MUCKE aims to mine a large volume of images, to structure them conceptually and to use this conceptual structuring in order to improve large-scale image retrieval. The last decade witnessed important progress concerning low-level image representations. However, there are a number problems which need to be solved in order to unleash the full potential of image mining in applications. The central problem with low-level representations is the mismatch between them and the human interpretation of image content. This problem can be instantiated, for instance, by the incapability of existing descriptors to capture spatial relationships between the concepts represented or by their incapability to convey an explanation of why two images are similar in a content-based image retrieval framework. We start by assessing existing local descriptors for image classification and by proposing to use co-occurrence matrices to better capture spatial relationships in images. The main focus in MUCKE is on cleaning large scale Web image corpora and on proposing image representations which are closer to the human interpretation of images. Consequently, we introduce methods which tackle these two problems and compare results to state of the art methods. Note: some aspects of this deliverable are withheld at this time as they are pending review. Please contact the authors for a preview.

http://ieeexplore.ieee.org/iel2/4524/12820/00592460.pdf

Image Processing

COVID-19 identification in chest X-ray images on flat and hierarchical classification scenarios.

The first € price and the £ and $ price are net prices, subject to local VAT. Prices indicated with * include VAT for books; the €(D) includes 7% for Germany, the €(A) includes 10% for Austria. Prices indicated with ** include VAT for electronic products; 19% for Germany, 20% for Austria. All prices exclusive of carriage charges. Prices and other details are subject to change without notice. All errors and omissions excepted. R. Lee (Ed.) Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing

Seventh ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing

Purpose – Many organizations are outsourcing their information technology (IT) related services to a third party vendor for quite some time. However, the IT services industry including outsourcing is going through rapid changes with the increasing adoption of Cloud computing. The purpose of this paper is to compare global IT outsourcing with Cloud computing along with the evolution of traditional IT services.Design/methodology/approach – Cloud computing is a model for provisioning and consuming IT capabilities on a need and pay by use basis. This helps in shifting the cost structure from capital expenditure to operating expenditure and also helps the IT systems to be more agile. This innovative model of acquiring IT related services has made organizations revisit their infrastructure and platform services strategy and optimize their IT spending while improving overall agility. This paper compares global IT outsourcing with Cloud computing along with the evolution of traditional IT services.Findings – The ...

From Outsourcing To Cloud Computing: Evolution of IT Services

A variety of screening approaches have been proposed to diagnose epileptic seizures, using electroencephalography (EEG) and magnetic resonance imaging (MRI) modalities. Artificial intelligence encompasses a variety of areas, and one of its branches is deep learning (DL). Before the rise of DL, conventional machine learning algorithms involving feature extraction were performed. This limited their performance to the ability of those handcrafting the features. However, in DL, the extraction of features and classification are entirely automated. The advent of these techniques in many areas of medicine, such as in the diagnosis of epileptic seizures, has made significant advances. In this study, a comprehensive overview of works focused on automated epileptic seizure detection using DL techniques and neuroimaging modalities is presented. Various methods proposed to diagnose epileptic seizures automatically using EEG and MRI modalities are described. In addition, rehabilitation systems developed for epileptic seizures using DL have been analyzed, and a summary is provided. The rehabilitation tools include cloud computing techniques and hardware required for implementation of DL algorithms. The important challenges in accurate detection of automated epileptic seizures using DL with EEG and MRI modalities are discussed. The advantages and limitations in employing DL-based techniques for epileptic seizures diagnosis are presented. Finally, the most promising DL models proposed and possible future works on automated epileptic seizure detection are delineated.

Epileptic Seizures Detection Using Deep Learning Techniques: A Review.

Identifying influential spreaders in complex networks is crucial in understanding, controlling and accelerating spreading processes for diseases, information, innovations, behaviors, and so on. Inspired by the gravity law, we propose a gravity model that utilizes both neighborhood information and path information to measure a node’s importance in spreading dynamics. In order to reduce the accumulated errors caused by interactions at distance and to lower the computational complexity, a local version of the gravity model is further proposed by introducing a truncation radius. Empirical analyses of the Susceptible-Infected-Recovered (SIR) spreading dynamics on fourteen real networks show that the gravity model and the local gravity model perform very competitively in comparison with well-known state-of-the-art methods. For the local gravity model, the empirical results suggest an approximately linear relation between the optimal truncation radius and the average distance of the network.

