gait features and is efficient in terms of the storage memory and speed for gait retrieval.

Processing Large Volume of Biometric Data in the Hadoop Single Cluster Node Environment

American Journal of Applied Sciences, Dec 1, 2017

In big data evolution, the analysis of large scale data and scrutinizing the required vital infor... more In big data evolution, the analysis of large scale data and scrutinizing the required vital information becomes very demanding task. The emerging cloud platform promises and gives hope in handling the enormous volume of data. Hence, a new kind of methodology is required to tap the full potential of leveraging the big data analytics over the biometric data. In this work, we are going to deal with the integration of Hadoop, a map reduce framework with the infamous powerful computer vision library tool, Opencv. The proposed setup will comparatively analyze the large set of biometric data; such as face over the pseudo distributed environment. We test the capacity of our methodology with a different data set and analyze various computational parameters. The results show the proposed method is applicable for dealing in the real distributed environment.

Optimal design of wireless sensor network for providing Qos

With respect to the increase in demand of wireless sensor network in various real time Applicatio... more With respect to the increase in demand of wireless sensor network in various real time Application. We present WISEN algorithm which is specially designed for WSN providing multiple communication architecture and mechanism to improve the Quality of service and routing performance. WISEN focuses on functions such as node synchronization, node localization for improving QoS. The design of WISEN algorithm evaluated through network simulator with various protocols proves better performance in terms of packet loss, throughput and delay metrics. Collectively design of WISEN assumed optimal for Researchers.

IoT Based Indoor Object Location Tracking Solution

Performance Modelling of Priority Queuing Discipline under Self-Similar and Poisson Traffic

The priority queueing discipline plays a crucial role in the differentiated services (DiffServ) a... more The priority queueing discipline plays a crucial role in the differentiated services (DiffServ) architecture. Analytical modelling and performance evaluation of priority queueing systems have received significant research efforts in the telecommunication community. However, most existing studies have primarily focused on the analysis of such systems under either short range dependent (SRD) or long range dependent (LRD) traffic only. With the aim of investigating the impact of heterogeneous traffic on the design and performance of telecommunication networks, this paper presents an analytical model for priority queueing systems subject to heterogeneous LRD self-similar and SRD Poisson traffic. We extend the application of the generalized Schilder's theorem to deal with heterogeneous traffic and further develop the analytical upper and lower bounds for the queue length distributions of individual traffic flows. Through extensive comparisons between analytical bounds and simulation results, we validate the effectiveness and accuracy of the developed model.

A Review on IoT with Big Data Analytics

2021 9th International Conference on Information and Communication Technology (ICoICT), 2021

The Internet of Things (IoT) is a powerful and transformative force for the convergence of the ph... more The Internet of Things (IoT) is a powerful and transformative force for the convergence of the physical and digital world of technology. The IoT is connecting things, businesses, and people in real-time and on a massive scale. The IoT is the network of interconnected devices that contains actuators, sensors, electronics, software and connectivity which lets these things connect, interact and transfer data. Connected devices and software work in ways that produce massive amounts of data where Big Data comes into the picture. The terminology of Big Data represents diverse sets of information that are both very large and complex in nature. Big data offers a better way of managing and using a large amount of data with the opportunity to conduct deeper and richer analysis. Although the extensive number of big data analytics and IoT studies has been performed, the overlapping of the two fields of study creates various possibilities for thriving data analysis in the IoT environment. This paper presents a thorough review of the recent advancement of IoT with big data and analytics. We also make a review of the relationship between these fields. We present a discussion on the application area of IoT and big data analytics as well as the opportunities created by enabling analytics in an IoT system.

