Data clustering is a basic technique to show the structure of a data set. K-means clustering is a widely acceptable method of data clustering, which follow a partitioned approach for dividing the given data set into non-overlapping groups. Unfortunately, it has the pitfall of randomly choosing the initial cluster centers. Due to its gradient nature, this algorithm is highly sensitive to the initial seed value. In this paper, we propose a kernel density-based method to compute an initial seed value for the k-means algorithm. The idea is to select an initial point from the denser region because they truly reflect the property of the overall data set. Subsequently, we are avoiding the selection of outliers as an initial seed value. We have verified the proposed method on real data sets with the help of different internal and external validity measures. The experimental analysis illustrates that the proposed method has better performance over the kmeans, k-means++ algorithm, and other recent initialization methods.

The use of machine learning for forgery detection in image forgeries poses a significant danger to modern multimedia technologies. These forgery detectors appear promising; however they are known to be vulnerable to evasion attacks. A "forgery" is a malicious user established by an attacker that may change its appearance while spreading; this is known as polymorphism and metamorphosis. Common behavioral characteristics showing the source and intended use of multimedia devices are displayed by image forgeries. However, the majority of novel forgery detection variations frequently elude detection by traditional forgery detection methods. This paper proposes a new approach to improve the utility of an existing framework. We need to create a unique model, PH-SIFT using PSO Algorithm, in order to improve the overall forgery analysis because the current deep learning algorithm forgery detection and analysis. Applying the PH-SIFT utilizing PSO Algorithm technique, which also aims to improve its accuracy and other metrics, achieves this. This method can get around the complexity and challenges presented by different concerns related to forgery detection and analysis. Here, the datasets from forgery detection and analysis are used to train the proposed algorithm. Following training, the datasets are pre-processed by reducing training mistakes. The PSO Algorithm-built PH-SIFT then carries out the detection procedure to anticipate any forgery activity.

The transport layer plays a crucial role in the Mobile Adhoc Network (MANET) protocol stack by controlling traffic flow, managing congestion, and enabling end-to-end delivery. With the help of congestion control mechanisms, numerous protocols are formed to enhance MANET performance. This paper focuses on a thorough analysis of the challenges the MANET protocol stack is facing as a result of congestion control issues such high overload, long delays, and increased packet loss. Finally, note that in order to increase MANET performance, research needs to concentrate on specific congestion control mechanisms.