Search

Due to the multimodal nature of real-world optimization problems, in recent years, there has been a great interest in multi-modal optimization algorithms. Multimodal optimization problems involve identifying multiple local/global optima. Niching techniques have been widely used to tackle multi-modal optimization problems. Most of the existing niching methods either require predefined niching parameters or extra information about the problem space. This paper presents a novel multimodal algorithm based on Butterfly Optimization Algorithm, which is constructed using the Fitness-Distance Balance (FDB) selection method. The purpose of applying the FDB selection method is to discover local/global optima with high potential as a fitness value along with the appropriate distance from solution candidates. Also, a local search scheme is used to enhance the convergence speed of the algorithm. Niching is a technique used in multimodal optimization to maintain diversity among multiple solutions in the population. The minimum distance between the solutions is called a "niche". The proper niching radius is the main challenge for existing approaches. Knowing the problem space helps determine the niche radius. This paper proposes a new multimodal optimization scheme that does not require prior knowledge of the problem space or the niching parameter. Seven state-of-the-art multi-modal optimization algorithms are compared to the multi-modal butterfly optimization algorithm (MBOA) on 16 benchmarks from the CEC 2013 and CEC 2015 competitions to evaluate its performance. Success rate, Number of function evaluations, Success performance, average number of optima found, Success accuracy, Maximum peak ratio, and Run-time performance criteria were measured over 25 runs to assess the efficiency of the proposed method. The experimental results demonstrate that MBOA outperforms other algorithms according to most of the performance criteria. [ABSTRACT FROM AUTHOR]

Many researchers have used tag information to improve the performance of recommendation techniques in recommender systems. Examining the tags of users will help to get their interests and leads to more accuracy in the recommendations. Since user-defined tags are chosen freely and without any restrictions, problems arise in determining their exact meaning and the similarity of tags. However, using thesaurus and ontologies to find the meaning of tags is not very efficient due to their free definition by users and the use of different languages in many data sets. Therefore, this article uses mathematical and statistical methods to determine lexical similarity and co-occurrence tags solution to assign semantic similarity. On the other hand, due to the change of users' interests over time this article has considered the time of tag assignments in co-occurrence tags for determining the similarity of tags. Then the graph is created based on similarity of tags. For modeling the interests of the users, the communities of tags are determined by using community detection methods. So, recommendations based on the communities of tags and similarity between resources are done. The performance of the proposed method has been evaluated using two criteria of precision and recall through evaluations on two public datasets. The evaluation results show that the precision and recall of the proposed method have significantly improved, compared to the other methods. According to the experimental results, the criteria of recall and precision have been improved, on average by 5% and 7%, respectively. [ABSTRACT FROM AUTHOR]

Nowadays, a tremendous amount of human communications occur on Internet-based communication infrastructures, like social networks, email, forums, organizational communication platforms, etc. Indeed, the automatic prediction or assessment of individuals' personalities through their written or exchanged text would be advantageous to ameliorate their relationships. To this end, this paper aims to propose KGrAt-Net, which is a Knowledge Graph Attention Network text classifier. For the first time, it applies the knowledge graph attention network to perform Automatic Personality Prediction (APP), according to the Big Five personality traits. After performing some preprocessing activities, it first tries to acquire a knowing-full representation of the knowledge behind the concepts in the input text by building its equivalent knowledge graph. A knowledge graph collects interlinked descriptions of concepts, entities, and relationships in a machine-readable form. Practically, it provides a machine-readable cognitive understanding of concepts and semantic relationships among them. Then, applying the attention mechanism, it attempts to pay attention to the most relevant parts of the graph to predict the personality traits of the input text. We used 2467 essays from the Essays Dataset. The results demonstrated that KGrAt-Net considerably improved personality prediction accuracies (up to 70.26% on average). Furthermore, KGrAt-Net also uses knowledge graph embedding to enrich the classification, which makes it even more accurate (on average, 72.41%) in APP. [ABSTRACT FROM AUTHOR]