IoT Based Parking Guidance Using WKNN Algorithm

Drivers are frequently challenged to find parking due to the limited parking lots available in in... more Drivers are frequently challenged to find parking due to the limited parking lots available in indoor and outdoor parking areas. The number of available parking lots is usually less compared to the number of vehicles. Hence, drivers spend more time looking for vacant parking lots, which eventually delays their tasks, such as paying bills, attending a meeting, visiting a patient at a hospital, etc. In this paper, an intelligent parking system that utilizes indoor positioning using Wi-Fi RSSI fingerprinting and the Weighted K Nearest Neighbors algorithm with real-time navigation is proposed to alleviate the problem. The proposed method applies an IoT approach to estimate user location and detect vacant parking lots using sensors technology. Data from sensors are sent to an IoT cloud in real-time and displayed via an app, allowing users to acquire the route and get navigated to their desired parking spot with minimal effort. In addition, the proposed system is cost-efficient since it uses existing technologies and infrastructure.

Heat stroke prediction: a perspective from the internet of things and machine learning approach

International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering, Jun 1, 2024

Heat stroke prediction: a perspective from the internet of things and machine learning approach

by International Journal of Electrical and Computer Engineering (IJECE) and Shohel Sayeed

International Journal of Electrical and Computer Engineering (IJECE), 2024

With the increasing occurrence of heat-related illnesses due to rising temperatures worldwide, th... more With the increasing occurrence of heat-related illnesses due to rising temperatures worldwide, there is a need for effective detection and prediction systems to mitigate the risks. Heat stroke, a life-threatening condition occurs when the body's temperature exceeds 104 °F (40 °C). It can happen due to prolonged exposure to temperatures. When the body struggles to cool itself down adequately. The internet of things (IoT) and machine learning (ML) are two advancing technologies that have the potential to revolutionize industries and enhance our lives in numerous ways. Currently, monitoring devices are primarily used to diagnose when individuals suffering from heatstroke are at the location. This paper delves into the exploration of utilizing the IoT and ML algorithms to predict heat strokes. It reviews existing studies in this field, focusing on how IoT has been deployed and the application of machine learning techniques. The research aims to define the integration of IoT devices and ML algorithms that has a great potential to detect and predict heat-related illnesses such as heat stroke at an early stage.

Design new hybrid system using differential phase shift keying (DPSK) modulation technique for optical access network

This research work focuses to design Hybrid OCDMA over WDM system with advance modulation techniq... more This research work focuses to design Hybrid OCDMA over WDM system with advance modulation techniques DPSK which will improve the performance for long distance fiber length. Hybrid system merges with different types of techniques like OCDMA and WDM. Modified double weight (MDW) code is proposed as the signature address code. Multiuser interference and increase the bit-error-rate might be condense satisfactory by this Hybrid System. OptiSystem 12.0 simulation software to be employs for comparing two different Bit Rate 622 Mb/s and 1 Gb/s. In this paper, spectral direct detection technique has been used. Simulation results illustrate that hybrid-DPSK system yield better performance and enhance the network competence.

Glove-Based Approach to Online Signature Verification

IEEE Transactions on Pattern Analysis and Machine Intelligence, Jun 1, 2008

Utilizing the multiple degrees of freedom offered by the data glove for each finger and the hand,... more Utilizing the multiple degrees of freedom offered by the data glove for each finger and the hand, a novel online signature verification system using the Singular Value Decomposition (SVD) numerical tool for signature classification and verification is presented. The proposed technique is based on the Singular Value Decomposition in finding r singular vectors sensing the maximal energy of glove data matrix A A A A, called principal subspace, so the effective dimensionality of A A A A can be reduced. Having modeled the data glove signature through its r-principal subspace, signature authentication is performed by finding the angles between the different subspaces. A demonstration of the data glove is presented as an effective highbandwidth data entry device for signature verification. This SVD-based signature verification technique is tested and its performance is shown to be able to produce Equal Error Rate (EER) of less than 2.37 percent.

