introduction Share on Introduction to Special Issue on Misinformation, Fake News and Rumor Detection in Low-Resource Languages Authors: Akshi Kumar View Profile , Christian Esposito View Profile , Dimitrios A. Karras View Profile Authors Info & Claims ACM Transactions on Asian and Low-Resource Language Information ProcessingVolume 21Issue 1January 2022 Article No.: 1epp 1–3https://doi.org/10.1145/3505588Online:24 December 2021Publication History 0citation247DownloadsMetricsTotal Citations0Total Downloads247Last 12 Months247Last 6 weeks39 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

Training a multilayer perceptron (MLP) with algorithms employing global search strategies has been an important research direction in the field of neural networks. Despite a number of significant results, an important matter concerning the bounds of the search region---typically defined as a box---where a global optimization method has to search for a potential global minimizer seems to be unresolved. The approach presented in this paper builds on interval analysis and attempts to define guaranteed bounds in the search space prior to applying a global search algorithm for training an MLP. These bounds depend on the machine precision and the term guaranteed denotes that the region defined surely encloses weight sets that are global minimizers of the neural network's error function. Although the solution set to the bounding problem for an MLP is in general non-convex, the paper presents the theoretical results that help deriving a box which is a convex set. This box is an outer approximation of the algebraic solutions to the interval equations resulting from the function implemented by the network nodes. An experimental study using well known benchmarks is presented in accordance with the theoretical results.

This research deals with several key issues concerning WSN design process. Due to the nature of these networks, debugging after deployment is unlikely, thus a wellorganized testing methodology is required. WSN simulators could achieve such a task, but still code implementing mote sensing and RF behaviour consists of layered and/or interacting protocols that for the sake of designing accuracy are tested working as a whole, running on specific hardware. Simulators that provide cross layer simulation and hardware cross platform emulation options may be regarded as a significant milestone for improved WSN design process. The herein proposed multi-agent simulation architecture aims at designing a novel generic WSN simulation system independent of specific hardware platforms but taking into account all hardware entities and sensor network events for testing and analysing the behaviour of a pragmatic WSN system. Details of the implementation are provided as well as a preliminary validation of the simulator in C# involving motes based on two different hardware platforms, randomly deployed over WSN motes.

This paper presents new strategies for offline traffic engineering, when designing multi-service UMTS networks, based on an efficient modification of basic genetic optimization techniques. More specifically, the problem setup includes Mobile Internet access as the most emergent trade in the telecommunications industry due to the advancement of cellular smartphones, tablets, notebooks and netbooks based computing and communicating technologies. In this real world set-up, the integration of deployed wireless networks with the Internet and the transparent interworking among different wireless technologies, namely EDGE/GPRS/CDMA2000 cellular and Wireless LAN (WLAN), appears to be a demanding objective for mobile communications service providers. The mixture and convergence of mobile and fixed infrastructures related services make it urgent to tackle quality of service and traffic engineering for mobile networks at the fundamental transport level with regards to OSI network analysis. A good solution to address such issues is the Multi-Protocol Label Switching (MPLS) technology providing a fused control instrument with connectionless multiprotocol capabilities, functioning over different communication media while supporting traffic engineering and quality of service (QoS) offering smooth traffic over the complex network of mobile and fixed communication infrastructures. Therefore, the MPLS architecture offers the required capacity to integrate different services over wired and wireless infrastructures in a unified convergent approach. Based on these concepts, this research proposal investigates traffic engineering solutions in MPLS involved wireless cellular networks. In this paper, two new methods for the solution of the off-line Traffic Engineering (TE) problem in multi-service wireless networks based on a certain efficient modification of basic genetic optimisation are presented. In the first method the off-line TE problem is formulated as an optimisation model with linear constraints and then solved using a new modified version of the Genetic Algorithm for Numerical Optimisation for Constraint Problems (GENOCOP). Besides, a hybrid method for the solution of the aforementioned problem involving modified GENOCOP and a heuristic TE algorithm is also provided. The performance of the above methods against a standard LP-based optimisation method is examined in terms of two different network topologies mixing both wireless and wired links and numerical test results are provided.

Reliable estimation of a neural network's domain of validity is important for a number of reasons such as assessing its ability to cope with a given problem, evaluating the consistency of its generalization etc. In this paper we introduce a new approach to estimate the domain of validity of a neural network based on Set Inversion Via Interval Analysis (SIVIA), the methodology established by Jaulin and Walter [1]. This approach was originally introduced in order to solve nonlinear parameter estimation problems in a bounded error context and proved to be effective in tackling several types of problems dealing with nonlinear systems analysis. The dependence of a neural network output on the pattern data is a nonlinear function and hence derivation of the impact of the input data to the neural network function can be addressed as a nonlinear parameter estimation problem that can be tackled by SIVIA. We present concrete application examples and show how the proposed method allows to delimit the domain of validity of a trained neural network. We discuss advantages, pitfalls and potential improvements offered to neural networks.