Named entity recognition (NER) is a subfield of natural language processing (NLP). It is able to identify proper nouns, such as person names, locations, and organizations, and has been widely used in various tasks. NER can be practical in extracting information from social media data. However, the unstructured and noisy nature of social media (such as grammatical errors and typos) causes new challenges for NER, especially for low-resource languages such as Persian, and existing NER methods mainly focus on formal texts and English social media. To overcome this challenge, we consider Persian NER as an optimization problem and use the binary Gray Wolf Optimization (GWO) algorithm to segment posts into small possible phrases of named entities. Later, named entities are recognized based on their score. Also, we prove that even human opinion can differ in the NER task and compare our method with other systems with the S e p _ T D _ T e l 01 dataset and the results show that our proposed system obtains a higher F1 score in comparison with other methods. [ABSTRACT FROM AUTHOR]

Human personality is significantly represented by those words which he/she uses in his/her speech or writing. As a consequence of spreading the information infrastructures (specifically the Internet and social media), human communications have reformed notably from face to face communication. Generally, Automatic Personality Prediction (or Perception) (APP) is the automated forecasting of the personality on different types of human generated/exchanged contents (like text, speech, image, video, etc.). The major objective of this study is to enhance the accuracy of APP from the text. To this end, we suggest five new APP methods including term frequency vector-based, ontology-based, enriched ontology-based, latent semantic analysis (LSA)-based, and deep learning-based (BiLSTM) methods. These methods as the base ones, contribute to each other to enhance the APP accuracy through ensemble modeling (stacking) based on a hierarchical attention network (HAN) as the meta-model. The results show that ensemble modeling enhances the accuracy of APP. [ABSTRACT FROM AUTHOR]

How people think, feel, and behave primarily is a representation of their personality characteristics. By being conscious of the personality characteristics of individuals whom we are dealing with or deciding to deal with, one can competently ameliorate the relationship, regardless of its type. With the rise of Internet-based communication infrastructures (social networks, forums, etc.), a considerable amount of human communications takes place there. The most prominent tool in such communications is the language in written and spoken form that adroitly encodes all those essential personality characteristics of individuals. Text-based Automatic Personality Prediction (APP) is the automated forecasting of the personality of individuals based on the generated/exchanged text contents. This paper presents a novel knowledge graph-enabled approach to text-based APP that relies on the Big Five personality traits. To this end, given a text, a knowledge graph, which is a set of interlinked descriptions of concepts, was built by matching the input text's concepts with DBpedia knowledge base entries. Then, due to achieving a more powerful representation, the graph was enriched with the DBpedia ontology, NRC Emotion Intensity Lexicon, and MRC psycholinguistic database information. Afterwards, the knowledge graph, which is now a knowledgeable alternative for the input text, was embedded to yield an embedding matrix. Finally, to perform personality predictions, the resulting embedding matrix was fed to four suggested deep learning models independently, which are based on convolutional neural network (CNN), simple recurrent neural network (RNN), long short-term memory (LSTM), and bidirectional long short-term memory (BiLSTM). The results indicated considerable improvements in prediction accuracies in all of the suggested classifiers. [ABSTRACT FROM AUTHOR]

Online social media such as Twitter are growing so rapidly. Recently, Twitter has become one of the popular microblogging services on the Internet. It lets millions of users to communicate and interact by sending short messages of up to 140 characters. The massive amount of information over the web from Twitter requires an automatic tool that can determine the topics that people are talking about. The Topic Detection task is concentrated on discovering the main topics automatically. In this article at first, we explore different approaches to detect topics of tweets. Then, we will classify these topic detection approaches to four classes of categories, including with word embedding or without word embedding, specified or unspecified, offline (RED) or online (NED), and supervised or unsupervised. Finally, we will discuss the studied approaches in detail. [ABSTRACT FROM AUTHOR]

