Hierarchical Self-organizing Map Research Articles

Distributed and explainable GHSOM for anomaly detection in sensor networks

The identification of anomalous activities is a challenging and crucially important task in sensor networks. This task is becoming increasingly complex with the increasing volume of data generated in real-world domains, and greatly benefits from the use of predictive models to identify anomalies in real time. A key use case for this task is the identification of misbehavior that may be caused by involuntary faults or deliberate actions. However, currently adopted anomaly detection methods are often affected by limitations such as the inability to analyze large-scale data, a reduced effectiveness when data presents multiple densities, a strong dependence on user-defined threshold configurations, and a lack of explainability in the extracted predictions. In this paper, we propose a distributed deep learning method that extends growing hierarchical self-organizing maps, originally designed for clustering tasks, to address anomaly detection tasks. The SOM-based modeling capabilities of the method enable the analysis of data with multiple densities, by exploiting multiple SOMs organized as a hierarchy. Our map-reduce implementation under Apache Spark allows the method to process and analyze large-scale sensor network data. An automatic threshold-tuning strategy reduces user efforts and increases the robustness of the method with respect to noisy instances. Moreover, an explainability component resorting to instance-based feature ranking emphasizes the most salient features influencing the decisions of the anomaly detection model, supporting users in their understanding of raised alerts. Experiments are conducted on five real-world sensor network datasets, including wind and photovoltaic energy production, vehicular traffic, and pedestrian flows. Our results show that the proposed method outperforms state-of-the-art anomaly detection competitors. Furthermore, a scalability analysis reveals that the method is able to scale linearly as the data volume presented increases, leveraging multiple worker nodes in a distributed computing setting. Qualitative analyses on the level of anomalous pollen in the air further emphasize the effectiveness of our proposed method, and its potential in determining the level of danger in raised alerts.

Genetic risks of Alzheimer's by APOE and MAPT on cortical morphology in young healthy adults.

Genetic risk factors such as APOE ε4 and MAPT (rs242557) A allele are associated with amyloid and tau pathways and grey matter changes at both early and established stages of Alzheimer's disease, but their effects on cortical morphology in young healthy adults remain unclear. A total of 144 participants aged from 18 to 24 underwent 3T MRI and genotyping for APOE and MAPT to investigate unique impacts of these genetic risk factors in a cohort without significant comorbid conditions such as metabolic and cardiovascular diseases. We segmented the cerebral cortex into 68 regions and calculated the cortical area, thickness, curvature and folding index for each region. Then, we trained machine learning models to classify APOE and MAPT genotypes using these morphological features. In addition, we applied a growing hierarchical self-organizing maps algorithm, which clustered the 68 regions into 4 subgroups representing different morphological patterns. Then, we performed general linear model analyses to estimate the interaction between APOE and MAPT on cortical patterns. We found that the classifiers using all cortical features could accurately classify individuals carrying genetic risks of dementia outperforming each individual feature alone. APOE ε4 carriers had a more convoluted and thinner cortex across the cerebral cortex. A similar pattern was found in MAPT A allele carriers only in the regions that are vulnerable for early tau pathology. With the clustering analysis, we found a synergetic effect between APOE ε4 and MAPT A allele, i.e. carriers of both risk factors showed the most deviation of cortical pattern from the typical pattern of that cluster. Genetic risk factors of dementia by APOE ε4 and MAPT (rs242557) A allele were associated with variations of cortical morphology, which can be observed in young healthy adults more than 30 years before Alzheimer's pathology is likely to occur and 50 years before dementia symptoms may begin.

