A new intrusion detection system using support vector machines and hierarchical clustering

Whenever an intrusion occurs, the security and value of a computer system is compromised. Network-based attacks make it difficult for legitimate users to access various network services by purposely occupying or sabotaging network resources and services. This can be done by sending large amounts of network traffic, exploiting well-known faults in networking services, and by overloading network hosts. Intrusion Detection attempts to detect computer attacks by examining various data records observed in processes on the network and it is split into two groups, anomaly detection systems and misuse detection systems. Anomaly detection is an attempt to search for malicious behavior that deviates from established normal patterns. Misuse detection is used to identify intrusions that match known attack scenarios. Our interest here is in anomaly detection and our proposed method is a scalable solution for detecting network-based anomalies. We use Support Vector Machines (SVM) for classification. The SVM is one of the most successful classification algorithms in the data mining area, but its long training time limits its use. This paper presents a study for enhancing the training time of SVM, specifically when dealing with large data sets, using hierarchical clustering analysis. We use the Dynamically Growing Self-Organizing Tree (DGSOT) algorithm for clustering because it has proved to overcome the drawbacks of traditional hierarchical clustering algorithms (e.g., hierarchical agglomerative clustering). Clustering analysis helps find the boundary points, which are the most qualified data points to train SVM, between two classes. We present a new approach of combination of SVM and DGSOT, which starts with an initial training set and expands it gradually using the clustering structure produced by the DGSOT algorithm. We compare our approach with the Rocchio Bundling technique and random selection in terms of accuracy loss and training time gain using a single benchmark real data set. We show that our proposed variations contribute significantly in improving the training process of SVM with high generalization accuracy and outperform the Rocchio Bundling technique.     

Effect of eggshell powder as nucleating agent on the structure,morphology and functional properties of normal corn starch foams

Corn starch and eggshell powder (with particle sizes of 4–5µm and 8–10µm) composite foams were prepared by extrusion. Effects of eggshell on the structure, morphology, physical properties (unit density and expansion ratio), mechanical properties (spring index and compressibility) and thermal behaviour (thermal transition and stability) of the foams were investigated. Foam cell size decreased and cell population increased with addition of eggshell into starch matrix. The foam unit density, expansion ratio and compressibility decreased significantly (p < 0.05), while the spring index increased significantly (p < 0.05) as the eggshell content increased from 0 to 6wt%. Further increasing eggshell content to 10wt% increased the unit density and compressibility and decreased the expansion ratio and spring index. The thermal transition and stability increased with the addition of eggshell. The optimum eggshell content was 6wt% and the smaller‐sized eggshell powder had a favourable effect on the functional properties of the foams. Copyright © 2006 John Wiley & Sons, Ltd.     

A new geometric model for clustering high-resolution satellite images

Image segmentation is a central process in image processing. There are many segmentation methods such as region growing, edge detection, split and merge and artificial neural networks (ANNs). However, the most important and popular are clustering methods. Normally, clustering methods select cluster centres randomly to segment an image into disjoint and homogeneous regions. The use of random cluster centres without a priori knowledge leads to degradation in the accuracy of the obtained results. However, combined with edge detection, shape representation can help in improving the clustering methods. The improvement is obtained by knowing the optimal location of the cluster centres at the beginning of the image segmentation process. In this article, a new geometric model for high-resolution satellite image segmentation is implemented that can overcome the problem encountered in random clustering processes. The proposed model uses Canny–Deriche edge detection and the modified non-uniform rational B-spline (NURBS) methods to generate the control points of the edges. These points are used to identify cluster centres that are necessary to create the population of the hybrid dynamic genetic algorithm (HDGA). The new geometric model is compared with the self-organizing maps (SOMs) method, which is an efficient unsupervised ANN method. Two experiments are conducted using high-resolution satellite images, and the results prove the high accuracy and reliability of the new evolutionary geometric model.     

Stress analysis of multi-layer electronic and mechanical systems (MEMS) under fatigue and impact loading conditions

Detailed micro electronic and mechanical systems (MEMS) for a mobile microprocessor complex shape were modeled using Finite Element (FE) processing. Fatigue and impact conditions were performed on the Ball Grid Array (BGA) Integrated Circuit (IC) using Abaqus\CAE finite element analysis software. The main objective of this research is to make sure that BGA products can endure the roughness of the daily usage, where a portable electronic product is habitually coupled with potential damage of functional failure when the device falls.     

Provenance studies of archaeological ceramics from Mar-Takla(Ain-Minin,Syria)using radioisotope X-ray fluorescence method

The radioisotope X-ray fluorescence method was applied to studies of the provenance of the ceramics fragments originated from the Mar-Takla site in Syria.The samples were irradiated 1000s by a ^109Cd radioisotope source and 13 elements (Ca,Ti,Mn,Fe,Zn,Ga,As,Rb,Sr,Y,Zr,Nb,and Pb)were determined in 35 samples.The data were subjected to two multivariate statistical methods,cluster and principal components analysis(PCA).It was shown from the combination of the statistical techniques and the determination of elemental composition of the samples that 94% of the ceramic samples analyzed can be considered to be manufactured using two sources of raw materials.     

Failure Analysis of Three Slow-Speed Induction Motors for Reciprocating Load Application

The North Atlantic Refinery has three 5400-hp slow-speed induction motors driving reciprocating compressors for a Hydrogen plant. All motors were rewound on-site over the period of 1998-2002, using magnetic wedges as per the original design, and the motor protection was upgraded to microprocessor-type relays. In summer 2003, during routine maintenance, it was noted that these motors suffered substantial loss of slot magnetic wedges and had to be scheduled for rewedging over 2004-2006. This paper presents failure analysis, field measurements covering heat-run test results, and motor-performance and vibration analysis before and after motor rewedging work. This paper reviews the key factors for using magnetic versus nonmagnetic wedges in these applications and techniques available to predict loss of wedges. The use of a microprocessor relay for slow-speed induction motor protection is discussed, particularly, the thermal protection feature to address the cyclic-load pattern caused by the reciprocating compressor     

Bioglass as a carrier for reindeer bone protein extract in the healing of rat femur defect

Bioactive glasses have been developed as scaffolds for bone tissue engineering but combination with reindeer bone protein extract has not been evaluated. We investigated the effects of bone protein extract implants (5–40 mg dosages) with bioglass (BG) carrier on the healing of rat femur defects. Bioglass implants and untreated defects served as controls. All doses of extract increased bone formation compared with the control groups, and bone union was enhanced with doses of 10 mg or more. In comparison with untreated defect, mean cross-sectional bone area at the defect site was greater when implants with BG + 15 mg of extract or bioglass alone were used, bone density at the defect site was higher in all bioglass groups with and without bone extract, and the BG + 15 mg extract dosage marginally increased bone torsional stiffness in mechanical testing. Bioglass performed well as a carrier candidate for reindeer bone protein extract.