Manolis Maragoudakis

Swarm intelligence in intrusion detection: A survey

Intrusion Detection Systems (IDS) have nowadays become a necessary component of almost every security infrastructure. So far, many different approaches have been followed in order to increase the efficiency of IDS. Swarm Intelligence (SI), a relatively new bio-inspired family of methods, seeks inspiration in the behavior of swarms of insects or other animals. After applied in other fields with success SI started to gather the interest of researchers working in the field of intrusion detection. In this paper we explore the reasons that led to the application of SI in intrusion detection, and present SI methods that have been used for constructing IDS. A major contribution of this work is also a detailed comparison of several SI-based IDS in terms of efficiency. This gives a clear idea of which solution is more appropriate for each particular case.

Preference Learning for Cognitive Modeling: A Case Study on Entertainment Preferences

Learning from preferences, which provide means for expressing a subjects desires, constitutes an important topic in machine learning research. This paper presents a comparative study of four alternative instance preference learning algorithms (both linear and nonlinear). The case study investigated is to learn to predict the expressed entertainment preferences of children when playing physical games built on their personalized playing features ( entertainment modeling). Two of the approaches are derived from the literature-the large-margin algorithm (LMA) and preference learning with Gaussian processes-while the remaining two are custom-designed approaches for the problem under investigation: meta-LMA and neuroevolution. Preference learning techniques are combined with feature set selection methods permitting the construction of effective preference models, given suitable individual playing features. The underlying preference model that best reflects children preferences is obtained through neuroevolution: 82.22% of cross-validation accuracy in predicting reported entertainment in the main set of game survey experimentation. The model is able to correctly match expressed preferences in 66.66% of cases on previously unseen data (p -value = 0.0136) of a second physical activity control experiment. Results indicate the benefit of the use of neuroevolution and sequential forward selection for the investigated complex case study of cognitive modeling in physical games.

Privacy Preservation by k-Anonymization of Weighted Social Networks

Privacy preserving analysis of a social network aims at a better understanding of the network and its behavior, while at the same time protecting the privacy of its individuals. We propose an anonymization method for weighted graphs, i.e., for social networks where the strengths of links are important. This is in contrast with many previous studies which only consider unweighted graphs. Weights can be essential for social network analysis, but they pose new challenges to privacy preserving network analysis. In this paper, we mainly consider prevention of identity disclosure, but we also touch on edge and edge weight disclosure in weighted graphs. We propose a method that provides k-anonymity of nodes against attacks where the adversary has information about the structure of the network, including its edge weights. The method is efficient, and it has been evaluated in terms of privacy and utility on real word datasets.

Improving SMS Usability Using Bayesian Networks

During the last years, the significant increase of mobile communications has resulted in the wide acceptance of a plethora of new services, like communication via written short messages (SMS). The limitations of the dimensions and the number of keys of the mobile phone keypad are probably the main obstacles of this service. Numerous intelligent techniques have been developed aiming at supporting users of SMS services. Special emphasis has been provided to the efficient and effective editing of words. In the presented research, we introduce a predictive algorithm that forecasts Greek letters occurrence during the process of compiling an SMS. The algorithm is based on Bayesian networks that have been trained with sufficient Greek corpus. The extracted network infers the probability of a specific letter in a word given one, two or three previous letter that have been keyed by the user with precision that reaches 95%. An important advantage, compared to other predictive algorithms is that the use of a vocabulary is not required, so the limited memory resources of mobile phones can easily accommodate the presented algorithm. The proposed method achieves improvement in the word editing time compared to the traditional editing method by a factor of 34.72%, as this has been proven by using Keystroke Level Modeling technique described in the paper.

Skin lesion diagnosis from images using novel ensemble classification techniques

Reduction of the error rate of melanoma diagnosis, a critical and very dangerous skin cancer that could be treated when early detected, is of major importance. Towards this direction, the present paper presents a novel ensemble classification technique, combining traditional Random Forests with the ‘Markov Blanket’ notion. The proposed algorithm performs an inherent feature selection phase where only truly informative features are carried forward, thus alleviating the curse of dimensionality and augmenting classification performance. It has been evaluated in a high-dimensional and imbalanced dataset of 1041 skin lesion images, which been preprocessed using the ABCD-rule of dermatology. The proposed ensemble classification technique exhibited a higher classification performance in comparison with the classical Random Forest algorithms, as well as other widely-used classification algorithms where standard feature reduction techniques, such as PCA and SVD, have been applied.