Biologically Inspired Self-Organizing Computational Model to Mimic Infant Learning

Recent technological advancements have fostered human–robot coexistence in work and residential environments. The assistive robot must exhibit humane behavior and consistent care to become an integral part of the human habitat. Furthermore, the robot requires an adaptive unsupervised learning model to explore unfamiliar conditions and collaborate seamlessly. This paper introduces variants of the growing hierarchical self-organizing map (GHSOM)-based computational models for assistive robots, which constructs knowledge from unsupervised exploration-based learning. Traditional self-organizing map (SOM) algorithms have shortcomings, including finite neuron structure, user-defined parameters, and non-hierarchical adaptive architecture. The proposed models overcome these limitations and dynamically grow to form problem-dependent hierarchical feature clusters, thereby allowing associative learning and symbol grounding. Infants can learn from their surroundings through exploration and experience, developing new neuronal connections as they learn. They can also apply their prior knowledge to solve unfamiliar problems. With infant-like emergent behavior, the presented models can operate on different problems without modifications, producing new patterns not present in the input vectors and allowing interactive result visualization. The proposed models are applied to the color, handwritten digits clustering, finger identification, and image classification problems to evaluate their adaptiveness and infant-like knowledge building. The results show that the proposed models are the preferred generalized models for assistive robots.

Growing Hierarchical Self-Organising Representation Map (GHSORM)

The Growing Hierarchical Self-Organising Representation Map (GHSORM) is a model fusing the denoising autoencoder, used to better represent a dataset, and the Growing Hierarchical Self-Organising Map, used for organizing and projecting the input data in clusters of varying detail. It is shown here that the GHSORM is instrumental in sub-grouping clusters that are not fully separable by a single SOM. This combined approach is first tested and illustrated on the problem of clustering handwritten digits where a modification of the Activation Maximisation method for use at the SOM output layer demonstrates the benefit of hierarchical growth in the GHSORM. In particular, the SOM Node Activation Maximisation method is used to visually represent the best approximation of each of the SOM nodes at the output layer. This demonstrates the improvement in representing difficult to separate digits in the hierarchical case. To test and measure the efficacy of the GHSORM hierarchical model in class and sub-class separation the method is applied to complex digital gene expression data. A cancer dataset, comprising of gene expression data that has samples of different classes and sub-classes, is used for this purpose. The GHSORM demonstrates robust capabilities to cluster and sub-cluster the different classes and subclasses of cancer, where the results are superior to both linear methods, currently in use, as well as the methods of its constituent algorithms.

Automatic Modulation Classification for MASK, MPSK, and MQAM Signals Based on Hierarchical Self-Organizing Map.

Automatic modulation classification (AMC) plays a fundamental role in common communication systems. Existing clustering models typically handle fewer modulation types with lower classification accuracies and more computational resources. This paper proposes a hierarchical self-organizing map (SOM) based on a feature space composed of high-order cumulants (HOC) and amplitude moment features. This SOM with two stacked layers can identify intrinsic differences among samples in the feature space without the need to set thresholds. This model can roughly cluster the multiple amplitude-shift keying (MASK), multiple phase-shift keying (MPSK), and multiple quadrature amplitude keying (MQAM) samples in the root layer and then finely distinguish the samples with different orders in the leaf layers. We creatively implement a discrete transformation method based on modified activation functions. This method causes MQAM samples to cluster in the leaf layer with more distinct boundaries between clusters and higher classification accuracies. The simulation results demonstrate the superior performance of the proposed hierarchical SOM on AMC problems when compared with other clustering models. Our proposed method can manage more categories of modulation signals and obtain higher classification accuracies while using fewer computational resources.

Social Network Community Detection by Combining Self-Organizing Maps and Genetic Algorithms

Social networks have become an important source of information from which we can extract valuable indicators that can be used in many fields such as marketing, statistics, and advertising among others. To this end, many research works in the literature offer users some tools that can help them take advantage of this mine of information. Community detection is one of these tools and aims to detect a set of entities that share some features within a social network. We have taken part in this effort, and we proposed an approach mainly based on pattern recognition techniques. The novelty of this approach is that we do not directly tackle the social networks to find these communities. We rather proceeded in two stages; first, we detected community cores through a special type of self-organizing map called the Growing Hierarchical Self-Organizing Map (GHSOM). In the second stage, the agglomerations resulting from GHSOM were grouped to retrieve the final communities. The quality of the final partition would be under the control of an evaluation function that is maximized through genetic algorithms. Our system was tested on real and artificial databases, and the obtained results are really encouraging.