Glove-based Approach To Dynamic Signature Verification

The purpose of this research is to develop a novel, accurate and efficient on-line signature veri... more

Low Latency, Area and Optimal Power Hybrid Lightweight Cryptography Authentication Scheme for Internet of Things Applications

Wireless Personal Communications, May 4, 2022

Modeling Consumers’ Usage Intention of Augmented Reality in Online Buying Context: Empirical Setting with Measurement Development

Journal of Global Marketing

Comprehensive Review on IoT: Indoor Objects Location Tracking

Solid State Technology, Feb 29, 2020

Internet of Things (IoT) is improving the quality of modern life. IoT-based objects tracking is a... more Internet of Things (IoT) is improving the quality of modern life. IoT-based objects tracking is a crying need for a smart indoor environment. There are several technologies to identify the location of particular objects. However, most of the technologies used to identify object locations at outdoor; GPS is the most renowned outdoor location tracking technology that directly depends on satellites. Furthermore, it is unable to determine accurate location of indoor objects. Several indoor location tracking systems have been implemented based on different techniques. Unfortunately, these technologies performance are not up to the mark, because of limitations and dependency. In this situation, research on indoor object tracking with IoT is a hot topic. This review provides a valued direction from existing techniques, for an accurate IoT-based indoor location tracking. The outcomes of the review might be beneficial to selecting appropriate directions for reliable indoor location tracking

Poverty Prediction Using Machine Learning Approach

Xinan Jiaotong Daxue Xuebao, Feb 28, 2022

The incidence of poverty is not a taboo topic. In fact, it happens in every country worldwide whe... more The incidence of poverty is not a taboo topic. In fact, it happens in every country worldwide where the policymakers and governments struggle to reduce their country’s poverty rate. However, the existing ways of finding the right targeted impoverished group to provide economic aid are often flawed because of multiple issues such as data transparency and incorrect, redundant, or imbalanced data. Even though there are solutions available for the problems, some of them are expensive and require a lot of human intervention efforts. Thus, there is an alternative solution to use machine learning approach to forecasting and providing cost efficiency and effective solution to address poverty issues. This research utilizes machine learning model which comprises several pre-processing techniques for data cleaning, feature engineering and then followed by three different regression methods for predictive analytics. The proposed machine learning model is evaluated using Costa Rican household poverty dataset and the experimental results indicate that out of three different regression models used, Random Forest Regressor has the best result in terms of R2 and RMSE score which are 0.9462 well-fitted and 0.2591 error rate, respectively. Moreover, we analyzed the importance of individual features for poverty prediction using SHAP value based on summary plot analysis.

Plant identification based on leaf shape and texture pattern using Local Graph Structure (LGS)

Australian journal of basic and applied sciences, 2013

Plant identification is the process of matching a specimen plant to a known plant group. Most of ... more Plant identification is the process of matching a specimen plant to a known plant group. Most of the people identify plants visually (morphologically). The natural key systems use morphological characteristics that can be compared with known databases to find the plants' genus. Characteristics observed to include general character, structures of stems, roots, leaves and flowers. People who are not intimately familiar with particular species require assistance in their identification. In this context, we proposed an approach called local graph structure to identify and classify the plant species based on their leaf shape and texture pattern. The experimental results have shown a significant level of accuracy of the plant species identification.

Smartic: A smart tool for Big Data analytics and IoT

F1000Research, Jan 7, 2022

The Internet of Things (IoT) is leading the physical and digital world of technology to converge.... more The Internet of Things (IoT) is leading the physical and digital world of technology to converge. Real-time and massive scale connections produce a large amount of versatile data, where Big Data comes into the picture. Big Data refers to large, diverse sets of information with dimensions that go beyond the capabilities of widely used database management systems, or standard data processing software tools to manage within a given limit. Almost every big dataset is dirty and may contain missing data, mistyping, inaccuracies, and many more issues that impact Big Data analytics performances. One of the biggest challenges in Big Data analytics is to discover and repair dirty data; failure to do this can lead to inaccurate analytics results and unpredictable conclusions. We experimented with different missing value imputation techniques and compared machine learning (ML) model performances with different imputation methods. We propose a hybrid model for missing value imputation combining ML and sample-based statistical techniques. Furthermore, we continued with the best missing value inputted dataset, chosen based on ML model performance for feature engineering and hyperparameter tuning. We used k-means clustering and principal component analysis. Accuracy, the evaluated outcome, improved dramatically and proved that the XGBoost model gives very high accuracy at around 0.125 root mean squared logarithmic error (RMSLE). To overcome overfitting, we used K-fold cross-validation.