A novel approach for spam e-mail filtering is herein considered based on information theoretic extraction of higher order features and the committee machines neural network models. An extensive experimental study is organized, the most extensive so far in the literature, based on widely accepted benchmarking e-mail data sets, comparing the proposed methodology with the Naïve Bayes spam filter as well as with the Boosting tree methodology, the linear models-based classification (classification via regression) and the nonlinear models-based classification using simple neural network models, including Multilayer Perceptrons. Moreover, several feature extraction approaches based on information theory are evaluated, comparing mainly the proposed higher order feature extraction methodology with information theoretic extraction of single features. It is shown that the former outperforms the latter and, moreover, that the proposed information theoretic Boolean features present a remarkably high spam categorization performance compared to that of their analog counterparts. Finally, it is shown that the committee machines mail categorization performance compares very favorably to the other rival methods' performance, including the Bayes spam filter which is the most widely used approach in the e-mail services market.

A novel approach for spam e-mail filtering is herein considered based on the committee machines neural network models and on information theoretic feature extraction. An extensive experimental study is organized, the most extensive so far in the literature, based on widely accepted benchmarking e-mail data sets, comparing the proposed methodology with the naive Bayes spam filter as well as with the boosting tree methodology, the linear models based classification (classification via regression) and the nonlinear models based classification using simple neural network models, including multilayer perceptrons. Moreover, several feature extraction approaches based on information theory are evaluated. It is shown that the committee machines mail categorization performance is compared very favorably to the other rival methods performance, including the Bayes spam filter which is the most widely used approach in the e-mail services market. It is, also, found that the proposed information theoretic Boolean features present a remarkably high spam categorization performance compared to their analog counterparts performance.

A novel approach for continuous function approximation using a two-stage neural network model, involving regularization techniques, is herein presented. The suggested method can be applied to real functions of many variables as in robot path planning problems. It involves a regularized Kohonen feature map (SOFM) in the first stage which aims at quantizing the input variable space into smaller regions representative of the input space probability distribution and preserving its original topology, while increasing, on the other hand, cluster distances. This is achieved through adapting not only the winning neuron and its neighboring neurons weights but, also, losing neurons weights during map's convergence phase. Losing neurons weights are adapted in a manner similar to that of LVQ, by increasing the distance between these weights vectors and the corresponding input data vectors. During convergence phase of the map a group of support vector machines (SVM), associated with its codebook vectors, is simultaneously trained in an online fashion so that each SVM learns to respond when the input data belongs to the topological space represented by its corresponding codebook vector. Moreover, these SVMs follow a task specific regularization strategy which aims at incorporating additional information in their training process. The proposed methodology is applied to the design of a neural-adaptive controller, by involving the computer-torque approach, which combines the regularized two-stage neural network model with a servo PD feedback controller. For this task, the regularization technique aims at filtering SVMs outputs so that their values become closer to that of a PD feedback controller, while compensating the nonlinear terms of the error, as regards the estimated torque, introduced in the robotic manipulator by employing the PD controller.

A novel procedure based on genetic algorithms and multilayer perceptron (MLP) neural networks (NN) is presented for the evaluation and production of digital fingerprints in the design of secure communication systems. These digital fingerprints are computed using the methodology of un-keyed one-way functions (hash functions). The problem of evaluating the quality of such functions is formulated as a global optimization one in the space spanned by all possible messages and is approached from a practical viewpoint by involving genetic algorithms and MLP neural networks, contrary to the very few similar research efforts existing in the literature that are of only theoretical interest. Moreover, the problem of producing digital fingerprints of good quality for use in communication systems is formulated in terms of a hash function constructed by involving the genetic algorithm procedure, exclusively utilizing the crossover operator and the steady-state reproduction method and omitting its random components, as well as by involving a suitably designed MLP NN. An extensive experimental study is conducted to assess the quality of the results produced by applying the herein suggested two novel approaches, employing SHA, the known one way hash function algorithm, in the comparison procedure. The promising results herein obtained illustrate the importance of applying genetic algorithms and neural networks in communication systems security design.