Cyberbullying Detection using Time Series Modeling

Cyber bullying is a new phenomenon resulting from the advance of new communication technologies including the Internet, cell phones and Personal Digital Assistants. It is a challenging bullying problem occurring in a new territory. Online bullying can be particularly damaging and upsetting because its usually anonymous or hard to trace. In this paper, the proposed method is utilizing a dataset of real world conversations (i.e. Pairs of questions and answers between cyber predator and the victim), in which each predator question is manually annotated in terms of severity using a numeric label. We approach the issue as a sequential data modelling approach, in which the predators questions are formulated using a Singular Value Decomposition representation. The motivation of this procedure is to study the accuracy of predicting the level of cyber bullying attack using classification methods and also to examine potential patterns between the lingustic style of each predator. More specifically, unlike previous approaches that consider a fixed window of a cyber-predators questions within a dialogue, we exploit the whole question set and model it as a signal, whose magnitude depends on the degree of bullying content. Using feature weighting and dimensionality reduction techniques, each signal is straightforwardly parsed by a neural network that forecasts the level of insult within a question given a window between two and three previous questions. Throughout the time series modeling experiments, an interesting discovery was made. By applying SVD on the time series data and taking into account the second dimension (since the first is usually modeling trivial dependencies between instances and attributes) we observed that its plot was very similar to the plot of the class attribute. By applying a Dynamic Time Warping algorithm, the similarity of the aforementioned signals was proved to exist, providing an immediate indicator for the severity of cyber bullying within a given dialogue.

Data Mining for Player Modeling in Videogames

In this paper we propose a method of videogame player modeling based on clustering of behavior data collected during game play. Based on the style of play, and game mechanics, we define two player types the action player and the tactical player. We then use the CURE clustering method to classify the game players according to their style of play. We demonstrate that the CURE algorithm can successfully assign the per-defined gamer type. The knowledge of the gamer type can then be used to adjust the game difficulty accordingly.

Learning automatic acquisition of subcategorization frames using Bayesian inference and support vector machines

Learning Bayesian belief networks (BBN) from corpora and support vector machines (SVM) have been applied to the automatic acquisition of verb subcategorization frames for Modern Greek. We are incorporating minimal linguistic resources, i.e. basic morphological tagging and phrase chunking, to demonstrate that verb subcategorization, which is of great significance for developing robust natural language human computer interaction systems, could be achieved using large corpora, without having any general-purpose, syntactic parser at all. In addition, apart from BBN and SVM, which have not previously used for this task, we have experimented with three well-known machine learning methods (feedforward backpropagation neural networks, learning vector quantization and decision tables), which are also being applied to the task of verb subcategorization frame defection for the first time. We argue that both BBN and SVM are well suited for learning to identify verb subcategorization frames. Empirical results will support this claim. Performance has been methodically evaluated using two different corpora types, one balanced and one domain-specific in order to determine the unbiased behaviour of the trained models. Limited training data are proved to endow with satisfactory results. We have been able to achieve precision exceeding 80% on the identification of subcategorization frames which were not known beforehand.

Heart sound screening in real-time assistive environments through MCMC Bayesian data mining

Emerging pervasive assistive environment applications for remote home healthcare monitoring of the elderly, disabled and also patients with various chronic diseases generate massive amounts of sensor signal data, which are transmitted from numerous homes to local health centers or hospitals. While it is critical to process this data efficiently (in a fast and accurate manner) and cost-effectively, in a large-scale application of the above technologies, it is not possible to do so manually by specialized human resources. This paper proposes a methodology for automatic real-time screening of heart sound signals (one of the most widely acquired signals from the human body for diagnostic purposes) and identification of those that are abnormal and require some action to be taken, which can be applied to many other similar types of bio-signals generated in assistive environments. It is based on a novel Markov Chain Monte Carlo Bayesian Inference approach, which estimates conditional probability distributions in structures obtained from a Tree-Augmented Naïve Bayes algorithm. It has been applied and validated in a highly ‘difficult’ heterogeneous dataset of 198 heart sound signals, which comes from both healthy medical cases and unhealthy ones having aortic stenosis, mitral regurgitation, aortic regurgitation or mitral stenosis. The proposed methodology achieved high classification performance in this difficult screening problem. It performs higher than other widely used classifiers, showing great potential for contributing to a cost-effective large-scale application of ICT-based assistive environment technologies.

A medical ontology for intelligent web-based skin lesions image retrieval

Researchers have applied increasing efforts towards providing formal computational frameworks to consolidate the plethora of concepts and relations used in the medical domain. In the domain of skin related diseases, the variability of semantic features contained within digital skin images is a major barrier to the medical understanding of the symptoms and development of early skin cancers. The desideratum of making these standards machine-readable has led to their formalization in ontologies. In this work, in an attempt to enhance an existing Core Ontology for skin lesion images, hand-coded from image features, high quality images were analyzed by an autonomous ontology creation engine. We show that by exploiting agglomerative clustering methods with distance criteria upon the existing ontological structure, the original domain model could be enhanced with new instances, attributes and even relations, thus allowing for better classification and retrieval of skin lesion categories from the web.