The escalating reliance of modern society on information and communication technology has rendered it vulnerable to an array of cyber-attacks, with distributed denial-of-service (DDoS) attacks emerging as one of the most prevalent threats. This paper delves into the intricacies of DDoS attacks, which exploit compromised machines numbering in the thousands to disrupt data services and online commercial platforms, resulting in significant downtime and financial losses. Recognizing the gravity of this issue, various detection techniques have been explored, yet the quantity and prior detection of DDoS attacks has seen a decline in recent methods. This research introduces an innovative approach by integrating evolutionary optimization algorithms and machine learning techniques. Specifically, the study proposes XGB-GA Optimization, RF-GA Optimization, and SVM-GA Optimization methods, employing Evolutionary Algorithms (EAs) Optimization with Tree-based Pipelines Optimization Tool (TPOT)-Genetic Programming. Datasets pertaining to DDoS attacks were utilized to train machine learning models based on XGB, RF, and SVM algorithms, and 10-fold cross-validation was employed. The models were further optimized using EAs, achieving remarkable accuracy scores: 99.99% with the XGB-GA method, 99.50% with RF-GA, and 99.99% with SVM-GA. Furthermore, the study employed TPOT to identify the optimal algorithm for constructing a machine learning model, with the genetic algorithm pinpointing XGB-GA as the most effective choice. This research significantly advances the field of DDoS attack detection by presenting a robust and accurate methodology, thereby enhancing the cybersecurity landscape and fortifying digital infrastructures against these pervasive threats.
Background: Drug targets in living beings perform pivotal roles in the discovery of potential drugs. Conventional wet-lab characterization of drug targets is although accurate but generally expensive, slow, and resource intensive. Therefore, computational methods are highly desirable as an alternative to expedite the large-scale identification of druggable proteins (DPs); however, the existing in silico predictor’s performance is still not satisfactory. Methods: In this study, we developed a novel deep learning-based model DPI_CDF for predicting DPs based on protein sequence only. DPI_CDF utilizes evolutionary-based (i.e., histograms of oriented gradients for position-specific scoring matrix), physiochemical-based (i.e., component protein sequence representation), and compositional-based (i.e., normalized qualitative characteristic) properties of protein sequence to generate features. Then a hierarchical deep forest model fuses these three encoding schemes to build the proposed model DPI_CDF. Results: The empirical outcomes on 10-fold cross-validation demonstrate that the proposed model achieved 99.13 % accuracy and 0.982 of Matthew’s-correlation-coefficient (MCC) on the training dataset. The generalization power of the trained model is further examined on an independent dataset and achieved 95.01% of maximum accuracy and 0.900 MCC. When compared to current state-of-the-art methods, DPI_CDF improves in terms of accuracy by 4.27% and 4.31% on training and testing datasets, respectively. We believe, DPI_CDF will support the research community to identify druggable proteins and escalate the drug discovery process. Availability: The benchmark datasets and source codes are available in GitHub: http://github.com/Muhammad-Arif-NUST/DPI_CDF.
Named data networking (NDN) is gaining momentum in vehicular ad hoc networks (VANETs) thanks to its robust network architecture. However, vehicular NDN (VNDN) faces numerous challenges, including security, privacy, routing, and caching. Specifically, the attackers can jeopardize vehicles’ cache memory with a Content Poisoning Attack (CPA). The CPA is the most difficult to identify because the attacker disseminates malicious content with a valid name. In addition, NDN employs request–response-based content dissemination, which is inefficient in supporting push-based content forwarding in VANET. Meanwhile, VNDN lacks a secure reputation management system. To this end, our contribution is three-fold. We initially propose a threshold-based content caching mechanism for CPA detection and prevention. This mechanism allows or rejects host vehicles to serve content based on their reputation. Secondly, we incorporate a blockchain system that ensures the privacy of every vehicle at roadside units (RSUs). Finally, we extend the scope of NDN from pull-based content retrieval to push-based content dissemination. The experimental evaluation results reveal that our proposed CPA detection mechanism achieves a 100% accuracy in identifying and preventing attackers. The attacker vehicles achieved a 0% cache hit ratio in our proposed mechanism. On the other hand, our blockchain results identified tempered blocks with 100% accuracy and prevented them from storing in the blockchain network. Thus, our proposed solution can identify and prevent CPA with 100% accuracy and effectively filters out tempered blocks. Our proposed research contribution enables the vehicles to store and serve trusted content in VNDN.
1 February 2022 in Chemometrics and Intelligent Laboratory Systems
In several projects in computational biology (CB), bioinformatics, health informatics(HI), precision medicine(PM) and precision agriculture(PA) machine learning(ML) has become a primary resource. In this paper we studied the use of machine learning in the development of computational methods for top five research aeras. The last few years have seen an increased interest in Artificial Intelligence (AI), comprehensive ML and DL techniques for computational method development. Over the years, an enormous amount of research has been biomedical scientists still don’t have more knowledge to handle a biomedical projects efficiently and may, therefore, adopt wrong methods, which can lead to frequent errors or inflated tests. Healthcare has become a fruitful ground for artificial intelligence (AI) and machine learning due to the increase in the volume, diversity, and complexity of data (ML). Healthcare providers and life sciences businesses already use a variety of AI technologies. The review summarizes a traditional machine learning cycle, several machine learning algorithms, various techniques to data analysis, and effective use in five research areas. In this comprehensive review analysis, we proposed 10 ten rapid and accurate practices to use ML techniques in health informatics, bioinformatics, computational and systems biology, precision medicine and precision agriculture, avoid some common mistakes that we have observed several hundred times in several computational method works.
5 March 2023 in VAWKUM Transactions on Computer Sciences
Immunoglobulin has a close connection to a number of disorders and is important in both biological and medicinal contexts. Therefore, it is crucial for illness research to employ efficient techniques to increase the categorization accuracy of immunoglobulins. Computational models have been used in a small number of research to address this important issue, but the accuracy of the predictions is not good enough. As a result, we use a cutting-edge deep learning technique with convolutional neural networks to enhance the performance results. In this study, the immunoglobulin features were extracted using the dipeptide acid composition (DPC) with the position-specific scoring matrix (DPC-PSSM) and position-specific scoring matrix-transition probability composition (PSSM-TPC) methods. we apply extracted features information from the DPC-PSSM profiles and PSSM-TPC profile by using a 1D-convolutional neural network (CNN) over an input shape. The outcomes demonstrated that the DeepImmuno-PSSM method based on sequential minimal optimization was able to properly predict DPC-PSSM accuracy score 93.44% obtained and of the immunoglobulins using the greatest feature subcategory produced by the PSSM-TPC feature mining approach accuracy score 89.92% obtained. Our findings indicate that we are able to provide a useful model for enhancing immunoglobulin proteins' capacity for prediction. Additionally, it implies that employing sequence data in deep learning and PSSM-based features may open up new path for biochemical modelling.