Papers by Shahrulniza Musa
Sensors
Social media applications, such as Twitter and Facebook, allow users to communicate and share the... more Social media applications, such as Twitter and Facebook, allow users to communicate and share their thoughts, status updates, opinions, photographs, and videos around the globe. Unfortunately, some people utilize these platforms to disseminate hate speech and abusive language. The growth of hate speech may result in hate crimes, cyber violence, and substantial harm to cyberspace, physical security, and social safety. As a result, hate speech detection is a critical issue for both cyberspace and physical society, necessitating the development of a robust application capable of detecting and combating it in real-time. Hate speech detection is a context-dependent problem that requires context-aware mechanisms for resolution. In this study, we employed a transformer-based model for Roman Urdu hate speech classification due to its ability to capture the text context. In addition, we developed the first Roman Urdu pre-trained BERT model, which we named BERT-RU. For this purpose, we exploi...
Sensors
Social media applications, such as Twitter and Facebook, allow users to communicate and share the... more Social media applications, such as Twitter and Facebook, allow users to communicate and share their thoughts, status updates, opinions, photographs, and videos around the globe. Unfortunately, some people utilize these platforms to disseminate hate speech and abusive language. The growth of hate speech may result in hate crimes, cyber violence, and substantial harm to cyberspace, physical security, and social safety. As a result, hate speech detection is a critical issue for both cyberspace and physical society, necessitating the development of a robust application capable of detecting and combating it in real-time. Hate speech detection is a context-dependent problem that requires context-aware mechanisms for resolution. In this study, we employed a transformer-based model for Roman Urdu hate speech classification due to its ability to capture the text context. In addition, we developed the first Roman Urdu pre-trained BERT model, which we named BERT-RU. For this purpose, we exploi...
Sustainability
The Internet and Big Data expansion have motivated the requirement for more generous stockpiling ... more The Internet and Big Data expansion have motivated the requirement for more generous stockpiling to hold and share information. Against the current era of information, guaranteeing protection and security to individuals sending data to each other is of utmost importance. The only file type that is instantly and widely used is the image. Therefore, to secure transmission, it is necessary to develop a mechanism to safeguard user data transmission. Considering this thought, it is necessary to analyze the best file type of image for essential criteria of image steganography, such as Payload, Robustness, Imperceptibility, etc., to challenge the weakness of the current algorithms. The widely used image formats are PNG, TIFF, JPEG, BMP, and GIF, which is the cause of existing methods. However, in this case, the critical softness is the credibility of the steganography, which plays a vital role in these format images to ensure the end users communicate. In this paper, a single algorithm pro...
VIII INTERNATIONAL ANNUAL CONFERENCE “INDUSTRIAL TECHNOLOGIES AND ENGINEERING” (ICITE 2021)
Artificial Intelligence (AI) will shape human civilization differently because AI is increasingly... more Artificial Intelligence (AI) will shape human civilization differently because AI is increasingly playing a key role in our lives. AI is an Art that facilitate computer, robot or machine to mimic human behavior. Deep learning is a subfield of AI that focuses more on how machine work mimics the human brain. Deep learning is one of the recent, trending and emerging field of study and research. Deep learning is revolutionizing every field from robotics to medicine, and has provided an easy solution to many state-of-the-art problems. The recent developments in deep learning are virtual assistant, self-driving cars, news aggregation, natural language processing, digital marketing, visual recognition and image recognition etc. Despite many developments in deep learning, there still exist some challenges. Many researchers are working on deep learning, but most of them focus on field-specific issues, while this paper presents a brief survey on deeplearning advancements from the last few decades. This work gives a clear and concise direction to new researchers and academicians. This article thoroughly discusses Deep Neural Network (DNN); it also covers different topics, i.e. Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), including most of the other deep learning deep algorithms. This article also covers the open research issues and recent trends in deep learning.
