A Quantum-Inspired Genetic Algorithm for Extractive Text Summarization

Automatic text summarization has recently become a key instrument for reducing the huge quantity of textual data. In this paper, the authors propose a quantum-inspired genetic algorithm (QGA) for extractive single-document summarization. The QGA is used inside a totally automated system as an optimizer to search for the best combination of sentences to be put in the final summary. The presented approach is compared with 11 reference methods including supervised and unsupervised summarization techniques. They have evaluated the performances of the proposed approach on the DUC 2001 and DUC 2002 datasets using the ROUGE-1 and ROUGE-2 evaluation metrics. The obtained results show that the proposal can compete with other state-of-the-art methods. It is ranked first out of 12, outperforming all other algorithms.

Symmetric and Asymmetric Encryption Algorithm Modeling on CPU Execution Time as Employed Over a Mobile Environment

This paper presents results on modelling of AES and RSA encryption algorithms in terms of CPU execution time, considering different modelling techniques such as linear, quadratic, cubic, and exponential mathematical models, each with the application of piecewise approximations. C#.net framework is used to implement this study. The authors consider the symmetric encryption algorithm named AES and the asymmetric encryption algorithm named RSA to carry out this study. This study recommends quadratic piecewise approximation modelling as the most optimized model for modelling the CPU execution time of AES and RSA towards encryption of data files. The model proposed in this study can be extended to other symmetric and asymmetric encryption algorithms, besides taking them over a mobile cloud environment.

Development of a Predictive Model for Textual Data Using Support Vector Machine Based on Diverse Kernel Functions Upon Sentiment Score Analysis

This research work specifically focusses on the development of a predictive model for movie review data using support vector machine (SVM) classifier with its improvisations using different kernel functions upon sentiment score estimation. The predictive model development proceeds with user level data input with the data processing with the data stream for analysis. Then formal calculation of TF-IDF evaluation has been made upon data clustering using simple k-means algorithm. Once the labeled data has been sorted out, then the SVM with kernel functions corresponding to linear, sigmoid, rbf, and polynomial have been applied over the clustered data with specific parameter setting for each type of library functions. Performance of each of the kernels has been measured using precision, recall, and F-score values for each of the specified kernel, and from the analysis, it has been found that sentiment analysis using SVM linear kernel with sentiment score analysis has been found to provide an improved accuracy of about 91.18%.

Decoding Algorithm by Cooperation Between Hartmann Rudolph Algorithm and a Decoder Based on Syndrome and Hash

In this paper, the authors present a concatenation of Hartmann and Rudolph (HR) partially exploited and a decoder based on hash techniques and syndrome calculation to decode linear block codes. This work consists firstly to use the HR with a reduced number of codewords of the dual code then the HWDec which exploits the output of the HR partially exploited. Researchers have applied the proposed decoder to decode some Bose, Chaudhuri, and Hocquenghem (BCH) and quadratic residue (QR) codes. The simulation and comparison results show that the proposed decoder guarantees very good performances, compared to several competitors, with a much-reduced number of codewords of the dual code. For example, for the BCH(31, 16, 7) code, the good results found are based only on 3.66% of the all codewords of the dual code space, for the same code the reduction rate of the run time varies between 78% and 90% comparing to the use of Hartmann and Rudolph alone. This shows the efficiency, the rapidity, and the reduction of the memory space necessary for the proposed concatenation.

Texture Features in Palmprint Recognition System

This paper describes the algorithm used for personal identification based on features extracted from the palmprint. The local Gabor XOR (LGXP) features is built using Gabor filter with orientation. Initially, the palm print images are preprocessed using median filter. The algorithm is then modified, where features are extracted with different orientations of the Gabor filter called the multiple orientation LGXP (MOLGXP) features. The PCA feature is extracted and fused with MOLGXP and PCA using sum rule.