Co-Maintained Database Based on Blockchain for IDSs: A Lifetime Learning Framework

Intrusion Detection System (IDS) is one of the most important approaches in cyber security to protect networks against both inner and outer threats. Apart from traditional networks, IDSs have been implemented in various emerging networks, such as mobile networks and Vehicle Ad hoc Networks (VANETs). However, a critical problem in IDSs is that the detection capacity is gradually decaying with the emergence of unknown attacks. It is necessary to constantly retrain IDSs with a more extensive database, but the security institutes usually lack the motivation to persistently update and maintain the database for public. Thus, in this paper, a lifetime learning framework is proposed for IDSs with a blockchain-based database (bc-DB). In the proposed framework, the blockchain-based database is multilaterally maintained by the security institutes and universities using Data Coins (DCoins) as the incentives. In addition, a Lifetime Learning IDS (LL-IDS) is further designed as the supplement of the bc-DB for common IDS users. For the LL-IDS, the Growing Hierarchical Self-Organizing Map with probabilistic relabeling (GHSOM-pr) having flexible and hierarchical architecture is employed as the classifier, which grows to make itself perfectly fit the changeable bc-DB. Security analysis and simulation experiments show that the proposed lifetime learning framework are both secure and effective in attacks detection.

Entropy-Defined Direct Batch Growing Hierarchical Self-Organizing Mapping for Efficient Network Anomaly Detection

This paper proposes a network anomaly detection model of direct batch growing hierarchical self-organizing mapping based on entropy, which facilitates clear topology representation for the asymmetrically-distributed data. Since the entropy-defined parameters dynamically vary with the incident dataset, that is, follow a data-adaptive manner, the proposed model is naturally valid in all cases with various data types. For fine-grained data distinguishing, a resemble entropy parameter is proposed for the first time to our best knowledge. The experimental results validate that the proposed model achieves a more efficient network anomaly detection than the conventional models, especially for real-world applications with unexpected anomaly data updating.

The two-phase scheduling based on deep learning in the Internet of Things

With the growth of data generation speed and its huge volume, the cloud-based infrastructure alone does not meet the needs of the Internet of Things (IoT), thereby leading to inefficiency. By combining fog and cloud computing on the IoT, some processing occurs near the data generation location with a higher speed and without needing a large bandwidth. Cloud and fog computing requires proper management in scheduling, increasing resource efficiency, and reducing consumption costs. In this article, a two-phase scheduling algorithm is presented based on deep learning methods in the field of IoT. The first phase is to decide on the location of the task execution using the clustering method. The second phase involves scheduling the task according to the execution location. In the clustering section, three ideas based on the Self-Organizing Map (SOM) clustering method have been proposed. In the first and second ideas, the SOM and hierarchical SOM are used to cluster the features of the tasks received from the IoT layer. In the third idea, the feature is extracted, and its dimensions are reduced using the Autoencoder, which is one of the deep learning methods, after which clustering is done. After scheduling each cluster's tasks, comparing the methods presented in the article, it is shown that the feature extraction using deep learning can improve clustering in such a way to reduce the missed rate of tasks in the cloud and fog, as well as their costs.

Rural\u2013urban scaling of age, mortality, crime and property reveals a loss of expected self-similar behaviour

The urban scaling hypothesis has improved our understanding of cities; however, rural areas have been neglected. We investigated rural–urban population density scaling in England and Wales using 67 indicators of crime, mortality, property, and age. Most indicators exhibited segmented scaling about a median critical density of 27 people per hectare. Above the critical density, urban regions preferentially attract young adults (25–40 years) and lose older people (> 45 years). Density scale adjusted metrics (DSAMs) were analysed using hierarchical clustering, networks, and self-organizing maps (SOMs) revealing regional differences and an inverse relationship between excess value of property transactions and a range of preventable mortality (e.g. diabetes, suicide, lung cancer). The most striking finding is that age demographics break the expected self-similarity underlying the urban scaling hypothesis. Urban dynamism is fuelled by preferential attraction of young adults and not a fundamental property of total urban population.