2020 First International Conference of Smart Systems and Emerging Technologies (SMARTTECH)
The fifth generation of mobile networks unrolls with it, substantial increases to bandwidth and a... more The fifth generation of mobile networks unrolls with it, substantial increases to bandwidth and a tremendous expansion in the number of connections whilst reducing latency. In simple terms, the 5G technology implies faster connections and greater communication capability. The fifth-generation network (5G) rollout is the result of years of rigorous research, development and planning. Nonetheless, researchers underscore that 5G network may have flaws and vulnerabilities that exists within the technology. The integration of resource constrained IoT devices with 5G has created numerous threat surfaces. This paper discusses the vulnerability surrounding the 5G communication network, particularly highlighting critical issue relating to location-tracking and network downgrading. Threats inherited from the previous generation networks as well as the unique challenges specifically targeting 5G networks have been discussed. Also, this paper indicates the improvements needed to the security of 5G at the edge networks, typically for IoT users.
Journal of Information Technology Research, 2022
Software development through teams at different geographical locations is a trend of modern era, ... more Software development through teams at different geographical locations is a trend of modern era, which is not only producing good results without costing lot of money but also productive in relation to its cost, low risk and high return. This shift of perception of working in a group rather than alone is getting stronger day by day and has become an important planning tool and part of their business strategy. In this research classification approaches like SVM and K-NN have been implemented to classify the true positive events of global software development project risk according to Time, Cost and Resource. Comparative analysis has also been performed between these two algorithms to determine the highest accuracy algorithms. Results proved that Support Vector Machine (SVM) performed very well in case of Cost Related Risk and Resource Related Risk. Whereas, KNN is found superior to SVM for Time Related Risk.
International Journal of Electrical and Computer Engineering (IJECE), 2021
Information technology fields are now more dominated by artificial intelligence, as it is playing... more Information technology fields are now more dominated by artificial intelligence, as it is playing a key role in terms of providing better services. The inherent strengths of artificial intelligence are driving the companies into a modern, decisive, secure, and insight-driven arena to address the current and future challenges. The key technologies like cloud, internet of things (IoT), and software-defined networking (SDN) are emerging as future applications and rendering benefits to the society. Integrating artificial intelligence with these innovations with scalability brings beneficiaries to the next level of efficiency. Data generated from the heterogeneous devices are received, exchanged, stored, managed, and analyzed to automate and improve the performance of the overall system and be more reliable. Although these new technologies are not free of their limitations, nevertheless, the synthesis of technologies has been challenged and has put forth many challenges in terms of scala...
Kajian ini diadakan bertujuan untuk mengetahui samada kurikulum automasi di kursus diploma Teknol... more Kajian ini diadakan bertujuan untuk mengetahui samada kurikulum automasi di kursus diploma Teknologi Pemasangan dan Peralatan Elektrik (EEIT) di MFI memenuhi keperluan industri berciri automasi dan melengkapkan pelajar-pelajar dengan kemahiran serta kebolehan dalam bidang itu sebelum menceburi dunia pekerjaan. Lebih jauh lagi, ianya juga adalah untuk mengetahui samada kurikulum itu melengkapkan pelajar-pelajarnya dengan pengetahuan yang cukup berkaitan dengan teknologi terbaru dalam industri berciri automasi. Sampel kajian terdiri daripada pengurus latihan, jurutera atau penyelia di jabatan kejuruteraan di empat puluh buah industri terbabit di sekitar Bandar Baru Bangi. Instrumen kajian merupakan soal selidik yang dibina sendiri penyelidik. Hasil kajian telah mendapati bahawa secara umumnya, pengetahuan teorik, kemahiran dan tecnical know-how yang telah diajar dalam kurikulum automasi di kursus diploma EEIT memenuhi keperluan industri. Walaubagaimanapun adalah dicadangkan agar beber...