Introduction: In recent decades, the growing rate of cancer incidence is a big concern for most societies. Due to the genetic origins of cancer disease, its internal structure is necessary for the study of this disease. Methods: In this research, cancer data are analyzed based on DNA sequences. The transition probability of occurring two pairs of nucleotides in DNA sequences has Markovian property. This property inspires the idea of feature dimension reduction of DNA sequence for overcoming the high computational overhead of genes analysis. This idea is utilized in this research based on the Markovian property of DNA sequences. This mapping decreases feature dimensions and conserves basic properties for discrimination of cancerous and noncancerous genes. Results: The results showed that a non-linear support vector machine (SVM) classifier with RBF and polynomial kernel functions can discriminate selected cancerous samples from non-cancerous ones. Experimental results based on the 10-fold cross-validation and accuracy metrics verified that the proposed method has low computational overhead and high accuracy. Conclusion: The proposed algorithm was successfully tested on related research case studies. In general, a combination of proposed Markovian-based feature reduction and non-linear SVM classifier can be considered as one of the best methods for discrimination of cancerous and noncancerous genes. [ABSTRACT FROM AUTHOR]

Cancer is one of the life-threatening diseases caused by changes in the structure of genetic components of the cell. DNA sequences are one of the most important factors in the formation and spread of this disease. The signal processing approach is one of the scientific fields that has been developed in the last two decades in the analysis of DNA sequences. In this research, a hybrid model of discrete Fourier transform and anti-notch digital filter has been used for this purpose. The aim of using these techniques is to model an approach that can distinguish cancerous samples from non-cancerous ones. In other words, a pattern recognition model is designed to discriminate cancerous cell samples based on the features of protein coding regions of DNA sequences. Some computational and statistical techniques have been used in feature extraction and feature selection stages. Despite the proposed model simplicity, it doesn't face conventional challenges such as high computational complexity or memory dissipation. Case studies have been tested with the least possible feature, depending on the nature of the features. Experimental results and features relationship led to the proposal of the SVM classifier to discriminate two categories. The output features and classification show good discrimination results among the cancerous and non-cancerous samples. One of the main advantages of the proposed model is the independence of its performance over the data length. Evaluation and validation results indicate the high accuracy and precision of the proposed method which emphasizes the biological genetic mutation nature of cancer. [ABSTRACT FROM AUTHOR]

Multi-modal optimization algorithms are one of the most challenging issues in the field of optimization. Most real-world problems have more than one solution; therefore, the potential role of multi-modal optimization algorithms is rather significant. Multi-modal problems consider several global and local optima. Therefore, during the search process, most of the points should be detected by the algorithm. The forest optimization algorithm has been recently introduced as a new evolutionary algorithm with the capability of solving unimodal problems. This paper presents the multi-modal forest optimization algorithm (MMFOA), which is constructed by applying a clustering technique, based on niching methods, to the unimodal forest optimization algorithm. The MMFOA operates by dividing the population of the forest into subpopulations to locate existing local and global optima. Subpopulations are generated by the Basic Sequential Algorithmic Scheme with a radius neighborhood. As population size is self-adaptive in MMFOA, population size can be increased in functions with too many local and global optima. The proposed algorithm is evaluated by a set of multi-modal benchmark functions. The experiment results show that not only is the population size low, but also that the convergence speed is high, and that the algorithm is efficient in solving multi-modal problems. [ABSTRACT FROM AUTHOR]

Nowadays, it is necessary that users have access to information in a concise form without losing any critical information. Document summarization is an automatic process of generating a short form from a document. In itemset‐based document summarization, the weights of all terms are considered the same. In this paper, a new approach is proposed for multidocument summarization based on weighted patterns and term association measures. In the present study, the weights of the terms are not equal in the context and are computed based on weighted frequent itemset mining. Indeed, the proposed method enriches frequent itemset mining by weighting the terms in the corpus. In addition, the relationships among the terms in the corpus have been considered using term association measures. Also, the statistical features such as sentence length and sentence position have been modified and matched to generate a summary based on the greedy method. Based on the results of the DUC 2002 and DUC 2004 datasets obtained by the ROUGE toolkit, the proposed approach can outperform the state‐of‐the‐art approaches significantly. [ABSTRACT FROM AUTHOR]