/pdf/identifying-influential-spreaders-by-gravity-model-9ug9f4ovmc.pdf

Identifying influential spreaders by gravity model

Background. Accurate classification for different non-Hodgkin lymphomas (NHL) is one of the main challenges in clinical pathological diagnosis due to its intrinsic complexity. Therefore, this paper proposes an effective classification model for three types of NHL pathological images, including mantle cell lymphoma (MCL), follicular lymphoma (FL), and chronic lymphocytic leukemia (CLL). Methods. There are three main parts with respect to our model. First, NHL pathological images stained by hematoxylin and eosin (H&E) are transferred into blue ratio (BR) and Lab spaces, respectively. Then specific patch-level textural and statistical features are extracted from BR images and color features are obtained from Lab images both using a hierarchical way, yielding a set of hand-crafted representations corresponding to different image spaces. A random forest classifier is subsequently trained for patch-level classification. Second, H&E images are cropped and fed into a pretrained google inception net (GoogLeNet) for learning high-level representations and a softmax classifier is used for patch-level classification. Finally, three image-level classification strategies based on patch-level results are discussed including a novel method for calculating the weighted sum of patch results. Different classification results are fused at both feature 1 and image levels to obtain a more satisfactory result. Results. The proposed model is evaluated on a public IICBU Malignant Lymphoma Dataset and achieves an improved overall accuracy of 0.991 and area under the receiver operating characteristic curve of 0.998. Conclusion. The experimentations demonstrate the significantly increased classification performance of the proposed model, indicating that it is a suitable classification approach for NHL pathological images.

/pdf/nhl-pathological-image-classification-based-on-hierarchical-epigmebtr6.pdf

NHL Pathological Image Classification Based on Hierarchical Local Information and GoogLeNet-Based Representations.

Zero day Cyber-attacks created potential impacts on the way information is held and protected, however one of the vital priorities for governments, agencies and organizations is to secure their network businesses, transactions and communications, simultaneously to avoid security policy and privacy violations under any circumstances. Covert Channel is used to in/ex-filtrate classified data secretly, whereas encryption is used merely to protect communication from being decoded by unauthorized access. In this paper, we propose a new Machine Learning approach to detect covert channel implementing an enhanced feature selection algorithm supporting Naive Bayesian classifier. NBC is one of the most prominent classification algorithm defining the highest probability in data mining area. The proposed framework uses Intelligent Heuristic Algorithm (IHA) to create novel primary training data, in addition to a modified Decision Tree C4.5 technique to detect and classify hidden channels in IPv6 network. The results showed better detection performance and high accuracy in True Positive Rate (TPR) and a low false negative rate (FNR) in comparison to other previous techniques. Keywords—CyberSecurity,Covert Channel, ICMPv6, IPv6,Naive Bayes, Dicision Trees C4.5, MLA ,DARPA, NSL-KDD.

Detection and classification of covert channels in IPv6 using enhanced machine learning

We report upon the development of sif (for service-oriented interoperability framework), a platform that has been developed to support the secure aggregation of medical data from disparate sources. By taking a data-agnostic approach to data access and transfer, sif provides a generic interface to data sources, which allows the current version to expose data from any relational database and any file system in a secure fashion. Application developers may then access and utilise such data via a simple API. sif is being developed within the GIMI (Generic Infrastructure for Medical Informatics) project; as such, we discuss its various applications within that context.

A healthcare−driven framework for facilitating the secure sharing of data across organisational boundaries

Intrusion detection systems offer monolithic way to detect attacks through monitoring, searching for abnormal characteristics, and malicious behavior in network communications. Cyber-attack is performed through using covert channel which currently is one of the most sophisticated challenges facing network security systems. Covert channel is used to ex/infiltrate classified information from legitimate targets; consequently, this manipulation violates network security policy and privacy. The New Generation Internet Protocol version 6 (IPv6) has certain security vulnerabilities and need to be addressed using further advanced techniques. Fuzzy rule is implemented to classify different network attacks as an advanced machine learning technique, meanwhile, Genetic algorithm is considered as an optimization technique to obtain the ideal fuzzy rule. This paper suggests a novel hybrid covert channel detection system implementing two Artificial Intelligence (AI) techniques, Fuzzy Logic and Genetic Algorithm (FLGA), to gain sufficient and optimal detection rule against covert channel. Our approach counters sophisticated network unknown attacks through an advanced analysis of deep packet inspection. Results of our suggested system offer high detection rate of 97.7 % and a better performance in comparison to previous tested techniques.

/pdf/implementation-of-hybrid-artificial-intelligence-technique-2eummv42gf.pdf

Xiaoqi Ma

Papers

Identifying influential spreaders by gravity model

NHL Pathological Image Classification Based on Hierarchical Local Information and GoogLeNet-Based Representations.

Detection and classification of covert channels in IPv6 using enhanced machine learning

A healthcare−driven framework for facilitating the secure sharing of data across organisational boundaries

Implementation of Hybrid Artificial Intelligence Technique to Detect Covert Channels Attack in New Generation Internet Protocol IPv6