2016 International Conference on Information and Communication Technology (ICICTM), 2016
Numerous applications are running in a distributed environment in today's large networked wor... more Numerous applications are running in a distributed environment in today's large networked world. Corporations really need a mechanism to monitor their own application(s) running on remote devices. One such mechanism by Trusted Computing Group (TCG) called remote attestation that can monitor and verify trustworthiness of remote applications. In this regard, many solutions have been provided on how to monitor remote applications. However, It becomes quite challenging task, when applications are running on millions of devices and it becomes necessary for the corporates to verify all of the applications. In this paper we have provided an efficient, scalable and privacy preserved mechanism to tackle the scalability of all these kinds of verifications. Machine learning algorithms are incorporated as Hadoop/MapReduce functions on the public cloud. The rest of low CPU intensive and privacy preserved verifications are performed on the private cloud.
Proceedings of the First EAI International Conference on Computer Science and Engineering, 2017
Security is an important factor in today's IT infrastructure due to complex and vast variety of m... more Security is an important factor in today's IT infrastructure due to complex and vast variety of malware threats. One way to tackle these malware is via signaturebased techniques. However, this requires human effort in identification of threats and is not scalable. The second way is to detect malware via behavior-based reference monitor so called 'O-Day' malware. In this paper, we have optimized behavior-based technique for a specific use-case, based on today's enterprise requirement. We have built behavior-based lightweight reference monitor to measure and report a complete system call sequences as well as its arguments. The measurements are stored into Trusted Platform Module (TPM) protected location. The reference monitor splits the sequences of system calls and its arguments. Arguments and their verification is performed independent of each other via machine learning techniques. The behavior monitor is designed and developed on the core Linux Security Module (LSM). The same monitor is also designed and developed for Android-based platform via a newly built architecture called Android Security Module (ASM).
Lecture Notes in Computer Science, 2016
Measurement and reporting of dynamic behavior of a target application is a pressing issue in the ... more Measurement and reporting of dynamic behavior of a target application is a pressing issue in the Trusted Computing paradigm. Remote attestation is a part of trusted computing, which allows monitoring and verification of a complete operating system or a specific application by a remote party. Several static remote attestation techniques have been proposed in the past but most of the feasible ones are static in nature. However, such techniques cannot cater to dynamic attacks such as the infamous Heartbleed bug. Dynamic attestation offers a solution to this issue but is impractical due to the infeasibility of measurement and reporting of enormous runtime data. To an extent, it is possible to measure and report the dynamic behavior of a single application but not the complete operating system. The contribution of this paper is to provide the design and implementation of a scalable dynamic remote attestation mechanism that can measure and report multiple applications from different stakeholders simultaneously while ensuring privacy of each stakeholder. We have implemented our reference monitor and tested on Linux Kernel. We show through empirical results that this design is high scalable and feasible for a large number of stakeholders.
Proceedings of the 6th International Conference on Ubiquitous Information Management and Communication, 2012
This research work is based on the location prediction of wireless nodes with the terrains/clutte... more This research work is based on the location prediction of wireless nodes with the terrains/clutters considerations. Multi-condition Bayesian decision theory is applied for precision in selected locations. Currecnt research is the continution of our previous research work in which twelve terrains were proposed based on the atmospharic atteunation. In the first phase of paper, basic parameters such as receive signal
Increasing numbers of alerts produced by network intrusion detection systems (NIDS) have burdened... more Increasing numbers of alerts produced by network intrusion detection systems (NIDS) have burdened the job of security analysts especially in identifying and responding to them. The tasks of exploring and analysing large quantities of communication network security data are also difficult. This thesis studied the application of visualisation in combination with alerts classifier to make the exploring and understanding of network security alerts data faster and easier. The prototype software, NSAViz, has been developed to visualise and to provide an intuitive presentation of the network security alerts data using interactive 3D visuals with an integration of a false alert classifier. The needs analysis of this prototype was based on the suggested needs of network security analyst's tasks as seen in the literatures. The prototype software incorporates various projections of the alert data in 3D displays. The overview was plotted in a 3D plot named as "time series 3D AlertGraph...