Information Visualization Using Clustering and Predictive Model

This paper aims to reveal the impact of rainfall on tea export from India, an issue that remained unexplored in the existing literature. This study explores a new model to predict India's tea export more accurately that would be helpful for Indian tea planters and exporters to plan their production as well as the inventory holding for deriving maximum value from tea export. A two-stage modelling approach has been developed. Firstly, an artificial intelligence-based growing hierarchical self-organising map algorithm is employed on the monthly time series of monthly frequency spreading over April 2005 to December 2013 to segregate India's monthly tea export data series into visual clusters of recognized pattern. Further, a predictive model using support vector machine has been developed and applied to forecast the tea export and then the importance of the predictor variables of the tea export have been identified. Finally, using the appropriate measures of performance a comparative analysis has been performed for each of the model. The newness of the study pertains to the two facts revealed from the study: firstly, India's tea export is embedded of complexity and nonlinearity, which could receive a successful clustering through growing hierarchical self organizing map that would make a deeper analysis easier with a further application of rich statistical techniques. Secondly, the analysis of prediction errors and the relative importance of the predictor variables establish rainfall as one of the most significant variable in predicting India's tea export, insight that has never surfaced in the literature developed thus far.

Direct Batch Growth Hierarchical Self-Organizing Mapping Based on Statistics for Efficient Network Intrusion Detection

A new evaluation mechanism was proposed to enhance the representation of data topology in the directed batch growth hierarchical self-organizing mapping. In the proposed mechanism, the growth threshold and the correlation worked in a case-sensitive manner through the statistic calculation of the input data. Since the proposed model enabled a more thorough representation of data topology from both the horizontal and the vertical directions, it naturally held great potential in detecting various traffic attacks. Numerical experiments of network intrusion detection were carried out on the datasets of KDD99, Moore and CICIDS2017, where the good performance validated the superiority of the proposed method.

Using Fuzzy Logic Algorithms and Growing Hierarchical Self-Organizing Maps to Define Efficient Security Inspection Strategies in a Container Terminal

Maritime transport is one of the oldest methods of moving various types of goods, and it continues to have an important role in our modern society. More than 20 million containers are transported across the oceans daily. However, this form of transportation is constantly threatened by illegal operations, such as the smuggling of goods or people and merchandise theft. Port security departments must be prepared to face the different threats and challenges that accompany the use of innovative techniques and devices to achieve efficient inspection strategies. Two inspection strategies are presented in this study. The first strategy is based on fuzzy logic (FL), and the second strategy is based on the growing hierarchical self-organizing map (GHSOM) approach. The weight variation and security index (SI) of a container and the readings from certain technologies, such as radio-frequency identification (RFID) and X-ray scanning, are considered as the input data. To minimize the inspection time and considering the costs associated with the security inspections of containers, the results of both inspection strategies are compared and analyzed. The findings indicate there is potential for improving the effectiveness of security inspections by employing both techniques, and the specific relevance in the case of GHSOMs is discussed.

A Survey on the Development of Self-Organizing Maps for Unsupervised Intrusion Detection

This paper describes a focused literature survey of self-organizing maps (SOM) in support of intrusion detection. Specifically, the SOM architecture can be divided into two categories, i.e., static-layered architectures and dynamic-layered architectures. The former one, Hierarchical Self-Organizing Maps (HSOM), can effectively reduce the computational overheads and efficiently represent the hierarchy of data. The latter one, Growing Hierarchical Self-Organizing Maps (GHSOM), is quite effective for online intrusion detection with low computing latency, dynamic self-adaptability, and self-learning. The ultimate goal of SOM architecture is to accurately represent the topological relationship of data to identify any anomalous attack. The overall goal of this survey is to comprehensively compare the primitive components and properties of SOM-based intrusion detection. By comparing with the two SOM-based intrusion detection systems, we can clearly understand the existing challenges of SOM-based intrusion detection systems and indicate the future research directions.

