On Modelling the Structural Quasiness of Complex Systems

Complex systems are usually represented by invariant models which at most admit only parametric variations. This approach assumes invariant idealized simplifications to model these systems. This standard approach is considered omitting crucial features of phenomenological interaction mechanisms related to processes of emergence of such systems. The quasiness of the structural dynamics that generate emergence of complex systems is considered as the main feature. Generation achieved through prevalently coherent sequences and combinations of interactions. Quasiness (dynamics of loss and recovery, equivalences, inhomogeneity, multiplicity, non-regularity, and partiality) represents the incompleteness of the interaction mechanisms, incompleteness necessary even if not sufficient for the establishment of processes of emergence. The emergence is extinguished by completeness. Complex systems possess local coherences corresponding to the phenomenological complexity. While quasi-systems are not necessarily complex systems, complex systems are considered quasi-systems, being not always systems, not always the same system, and not only systems. It is addressed the problem of representing the quasiness of coherence (quasicoherence), such as the ability to recover and tolerate temporary levels of incoherence. The main results of the study focus on research approaches to model quasicoherence through the changing of rules in models of emergence. It is presented a version of standard analytical approaches compatible with quasiness of systemic emergence and related mathematical issues. The same approach is considered for networks, artificial neural networks, and it is introduced the concept of quasification for fixed models. Finally, it is considered that suitable representations of structural dynamics and its quasiness are needed to model, simulate, and adopt effective interventions on emergence of complex systems.

Multilevel Deep Feature Generation Framework for Automated Detection of Retinal Abnormalities Using OCT Images

Optical coherence tomography (OCT) images coupled with many learning techniques have been developed to diagnose retinal disorders. This work aims to develop a novel framework for extracting deep features from 18 pre-trained convolutional neural networks (CNN) and to attain high performance using OCT images. In this work, we have developed a new framework for automated detection of retinal disorders using transfer learning. This model consists of three phases: deep fused and multilevel feature extraction, using 18 pre-trained networks and tent maximal pooling, feature selection with ReliefF, and classification using the optimized classifier. The novelty of this proposed framework is the feature generation using widely used CNNs and to select the most suitable features for classification. The extracted features using our proposed intelligent feature extractor are fed to iterative ReliefF (IRF) to automatically select the best feature vector. The quadratic support vector machine (QSVM) is utilized as a classifier in this work. We have developed our model using two public OCT image datasets, and they are named database 1 (DB1) and database 2 (DB2). The proposed framework can attain 97.40% and 100% classification accuracies using the two OCT datasets, DB1 and DB2, respectively. These results illustrate the success of our model.

LEDPatNet19: Automated Emotion Recognition Model based on Nonlinear LED Pattern Feature Extraction Function using EEG Signals

Electroencephalography (EEG) signals collected from human brains have generally been used to diagnose diseases. Moreover, EEG signals can be used in several areas such as emotion recognition, driving fatigue detection. This work presents a new emotion recognition model by using EEG signals. The primary aim of this model is to present a highly accurate emotion recognition framework by using both a hand-crafted feature generation and a deep classifier. The presented framework uses a multilevel fused feature generation network. This network has three primary phases, which are tunable Q-factor wavelet transform (TQWT), statistical feature generation, and nonlinear textural feature generation phases. TQWT is applied to the EEG data for decomposing signals into different sub-bands and create a multilevel feature generation network. In the nonlinear feature generation, an S-box of the LED block cipher is utilized to create a pattern, which is named as Led-Pattern. Moreover, statistical feature extraction is processed using the widely used statistical moments. The proposed LED pattern and statistical feature extraction functions are applied to 18 TQWT sub-bands and an original EEG signal. Therefore, the proposed hand-crafted learning model is named LEDPatNet19. To select the most informative features, ReliefF and iterative Chi2 (RFIChi2) feature selector is deployed. The proposed model has been developed on the two EEG emotion datasets, which are GAMEEMO and DREAMER datasets. Our proposed hand-crafted learning network achieved 94.58%, 92.86%, and 94.44% classification accuracies for arousal, dominance, and valance cases of the DREAMER dataset. Furthermore, the best classification accuracy of the proposed model for the GAMEEMO dataset is equal to 99.29%. These results clearly illustrate the success of the proposed LEDPatNet19.

An Optimized Machine Learning and Big Data Approach to Crime Detection

Crime detection is one of the most important research applications in machine learning. Identifying and reducing crime rates is crucial to developing a healthy society. Big Data techniques are applied to collect and analyse data: determine the required features and prime attributes that cause the emergence of crime hotspots. The traditional crime detection and machine learning-based algorithms lack the ability to generate key prime attributes from the crime dataset, hence most often fail to predict crime patterns successfully. This paper is aimed at extracting the prime attributes such as time zones, crime probability, and crime hotspots and performing vulnerability analysis to increase the accuracy of the subject machine learning algorithm. We implemented our proposed methodology using two standard datasets. Results show that the proposed feature generation method increased the performance of machine learning models. The highest accuracy of 97.5% was obtained when the proposed methodology was applied to the Naïve Bayes algorithm while analysing the San Francisco dataset.