A Framework of Modified Adaptive Fuzzy Inference Engine (MAFIE) and Its Application

This paper introduces a complete framework of Modified Adaptive Fuzzy Inference Engine (MAFIE) an... more This paper introduces a complete framework of Modified Adaptive Fuzzy Inference Engine (MAFIE) and its application. The fuzzy with hybridization schemes has become of research interest in versatile applications over the past decade. The fuzzy hybridizations models are quite popular among practitioners or researchers in various advanced promising fields to help solve problems with a small number of inputs. However, there are limitations faced by all popular fuzzy systems when they are applied to systems with a large number of inputs. A modified apriori algorithm technique is utilized to reduce a minimal set of decision rules based on input-output dataset. A TSK type fuzzy inference system is constructed by the automatic generation of membership functions and fuzzy rules by the hybrid fuzzy clustering (Fuzzy C-Means and Subtractive Clustering) and apriori algorithms techniques, respectively. The generated adaptive fuzzy inference engine is adjusted by the least-square estimator and a conjugate gradient descent algorithm towards better performance with a minimal set of fuzzy rules. The proposed MAFIE is able to reduce the number of fuzzy rules which increases exponentially when large input dimensions are involved. The performance of the proposed MAFIE is compared with other existing models when applied to pattern classification schemes using Fisher's Iris and Wisconsin Breast Cancer benchmark datasets. The results are shown to be very competitive and MAFIE is ready for high dimension practical applications.

Classification Algorithms in Human Activity Recognition using Smartphones

Zenodo (CERN European Organization for Nuclear Research), Aug 27, 2012

Rapid advancement in computing technology brings computers and humans to be seamlessly integrated... more Rapid advancement in computing technology brings computers and humans to be seamlessly integrated in future. The emergence of smartphone has driven computing era towards ubiquitous and pervasive computing. Recognizing human activity has garnered a lot of interest and has raised significant researches' concerns in identifying contextual information useful to human activity recognition. Not only unobtrusive to users in daily life, smartphone has embedded built-in sensors that capable to sense contextual information of its users supported with wide range capability of network connections. In this paper, we will discuss the classification algorithms used in smartphone-based human activity. Existing technologies pertaining to smartphone-based researches in human activity recognition will be highlighted and discussed. Our paper will also present our findings and opinions to formulate improvement ideas in current researches' trends. Understanding research trends will enable researchers to have clearer research direction and common vision on latest smartphone-based human activity recognition area.

Deep supervised hashing for gait retrieval

F1000Research, Jun 22, 2022

Background: Gait recognition is perceived as the most promising biometric approach for future dec... more Background: Gait recognition is perceived as the most promising biometric approach for future decades especially because of its efficient applicability in surveillance systems. Due to recent growth in the use of gait biometrics across surveillance systems, the ability to rapidly search for the required data has become an emerging need. Therefore, we addressed the gait retrieval problem, which retrieves people with gaits similar to a query subject from a large-scale dataset. Methods: This paper presents the deep gait retrieval hashing (DGRH) model to address the gait retrieval problem for large-scale datasets. Our proposed method is based on a supervised hashing method with a deep convolutional network. We use the ability of the convolutional neural network (CNN) to capture the semantic gait features for feature representation and learn the compact hash codes with the compatible hash function. Therefore, our DGRH model combines gait feature learning with binary hash codes. In addition, the learning loss is designed with a classification loss function that learns to preserve similarity and a quantization loss function that controls the quality of the hash codes Results: The proposed method was evaluated against the CASIA-B, OUISIR-LP, and OUISIR-MVLP benchmark datasets and received the promising result for gait retrieval tasks. Conclusions: The end-to-end deep supervised hashing model is able to learn discriminative gait features and is efficient in terms of the storage memory and speed for gait retrieval.