A new model for retinal lesion detection of diabetic retinopathy using hierarchical self-organizing maps

Diabetes is a disease that impairs blood flow throughout the body. In this disease, the retinal blood vessels may leak and cause retinal swelling known as edema. The person’s sight might be affected if this swelling happens in the central vision area of retina, the macula. In this paper, we proposed a classification system, including a novel combination of Self-Organizing Maps (SOM) for detecting retinal lesions. The proposed system consists of a fast pre-processing step followed by lesion feature extraction and, finally, a detailed classification model. In the pre-processing stage, the system is divided into the three procedures of initial target lesion extraction, optical disk extraction, and eventually extracting retinal blood vessels from the retina. The second step is a combination of multiple features such as morphology, color, intensity, and moments. The classifier is a model of Hierarchical Self-Organizing Maps (HSOM), which aims to increase the accuracy and speed of classifying the lesions while considering the high amount of data in extracting the features. The overall accuracy and sensitivity of the proposed method according to the MESSIDOR database is 97.87% and 98.51%, respectively. The results show that the proposed model can detect and classify the Lesions in HDR images accurately.

Multiple-fault detection and identification scheme based on hierarchical self-organizing maps applied to an electric machine

Strategies of condition monitoring applied to electric motors play an important role in the competitiveness of multiple industrial sectors. However, the risk of faults coexistence in an electric motor and the overlapping of their effects in the considered physical magnitudes represent, currently, a critical limitation to provide reliable diagnosis outcomes. In this regard, additional investigation efforts are required towards high-dimensional data fusion schemes, particularly over the features calculation and features reduction, which represent two decisive stages in such data-driven approaches. In this study, a novel multiple-fault detection and identification methodology supported by a feature-level fusion strategy and a Self-Organizing Maps (SOM) hierarchical structure is proposed. The condition diagnosis as well as the corresponding estimated probability are obtained. Moreover, the proposed method allows the visualization of the results while preserving the underlying physical phenomenon of the system under monitoring. The proposed scheme is performed sequentially; first, a set of statistical-time based features is estimated from different physical magnitudes. Second, a hybrid feature reduction method is proposed, composed by an initial soft feature reduction, based on sequential floating forward selection to remove the less informative features, and followed by a hierarchical SOM structure which reveals directly the diagnosis and probability assessment. The effectiveness of the proposed detection and identification scheme is validated with a complete set of experimental data including healthy and five faulty conditions. The accuracy’s results are compared with classical condition monitoring approaches in order to validate the competency of the proposed method.

Efficient Segmentation of Brain Tumor Using FL-SNM with a Metaheuristic Approach to Optimization.

Nowadays, automatic tumor detection from brain images is extremely significant for many diagnostic as well as therapeutic purposes, due to the unpredictable shape and appearance of tumors. In medical image analysis, the automatic segmentation of tumors from brain using magnetic resonance imaging (MRI) data is the most critical issue. Existing research has some limitations, such as high processing time and lower accuracy, because of the time required for the training process. In this research, a new automatic segmentation process is introduced using machine learning and a swarm intelligence scheme. Here, a fuzzy logic with spiking neuron model (FL-SNM) is proposed for segmenting the brain tumor region in MR images. Initially, input images are preprocessed to remove Gaussian and Poisson noise using a modified Kuan filter (MKF). In the MKF, the optimal selection of the minimum MSE of image pixels is achieved using a random search algorithm (RSA), which improves the peak signal-to-noise ratio (PSNR). Then, the image is smoothed using an anisotropic diffusion filter (ADF) to reduce the over-filtering problem. Afterwards, to extract statistical texture features, Fisher's linear-discriminant analysis (FLDA) is used. Finally, extracted features are transferred to the FL-SNM process and this scheme effectively segments the tumor region. In FL-SNM, the consequent parameters such as weight and bias play an important role in segmenting the region. Therefore, optimizing the weight parameter values using a chicken behavior-based swarm intelligence (CSI) algorithm, is proposed. The proposed (FL-SNM) scheme attained better performance in terms of high accuracy (94.87%), sensitivity (92.07%), specificity (99.34%), precision rate (89.36%), recall rate (88.39%), F-measure (95.06%), G-mean (95.63%), and DSC rate (91.2%), compared to existing convolutional neural networks (CNNs) and hierarchical self-organizing maps (HSOMs).