Owing to advances in information technology, online communications between people living in different parts of the world have considerably increased. The subsequent emergence of social networks helped this kind of communications to be further organized. One of the most important issues considered when analyzing these kinds of networks is community detection, in which a majority of studies tend to detect disjoint communities through analyzing linkages of networks. What this paper aims to achieve is to obtain overlapping communities in which the members have the same topics of interest, and where the strengths of connections between them are the consequence of their communications’ content analysis. Consequently, we have hereby proposed a generic framework for overlapping community detection in social networks with special focus on rating-based social networks. This framework considers the information shared by the users (ratings), as well as their topics of interest, for the sake of finding meaningful communities. This will lead us to topical communities in which members are interested in the same topics, and the strengths of their relationships are directly based on the rate of their viewpoints’ unity. Quantitative evaluations also reveal that the framework presented in this study achieves favorable results which are quite superior to the results of 3 other relevant frameworks in the literature. [ABSTRACT FROM AUTHOR]

Detecting communities in complex networks is one of the most important issues considered when analyzing these kinds of networks. A majority of studies in the field of community detection tend to detect communities through analyzing linkages of the networks. What this paper aims to achieve is to reach to a trade-off between similarity of nodes' attributes and density of connections in finding communities of social networks with node attributes. Since the community detection problem can be modeled as a seriously non-linear discrete optimization problem, we have hereby proposed a multi-objective discrete Biogeography Based Optimization (BBO) algorithm to find communities in social networks with node attributes. This algorithm uses the Pareto-based approach for community detection. Also, we introduced a new metric, SimAtt, to measure the similarity of node attributes in a community of a network and used it along with Modularity, which considers the linkage structure of a network to detect communities, as the two objective functions of the proposed method to be maximized. In the proposed method, a two phase sorting strategy is introduced which uses the non-dominated sorting and Crowding-distance to sort the generated solution of a population in each iteration. Moreover, this paper introduces a method for mutation probability approximation and uses a chaotic mechanism to dynamically tune the mutation probability in each iteration. Additionally, two novel strategies are introduced for mutation in unweighted and weighted networks. Since the final output of the proposed method is a set of non-dominated (Pareto-optimal) solutions, a metric named alpha_SAM is proposed to determine the best compromise solution among these non-dominated ones. Quantitative evaluations based on extensive experiments on 14 real-life data sets reveals that the method presented in this study achieves favorable results which are quite superior to other relevant algorithms in the literature. [ABSTRACT FROM AUTHOR]

Reinforcement Learning (RL) is used to find the best policy. A policy is a rule that maps a given state to an appropriate action. The RL is used to learn utility functions for dynamic resource allocation. According to the future demand of resources, the learned policy maps the appropriate resources in a way that wasting energy and resources is stopped and Service Level Agreements (SLA) violation and Quality of Service (QoS) dropping are also avoided. However, The RL encounters a lot of problems in this field such as: having good policies in the early phases of learning and the learning time to converge to the optimal policy. This paper deals with these problems using the appropriate initialization of Q learning and a new fuzzy approach to increase the convergence speed to the optimal policy. The fuzzy approach presented in this paper improves the accuracy and speed of convergence of Q learning. Firstly, the proposed method predicts the future workload, then determines the appropriate number of physical machines by using the optimal policy learned by improved Q learning. The evaluation results show the advantages of accuracy and convergence speed of the proposed method in comparison with the similar methods. [ABSTRACT FROM AUTHOR]

Feature selection as a combinatorial optimization problem is an important preprocessing step in data mining; which improves the performance of the learning algorithms with the help of removing the irrelevant and redundant features. As evolutionary algorithms are reported to be suitable for optimization tasks, so Forest Optimization Algorithm (FOA) – which is initially proposed for continuous search problems – is adapted to be used for feature selection as a discrete search space problem. As the result, Feature Selection using Forest Optimization Algorithm (FSFOA) is proposed in this article in order to select the more informative features from the datasets. The proposed FSFOA is validated on several real world datasets and it is compared with some other methods including HGAFS, PSO and SVM-FuzCoc. The results of the experiments show that, FSFOA can improve the classification accuracy of classifiers in some selected datasets. Also, we have compared the dimensionality reduction of the proposed FSFOA with other available methods. [ABSTRACT FROM AUTHOR]