Three new methods for the solution of the offline traffic engineering. (TE) problem in multi-service networks based on genetic optimisation and simulated annealing optimization techniques are presented and compared. In the first method, the off-line TE problem is formulated as an optimisation model with linear constraints and then solved using the Genetic Algorithm for Numerical Optimisation for Constraint Problems (GENOCOP). In the second method the same problem is solved using simulated annealing. Besides, a third hybrid method for the solution of the aforementioned problem involving GENOCOP and a heuristic TE algorithm is also provided. The performance of the above methods against a standard LP-based optimisation method is examined in terms of two different network topologies and numerical test results are provided. The contribution of the paper lies on the fact that for the first time genetic optimization and simulated annealing methods are involved in traffic engineering problems. In addition, a novel hybrid method based on genetic optimization is proposed with performance comparable to that obtained by linear programming techniques (Simplex), which are the optimum solvers in the case of linear cost functions optimization under linear constraints as it takes place in the herein proposed traffic engineering problem formulations. Finally, the contribution of the paper is that for the first time genetic optimization and simulated annealing techniques are used to solve real world problems of thousands of variables, achieving in the case of genetic algorithms, near optimal results.

This paper aims at investigating a novel solution to the problem of defect detection from images using the Support Vector Machines (SVM) classification approach, that can find applications in the design of robust quality control systems for the production of furniture, textile, integrated circuits, etc. The suggested solution focuses on detecting defects for manufacturing applications from their wavelet transformation and vector quantization related properties of the associated wavelet coefficients. More specifically, a novel methodology is investigated for discriminating defects by applying a supervised neural classification technique, namely SVM, to innovative multidimensional wavelet based feature vectors. These vectors are extracted from the K-Level 2-D DWT (Discrete Wavelet Transform) transformed original image using Vector Quantization techniques and a Singular Value Decomposition (SVD) Analysis applied to these wavelet domain quantization vectors. The results of the proposed methodology are illustrated in defective textile images where the defective areas are recognized with higher accuracy than the one obtained by applying two rival defect detection methodologies. The first one of them uses all the wavelet coefficients derived from the k-Level 2-D DWT and involves SVM again in the classification stage, while the second one uses again all the wavelet coefficients derived from the k-Level 2-D DWT but involves a Multilayer Perceptron (MLP) neural network in the classification stage. The promising results herein shown outline the importance of judicious selection and processing of 2-D DWT wavelet coefficients for industrial pattern recognition applications as well as the generalization performance benefits obtained by involving SVM neural networks instead of other ANN models.

The goal of this paper is to present the development of a new reconstruction methodology for restoring magnetic resonance images (MRI) from reduced scans in k-space. The proposed approach considers the combined use of regularized neural network models and Bayesian restoration, in the problem of MRI image extraction from sparsely sampled k-space, following several different sampling schemes, including spiral and radial. Effective solutions to this problem are indispensable especially when dealing with MRI of dynamic phenomena since then, rapid sampling in k-space is required. The goal is such a case is to make measurement time smaller by reducing scanning trajectories as much as possible. In this way, however, underdetermined equations are introduced and poor image reconstruction follows. It is suggested here that significant improvements could be achieved concerning quality of the extracted image, by judiciously applying regularization based neural networks and Bayesian estimation methods to the k-space data. More specifically, it is demonstrated that regularization based neural network techniques could construct efficient priors and introduce them in the procedure of Bayesian reconstruction. These regularized ANN priors are independent of specific image properties and probability distributions. They are based on training supervised multilayer perceptron (MLP) regularized neural filters to estimate the missing samples of complex k-space and thus, to improve k-space information capacity. Such a neural filter based prior is integrated to the maximum likelihood procedure involved in the Bayesian reconstruction. It is found that the proposed methodology leads to enhanced image extraction results favorably compared to the ones obtained by the traditional Bayesian MRI reconstruction approach as well as by the pure MLP based reconstruction approach.

The strength of data integrity, message authentication and pseudonym generation mechanisms in the design of secure multimedia communication applications over the Internet relies on the quality of the message digest algorithms used in the digital signatures construction/verification process. In this paper, we propose neural network based evaluation benchmarks to assess the message digest function quality since there is lack of practical tests to be applied to message digest algorithms in the emerging field of designing secure information and communication systems especially for the delivery of multimedia content, where the issues of copyright protection and security in transactions are outstanding. These assessment tests are suggested here along with other ones derived from well known statistical and information theoretic methods, such as entropy test, and thus comprise a suitable practical evaluation methodology.