Estimation of ASR Parameterization for Interactive System

In this study, the authors explore the integration of speaker-independent automatic Amazigh speech recognition technology into interactive applications to extract data remotely from a distance database. Based on the combined interactive voice response (IVR) and automatic speech recognition (ASR) technologies, the authors built an interactive speech system to allow users to interact with the interactive system through voice commands. The hidden Markov models (HMMs), Gaussian mixture models (GMMs), and Mel frequency spectral coefficients (MFCCs) are used to develop a speech system based on the ten first Amazigh digits and six Amazigh words. The best-obtained performance is 89.64% by using 3 HMMs and 16 GMMs.

An Unsupervised Hybrid Symbolic Fuzzy Clustering Approach for Efficient Sclera Segmentation

This article performs the sclera segmentation task by proposing a new hybrid symbolic fuzzy c-means (HSFCM) clustering method. Practically, though the data point exhibits some sort of similarity, unfortunately they are not undistinguishable and exhibit some sort of dissimilarity. Thus, to capture these disparities, the proposed work uses symbolic interval valued representation method. Further, to handle uncertainty and imprecision, the paper has proposed to use symbolic fuzzy clustering methods. To assess the performance of the proposed method, extensive experimentation is conducted on SSRBC2016 dataset. The proposed clustering method is compared with existing FCM, KFCM, RSKFCM method in terms of cluster validity indices and accuracy. The obtained outcomes demonstrated that the proposed method performed better compared to the contemporary methods.

Classification of Rusty and Non-Rusty Images

Rust detection is necessary for proper working and maintenance of machines for security purposes. Images are one of the suggested platforms for rust detection in which rust can be detected even though the human can't reach to the area. However, there are a lack of online databases available that can provide a sizable dataset to identify the most suitable model that can be used further. This paper provides a data augmentation technique by using Perlin noise, and further, the generated images are tested on standard features (i.e., statistical values, entropy, along with SIFT and SURF methods). The two most generalized classifiers, naïve Bayes and support vector machine, are identified and tested to obtain the performance of classification of rusty and non-rusty images. The support vector machine provides better classification accuracy, which also suggests that that the combined features of statistics, SIFT, and SURF are able to differentiate the images. Hence, it can be further used to detect the rust in different parts of machines.

Automatic Facial Expression Recognition System Using Shape-Information-Matrix (SIM)

Automatic recognition of facial expressions and modeling of human expressions are very essential in the field of affective computing. The authors have introduced a novel geometric and texture-based method to extract the shapio-geometric features from an image computed by landmarking the geometric locations of facial components using the active appearance model (AAM). Expression-specific analysis of facial landmark points is carried out to select a set of landmark points for each expression to identify features for each specific expression. The shape information matrix (SIM) is constructed the set salient landmark points assign to an expression. Finally, the histogram-oriented gradients (HoG) of SIM are computed which is used for classification with multi-layer perceptron (MLP). The proposed method is tested and validated on four well-known benchmark databases, which are CK+, JAFFE, MMI, and MUG. The proposed system achieved 98.5%, 97.6%, 96.4%, and 97.0% accuracy in CK+, JAFFE, MMI, and MUG database, respectively.

A Power-Efficient and Quantum-Resistant N-Bit Cryptography Algorithm

With rapid technological advancements and enhanced network growth, security contends to play a crucial role. A powerful network security tends to point out diverse mixture of threats and intimidations and blocks them from creeping and getting circulated into the network to preserve the reliability, confidentiality, integrity, and accessibility of computer networks by annihilating illegitimate admittance and corruption of critical information. Secure hash algorithms (SHA) are cryptographic hash functions used to produce a hash value of fixed output bit sizes. In this paper, an algorithm is proposed to strengthen the cryptographic systems by using reversible logic to generate higher and variable hash values, making it difficult to trace the keys. The proposed scheme is simulated and verified using FPGA Virtex ML505 board, the analysis of power and time of which is carried out using Genus tool, proving it to be efficient in terms of power, gate usage, garbage, and quantum cost.