Processing Large Volume of Biometric Data in the Hadoop Single Cluster Node Environment

American Journal of Applied Sciences, Dec 1, 2017

In big data evolution, the analysis of large scale data and scrutinizing the required vital infor... more In big data evolution, the analysis of large scale data and scrutinizing the required vital information becomes very demanding task. The emerging cloud platform promises and gives hope in handling the enormous volume of data. Hence, a new kind of methodology is required to tap the full potential of leveraging the big data analytics over the biometric data. In this work, we are going to deal with the integration of Hadoop, a map reduce framework with the infamous powerful computer vision library tool, Opencv. The proposed setup will comparatively analyze the large set of biometric data; such as face over the pseudo distributed environment. We test the capacity of our methodology with a different data set and analyze various computational parameters. The results show the proposed method is applicable for dealing in the real distributed environment.

Optimal design of wireless sensor network for providing Qos

With respect to the increase in demand of wireless sensor network in various real time Applicatio... more With respect to the increase in demand of wireless sensor network in various real time Application. We present WISEN algorithm which is specially designed for WSN providing multiple communication architecture and mechanism to improve the Quality of service and routing performance. WISEN focuses on functions such as node synchronization, node localization for improving QoS. The design of WISEN algorithm evaluated through network simulator with various protocols proves better performance in terms of packet loss, throughput and delay metrics. Collectively design of WISEN assumed optimal for Researchers.

IoT Based Indoor Object Location Tracking Solution

Performance Modelling of Priority Queuing Discipline under Self-Similar and Poisson Traffic

The priority queueing discipline plays a crucial role in the differentiated services (DiffServ) a... more The priority queueing discipline plays a crucial role in the differentiated services (DiffServ) architecture. Analytical modelling and performance evaluation of priority queueing systems have received significant research efforts in the telecommunication community. However, most existing studies have primarily focused on the analysis of such systems under either short range dependent (SRD) or long range dependent (LRD) traffic only. With the aim of investigating the impact of heterogeneous traffic on the design and performance of telecommunication networks, this paper presents an analytical model for priority queueing systems subject to heterogeneous LRD self-similar and SRD Poisson traffic. We extend the application of the generalized Schilder's theorem to deal with heterogeneous traffic and further develop the analytical upper and lower bounds for the queue length distributions of individual traffic flows. Through extensive comparisons between analytical bounds and simulation results, we validate the effectiveness and accuracy of the developed model.

A Review on IoT with Big Data Analytics

2021 9th International Conference on Information and Communication Technology (ICoICT), 2021

The Internet of Things (IoT) is a powerful and transformative force for the convergence of the ph... more The Internet of Things (IoT) is a powerful and transformative force for the convergence of the physical and digital world of technology. The IoT is connecting things, businesses, and people in real-time and on a massive scale. The IoT is the network of interconnected devices that contains actuators, sensors, electronics, software and connectivity which lets these things connect, interact and transfer data. Connected devices and software work in ways that produce massive amounts of data where Big Data comes into the picture. The terminology of Big Data represents diverse sets of information that are both very large and complex in nature. Big data offers a better way of managing and using a large amount of data with the opportunity to conduct deeper and richer analysis. Although the extensive number of big data analytics and IoT studies has been performed, the overlapping of the two fields of study creates various possibilities for thriving data analysis in the IoT environment. This paper presents a thorough review of the recent advancement of IoT with big data and analytics. We also make a review of the relationship between these fields. We present a discussion on the application area of IoT and big data analytics as well as the opportunities created by enabling analytics in an IoT system.

IoT Based Parking Guidance Using WKNN Algorithm

Drivers are frequently challenged to find parking due to the limited parking lots available in in... more Drivers are frequently challenged to find parking due to the limited parking lots available in indoor and outdoor parking areas. The number of available parking lots is usually less compared to the number of vehicles. Hence, drivers spend more time looking for vacant parking lots, which eventually delays their tasks, such as paying bills, attending a meeting, visiting a patient at a hospital, etc. In this paper, an intelligent parking system that utilizes indoor positioning using Wi-Fi RSSI fingerprinting and the Weighted K Nearest Neighbors algorithm with real-time navigation is proposed to alleviate the problem. The proposed method applies an IoT approach to estimate user location and detect vacant parking lots using sensors technology. Data from sensors are sent to an IoT cloud in real-time and displayed via an app, allowing users to acquire the route and get navigated to their desired parking spot with minimal effort. In addition, the proposed system is cost-efficient since it uses existing technologies and infrastructure.