In this article, a new evolutionary algorithm, Forest Optimization Algorithm (FOA), suitable for continuous nonlinear optimization problems has been proposed. It is inspired by few trees in the forests which can survive for several decades, while other trees could live for a limited period. In FOA, seeding procedure of the trees is simulated so that, some seeds fall just under the trees, while others are distributed in wide areas by natural procedures and the animals that feed on the seeds or fruits. Application of the proposed algorithm on some benchmark functions demonstrated its good capability in comparison with Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). Also we tested the performance of FOA on feature weighting as a real optimization problem and the results of the experiments showed the good performance of FOA in some data sets from the UCI repository. [ABSTRACT FROM AUTHOR]

□ This paper is a review of automated text summarization, which is concerned with a subfield of natural language processing (NLP). With the daily increase of available information on the Internet, the need for systems that are able to automatically summarize text is greater than before. The continual study of a significant number of documents is a laborious and time-consuming process, so existence of an alternative procedure will be very effective. In this article, our consideration of the effective aspects of the summarization process will be discussed in detail. Three classifications of approaches to text summarization will be presented. Also, we will review some text summarization systems and their characteristics, and finally, the manner of evaluating text summarization systems will be surveyed. [ABSTRACT FROM PUBLISHER]

Cross-lingual word embeddings display words from different languages in the same vector space. They provide reasoning about semantics, compare the meaning of words across languages and word meaning in multilingual contexts, necessary to bilingual lexicon induction, machine translation, and cross-lingual information retrieval. This paper proposes an efficient approach to learn bilingual transform mapping between monolingual word embeddings in language pairs. We choose ten different languages from three different language families and downloaded their last update Wikipedia dumps1 Then, with some pre-processing steps and using word2vec, we produce word embeddings for them. We select seven language pairs from chosen languages. Since the selected languages are relative, they have thousands of identical words with similar meanings. With these identical dictation words and word embedding models of each language, we create training, validation and, test sets for the language pairs. We then use a generative adversarial network (GAN) to learn the transform mapping between word embeddings of source and target languages. The average accuracy of our proposed method in all language pairs is 71.34%. The highest accuracy is achieved for the Turkish-Azerbaijani language pair with the accuracy 78.32%., which is noticeably higher than prior methods. [ABSTRACT FROM AUTHOR]

Abstract This article is improved the random forest algorithm by selecting the most appropriate penalized regression methods, and it is tried to improve the post-selection boosting random forest (PBRF) algorithm using elastic net regression. The proposed method with the highest efficiency is called Reducing and Aggregating Random Forest Trees by Elastic Net (RARTEN). The introduced method consists of three steps. In the first step, the random forest algorithm is used as a predictor. In the second step, Elastic Net, as a penalized regression method, is applied to reduce the number of trees and improve the random forest and PBRF. In the last step, selected trees are aggregated. The obtained results of the real data and Monte Carlo simulation are evaluated using various statistical performance criteria. The simulation study shows that the RARTEN with 7%, 5%, and 8.5% reduction in the linear, nonlinear, and noise model, respectively improve the accuracy of the traditional random forest and the proposed method by Wang. In addition, this method has a significant reduction compared to other penalized regression methods. Moreover, the real data results show that the proposed method in our study with a reduction of almost 16% confirms the validity of the proposed model. [ABSTRACT FROM AUTHOR]

