The ever-growing network traffic threat landscape necessitates adopting accurate and robust intrusion detection systems(IDSs).IDSs have become a research hotspot and have seen remarkable performance improvements.Gener...The ever-growing network traffic threat landscape necessitates adopting accurate and robust intrusion detection systems(IDSs).IDSs have become a research hotspot and have seen remarkable performance improvements.Generative adversarial networks(GANs)have also garnered increasing research interest recently due to their remarkable ability to generate data.This paper investigates the application of(GANs)in(IDS)and explores their current use within this research field.We delve into the adoption of GANs within signature-based,anomaly-based,and hybrid IDSs,focusing on their objectives,methodologies,and advantages.Overall,GANs have been widely employed,mainly focused on solving the class imbalance issue by generating realistic attack samples.While GANs have shown significant potential in addressing the class imbalance issue,there are still open opportunities and challenges to be addressed.Little attention has been paid to their applicability in distributed and decentralized domains,such as IoT networks.Efficiency and scalability have been mostly overlooked,and thus,future works must aim at addressing these gaps.展开更多
There are several ethical issues that have arisen in recent years due to the ubiquity of the Internet and the popularity of social media and community platforms.Among them is cyberbullying,which is defined as any viol...There are several ethical issues that have arisen in recent years due to the ubiquity of the Internet and the popularity of social media and community platforms.Among them is cyberbullying,which is defined as any violent intentional action that is repeatedly conducted by individuals or groups using online channels against victims who are not able to react effectively.An alarmingly high percentage of people,especially teenagers,have reported being cyberbullied in recent years.A variety of approaches have been developed to detect cyberbullying,but they require time-consuming feature extraction and selection processes.Moreover,no approach to date has examined the meanings of words and the semantics involved in cyberbullying.In past work,we proposed an algorithm called Cyberbullying Detection Based on Deep Learning(CDDL)to bridge this gap.It eliminates the need for feature engineering and generates better predictions than traditional approaches for detecting cyberbullying.This was accomplished by incorporating deep learning—specifically,a convolutional neural network(CNN)—into the detection process.Although this algorithm shows remarkable improvement in performance over traditional detection mechanisms,one problem with it persists:CDDL requires that many parameters(filters,kernels,pool size,and number of neurons)be set prior to classification.These parameters play a major role in the quality of predictions,but a method for finding a suitable combination of their values remains elusive.To address this issue,we propose an algorithm called firefly-CDDL that incorporates a firefly optimisation algorithm into CDDL to automate the hitherto-manual trial-and-error hyperparameter setting.The proposed method does not require features for its predictions and its detection of cyberbullying is fully automated.The firefly-CDDL outperformed prevalent methods for detecting cyberbullying in experiments and recorded an accuracy of 98%within acceptable polynomial time.展开更多
Two traditional recommendation techniques, content-based and collaborative filtering （CF）, have been widely used in a broad range of domain areas. Both meth- ods have their advantages and disadvantages, and some of ...Two traditional recommendation techniques, content-based and collaborative filtering （CF）, have been widely used in a broad range of domain areas. Both meth- ods have their advantages and disadvantages, and some of the defects can be resolved by integrating both techniques in a hybrid model to improve the quality of the recommendation. In this article, we will present a problem-oriented approach to design a hybrid immunizing solution for job recommen- dation problem from applicant＇s perspective. The proposed approach aims to recommend the best chances of opening jobs to the applicant who searches for job. It combines the artificial immune system （AIS）, which has a powerful explo- ration capability in polynomial time, with the collaborative filtering, which can exploit the neighbors＇ interests. We will discuss the design issues, as well as the hybridization process that should be applied to the problem. Finally, experimental studies are conducted and the results show the importance of our approach for solving the job recommendation problem.展开更多
文摘The ever-growing network traffic threat landscape necessitates adopting accurate and robust intrusion detection systems(IDSs).IDSs have become a research hotspot and have seen remarkable performance improvements.Generative adversarial networks(GANs)have also garnered increasing research interest recently due to their remarkable ability to generate data.This paper investigates the application of(GANs)in(IDS)and explores their current use within this research field.We delve into the adoption of GANs within signature-based,anomaly-based,and hybrid IDSs,focusing on their objectives,methodologies,and advantages.Overall,GANs have been widely employed,mainly focused on solving the class imbalance issue by generating realistic attack samples.While GANs have shown significant potential in addressing the class imbalance issue,there are still open opportunities and challenges to be addressed.Little attention has been paid to their applicability in distributed and decentralized domains,such as IoT networks.Efficiency and scalability have been mostly overlooked,and thus,future works must aim at addressing these gaps.
文摘There are several ethical issues that have arisen in recent years due to the ubiquity of the Internet and the popularity of social media and community platforms.Among them is cyberbullying,which is defined as any violent intentional action that is repeatedly conducted by individuals or groups using online channels against victims who are not able to react effectively.An alarmingly high percentage of people,especially teenagers,have reported being cyberbullied in recent years.A variety of approaches have been developed to detect cyberbullying,but they require time-consuming feature extraction and selection processes.Moreover,no approach to date has examined the meanings of words and the semantics involved in cyberbullying.In past work,we proposed an algorithm called Cyberbullying Detection Based on Deep Learning(CDDL)to bridge this gap.It eliminates the need for feature engineering and generates better predictions than traditional approaches for detecting cyberbullying.This was accomplished by incorporating deep learning—specifically,a convolutional neural network(CNN)—into the detection process.Although this algorithm shows remarkable improvement in performance over traditional detection mechanisms,one problem with it persists:CDDL requires that many parameters(filters,kernels,pool size,and number of neurons)be set prior to classification.These parameters play a major role in the quality of predictions,but a method for finding a suitable combination of their values remains elusive.To address this issue,we propose an algorithm called firefly-CDDL that incorporates a firefly optimisation algorithm into CDDL to automate the hitherto-manual trial-and-error hyperparameter setting.The proposed method does not require features for its predictions and its detection of cyberbullying is fully automated.The firefly-CDDL outperformed prevalent methods for detecting cyberbullying in experiments and recorded an accuracy of 98%within acceptable polynomial time.
文摘Two traditional recommendation techniques, content-based and collaborative filtering （CF）, have been widely used in a broad range of domain areas. Both meth- ods have their advantages and disadvantages, and some of the defects can be resolved by integrating both techniques in a hybrid model to improve the quality of the recommendation. In this article, we will present a problem-oriented approach to design a hybrid immunizing solution for job recommen- dation problem from applicant＇s perspective. The proposed approach aims to recommend the best chances of opening jobs to the applicant who searches for job. It combines the artificial immune system （AIS）, which has a powerful explo- ration capability in polynomial time, with the collaborative filtering, which can exploit the neighbors＇ interests. We will discuss the design issues, as well as the hybridization process that should be applied to the problem. Finally, experimental studies are conducted and the results show the importance of our approach for solving the job recommendation problem.