Routing protocols are used to transmit the packets from the source to the destination node in mob... more Routing protocols are used to transmit the packets from the source to the destination node in mobile ad hoc networks. The intruders seek chance to pierce into the network and becomes a cause of malfunctioning in the network. These protocols are always prone to attacks. During the phases of routing in different types of protocols, each of the attack finds a way to degrade the performance of the routing protocols. The reactive routing protocols DSR and AODV have lot of similar features and so are considered in this study. In order to transmit the packets safely, a secured approach using triple factor has been proposed. This triple factor computes the trust by using the direct information then verifies the reputation by collecting the information from the neighbouring nodes called distributed factor and uses cryptographic algorithm to ensure security. And to ensure that there are routes available to perform the routing process, the reasons for such attacks are studied so as to re-integ...
The Internet of Things (IoT) is undergoing rapid growth in the IT industry, but, it continues to ... more The Internet of Things (IoT) is undergoing rapid growth in the IT industry, but, it continues to be associated with several security and privacy concerns as a result of its massive scale, decentralised topology, and resource-constrained devices. Blockchain (BC), a distributed ledger technology used in cryptocurrency has attracted significant attention in the realm of IoT security and privacy. However, adopting BC to IoT is not straightforward in most cases, due to overheads and delays caused by BC operations. In this paper, we apply a BC technology known as Hyperledgder Fabric, to an IoT network. This technology introduces an execute-order technique for transactions that separates the transaction execution from consensus, resulting in increased efficiency. We demonstrate that our proposed IoT-BC architecture is sufficiently secure with regard to fundamental security goals i.e., confidentiality, integrity, and availability. Finally, the simulation results are highlighted that shows t...
Distributed control systems (DCSs) and Supervisory Control and Data Acquisition (SCADA) systems a... more Distributed control systems (DCSs) and Supervisory Control and Data Acquisition (SCADA) systems are widely used in real time industry infrastructures such as water pumping stations, gas, and oil and other real time deployments. SCADA implementation within a cloud computing environment is new and beneficial for real time infrastructures. Using cloud computing, real time infrastructure saves cost, achieves more reliability and functionality related to the current requirements of industry in the term of control and acquisition. In current paper, the cryptyography algorithms are implements to secure data, and a conception model is proposed to implement SCADA within cloud environment and also discuss the security issues related to SCADA, and cloud computing. At the end; cryptographic solution is implemented in SCADA communication without or/and within the cloud infrastructure (Delivery of data/message from the Master terminal unit or MTU to Remote terminal unit or RTU) to achieve securit...
As a result of integration of many techniques such as grading, clustering, utilization computing ... more As a result of integration of many techniques such as grading, clustering, utilization computing and resource's sharing, cloud computing has been appeared as multi element's composition technology, it offers several computing services such as IaaS (infrastructure as service), PaaS (platform as service) and SaaS (software as service) based on pay as you use rule, but nevertheless, and because of cloud computing end users participate in computing resources (co_ tenancy) , and by which infrastructure computing can be shared by a number of users, and as a result to this feature, some security challenges has been existed and one of the most serious security threats is flooding attack, which prevent other users from using cloud infrastructure services, that kind of attack can be done by a legitimate or illegitimate cloud computing users. To overcome this problem various approaches have been proposed based on Artificial intelligence and statistical methods, but most of them conce...
Cloud computing is a new and most demandable technology in communication environment. Where compu... more Cloud computing is a new and most demandable technology in communication environment. Where computing resources such as hardware or/and software are processed as service over networks. SCADA implementation within cloud environment is relatively new and demandable over real time infrastructure (industrial infrastructure).The shifting (moving) of SCADA system (applications and resources) within cloud based infrastructure,meanfully overcome the cost and improve the reliability and performance of whole system. Cloud computing provides on-demand network access and batch of computing services for SCADA system. The current research paper takes two conceptual ideas to implement SCADA system within cloud computing (Hybrid Cloud) environment. In the first phase, SCADA applications are processed entirely inside the hybrid
Uploads
Papers by Shahrulniza Musa