A novel Intrusion Detection System for Vehicular Ad Hoc Networks (VANETs) based on differences of traffic flow and position

Vehicle ad hoc networks (VANETs) have attracted great interests from both industry and academia, but a number of issues, particularly security, have not been readily addressed. Intrusion Detection System (IDS) as one of the most important approaches to protect network security has been studied adequately in previous literatures. However, the performance of IDSs still needs to be improved to adapt the scenario of VANETs which are very fast moving and highly dynamic. In this paper, we propose a novel IDS that is able to be appropriately used in the wireless and dynamic networks, like VANETs. It mainly contains a novel feature extraction algorithm and a classifier based on an improved growing hierarchical self-organizing map (I-GHSOM) for IDS in VANETs. The proposed feature extraction algorithm is used to quickly extract distinct features from vehicle messages for IDS’s training and test. In the proposed algorithm, two key features including the differences of traffic flow and of position are extracted. The former feature is calculated according to the range of the distance between vehicles, while both a voting filter mechanism and a semi-cooperative mechanism are designed to get the latter feature. Furthermore, in the I-GHSOM-based classifier, for quickly attaining precise classification results, two novel mechanisms (relabeling and recalculating mechanisms) are proposed to relabel the units of GHSOM and check whether the balance of GHSOM structure is broken or not. Simulation results show that the proposed IDS is better than others in the measurement of accuracy, stability, processing efficiency and message scales.

Spark-GHSOM: Growing Hierarchical Self-Organizing Map for large scale mixed attribute datasets

The Growing Hierarchical Self-Organizing Map (GHSOM) algorithm has shown its potential for performing several tasks such as exploratory analysis, anomaly detection and forecasting on a variety of domains including the financial and cyber-security domains. GHSOM is a dynamic variant of the SOM algorithm which generates a multi-level hierarchy of SOM maps based solely on input data. However, in order to generate this multi-level structure, GHSOM requires multiple iterations over the input dataset, thus making it intractable on large datasets. Moreover, the conventional GHSOM algorithm is designed to handle datasets with numeric attributes only. This represents an important limitation as most modern real-world datasets are characterized by mixed attributes - numerical and categorical. In this work, we propose an extension of the conventional GHSOM algorithm called Spark-GHSOM, which exploits the Spark platform to process massive datasets in a distributed manner. Moreover, we leverage a method known as the distance hierarchy approach to modify the optimization function of GHSOM so that it can (also) coherently handle mixed-attribute datasets. We test our new method with respect to accuracy, scalability and descriptive power. The results obtained using different datasets demonstrate the superior predictive and descriptive capabilities of Spark-GHSOM, as well as its applicability to large-scale datasets which could not be analyzed before.

An investigation on establishing small- and medium-sized enterprises communities under the environment of social manufacturing

Manufacturing industry has become professionalization, socialized, service-oriented, and collaborative, so lots of professional and socialized small- and medium-sized enterprises spring up to provide product–service for satisfying customers’ requirements. Faced with this trend, a new manufacturing mode called social manufacturing has been proposed to aggregate these small- and medium-sized enterprises into communities for mass individualization manufacturing. As the basis of social manufacturing, small- and medium-sized enterprises communities are established by two steps. First, the socialized manufacturing resources of the small- and medium-sized enterprises are described and similar socialized manufacturing resources are clustered into one socialized manufacturing resources community by growing hierarchical self-organizing map algorithm. Suitable socialized manufacturing resources communities are selected and then the socialized manufacturing resources are mapped into small- and medium-sized enterprises. Second, these small- and medium-sized enterprises form a community to complete the order together. Product cost and delivery time serve as the indicator for order allocating in the small- and medium-sized enterprises community. A multi-objective algorithm is proposed to tackle the order allocation problem. A case from a professional printing firm is analyzed to validate the proposed methodology and model. Some rules are revealed and these rules are useful for guiding practical production. The study endorses the establishing small- and medium-sized enterprises communities under the environment of social manufacturing and suggests extensive practical implications.