Random components play an especially important role in secure electronic commerce and Internet communications. For this reason, the existence of strong pseudo random number generators is highly required. This paper presents novel techniques, which rely on artificial neural network architectures, to strengthen traditional generators such as ANSI X.9 based DES and IDEA. Additionally, this paper proposes a test method for evaluating the required non-predictability property, which also relies on neural networks. This non-predictability test method along with commonly used statistical and non-linearity tests are proposed as methodology for the evaluation of strong pseudo random number generators. By means of this methodology, traditional and proposed generators are evaluated. The results show that the proposed generators behave significantly better than the traditional, in particular, in terms of non-predictability.

We present an approach to the detection of tumors in colonoscopic video. It is based on a new color feature extraction scheme to represent the different regions in the frame sequence. This scheme is built on the wavelet decomposition. The features named as color wavelet covariance (CWC) are based on the covariances of second-order textural measures and an optimum subset of them is proposed after the application of a selection algorithm. The proposed approach is supported by a linear discriminant analysis (LDA) procedure for the characterization of the image regions along the video frames. The whole methodology has been applied on real data sets of color colonoscopic videos. The performance in the detection of abnormal colonic regions corresponding to adenomatous polyps has been estimated high, reaching 97% specificity and 90% sensitivity.

Video processing for classification applications in medical imaging is an area with great importance. In this paper a framework for classification of suspicious lesions using the video produced during an endoscopic session is presented. The proposed approach is based on a feature extraction scheme that uses second order statistical information of the wavelet transformation. These features are used as input to a multilayer feedforward neural network (MFNN) architecture, which has been trained using features of normal and tumor regions. The system uses a limited number of frames with a rather small population of training vectors. The classification results are promising, since the system has been proven to be capable to classify and locate regions, that correspond to lesions with a success of 94 up to 99%, in a sequence of the video-frames. The proposed methodology can be used as a valuable diagnostic tool that may assist physicians to identify possible tumor regions or malignant formations.

A few approaches have been presented in the literature towards the discrimination of texture in medical images. Medical experts proposed that the more valuable information for discriminating among normal and suspicious cancer regions in endoscopic images is the texture of the examined tissue. Texture can be encoded by a number of mathematical descriptors. Three well-known textural descriptors, as well as a new wavelet-based one are used in this paper for an accurate study and evaluation of the methodologies encountered. Experiments conducted include tests with various images from the Brodatz album, as well as interpretation of tissue regions in endoscopic image. In all cases the recognition task is supported by multilayer perceptron type neural network architectures.

This paper concerns a novel application of neural networks to magnetic resonance imaging (MRI) by considering regularized neural network models for the problem of image reconstruction from sparsely sampled k-space. Effective solutions to this problem are indispensable especially when dealing with MRI of dynamic phenomena since then, rapid sampling in k-space is required. The goal in such a case is to reduce the measurement time by omitting as many scanning trajectories as possible. This approach, however, entails underdetermined equations and leads to poor image reconstruction. It is proposed here that significant improvements could be achieved concerning image reconstruction if a procedure, based on neural network function approximation methodology and involving regularization techniques, for estimating the missing samples of complex k-space were introduced. To this end, the viability of involving neural network algorithms with/without regularization for such a problem is considered and it is found that their image reconstruction results are very favorably compared to the ones obtained by the trivial zero-filled k-space approach or traditional more sophisticated interpolation approaches. Moreover, it is found that regularized multilayer perceptrons outperform the ones not involving regularization during their training.

A novel algorithm is presented which supplements the training phase in feedforward networks with various forms of information about desired learning properties. This information is represented by conditions which must be satisfied in addition to the demand for minimization of the usual mean square error cost function. The purpose of these conditions is to improve convergence, learning speed, and generalization properties through prompt activation of the hidden units, optimal alignment of successive weight vector offsets, elimination of excessive hidden nodes, and regulation of the magnitude of search steps in the weight space. The algorithm is applied to several small- and large-scale binary benchmark training tasks, to test its convergence ability and learning speed, as well as to a large-scale OCR problem, to test its generalization capability. Its performance in terms of percentage of local minima, learning speed, and generalization ability is evaluated and found superior to the performance of the backpropagation algorithm and variants thereof taking especially into account the statistical significance of the results.

A general method for designing two-layer neural networks which can be used as pattern classifiers is presented. The method is based on coding techniques. It is independent of the number of pattern elements and results in a minimal-cost implementation for correcting and detecting random errors from black to white or from white to black. It can also detect all the unidirectional errors from white to black and all the unidirectional errors from black to white, preventing incorrectly recognized input images. It reduces the number of connections needed by the resulting neural networks by more than 80%, as is shown in an example. compared with the other currently available methods