Novel metaheuristic algorithms are now considered an appealing collection of methods for solving complex optimization problems, in which the challenging objective is to find a better solution in a shorter computation time. Focusing on the same objective, this paper proposes a novel metaheuristic optimization algorithm inspired by heat transfer relationships based on the second law of thermodynamics. Imitating the heat transfer behavior of solid objects, the proposed method is called the heat transfer relation-based optimization algorithm (HTOA). This behavior was modeled on a heat transfer function used to measure temperature differences between the selected solutions and the best solution. This function was employed to determine and add the heat capacity transferred between those solutions. Finally, all the solutions were heat-exchanged with the best solution to select the fittest solution and exclude the rest. This procedure continued until the best solution or solutions were found. The proposed method is challenged by many famous benchmark problems in two categories as well as two real-world problems (PID controller and linear regression). The HTOA was then compared with a number of well-known and state-of-the-art optimization algorithms. Selecting better solutions and requiring shorter computation time, the proposed HTOA outperformed the other algorithms. [ABSTRACT FROM AUTHOR]

Abstract Present paper introduces a polynomial combination of one dimensional chaotic maps that is blended in a dynamic image encryption algorithm. It is special because not only this combination has butterfly folding effect but also it shows generalization property over any polynomial combination. Hence, the butterfly folding effect is caused by governed parameters of polynomial combination. Moreover, multiple simulations and evaluations show the superiority of the proposed chaotic system. An application of this system, which we propose in cryptography, is a novel image encryption algorithm based on dynamic function generation. Compared to the state of the art algorithms, our image encryption algorithm has higher statistical and cryptanalytic properties. Even though this algorithm is not suitable for real-time applications such as streaming video encryption, it makes a good use of the proposed chaotic system. Uppermost cryptanalytic properties that are proven by statistical/numeric tests show good performance and reliability of proposed algorithm for image encryption tasks while unlike any other chaotic image encryption system, our algorithm uses a string input for secret key. [ABSTRACT FROM AUTHOR]

In recent years, hybrid genetic algorithms (GAs) have received significant interest and are widely being used to solve real-world problems. The hybridization of heuristic methods aims at incorporating benefits of stand-alone heuristics in order to achieve better results for the optimization problem. In this paper, we propose a hybridization of GAs and Multiagent Reinforcement Learning (MARL) heuristic for solving Traveling Salesman Problem (TSP). The hybridization process is implemented by producing the initial population of GA, using MARL heuristic. In this way, GA with a novel crossover operator, which we have called Smart Multi-point crossover, acts as tour improvement heuristic and MARL acts as construction heuristic. Numerical results based on several TSP datasets taken from the TSPLIB demonstrate that proposed method found optimum solution of many TSP datasets and near optimum of the others and enable to compete with nine state-of-the-art algorithms, in terms of solution quality and CPU time. [ABSTRACT FROM AUTHOR]

Objective: To design an ontology-based clinical decision support system based on a clinical guideline for urinary incontinence. Materials and Methods: The study will be conducted in four phases: Updating the adapted clinical guideline for urinary incontinence for Iranian clinicians; Developing an ontology based on the adapted guideline; Developing a guideline-oriented ontological decision support system for urinary incontinence; and evaluating both the ontology and the decision support system. The GRADE-Adolopment methodology will be used for updating the adapted guideline. The researcher will deploy Protégé 5.5.0 ontology editor for developing the ontology. The rules will be extracted from the adapted guideline for urinary incontinence, and the rule language will be Semantic Web Rule Language. Ontology consistency will be evaluated with Pellet reasoner. The system will be evaluated and analyzed by the GUIDES checklist. Results: The results of the study will be published and disseminated in peer-reviewed journals. Conclusion: UrInO-DSS will offer a tool to support clinicians in providing personalized treatment for patients who suffer from urinary incontinence. It can also help the residents and medical students to learn how to diagnose and manage urinary incontinence in the best way. The system can be implemented as an international decision support system for the diagnosis and management of urinary incontinence. [ABSTRACT FROM AUTHOR]

Background: Clinical practice guidelines are statements which are based on the best available evidence, and their goal is to improve the quality of patient care. Integrating clinical practice guidelines into computer systems can help physicians reduce medical errors and help them to have the best possible practice. Guideline-based clinical decision support systems play a significant role in supporting physicians in their decisions. Meantime, system errors are the most critical concerns in designing decision support systems that can affect their performance and efficacy. A well-developed ontology can be helpful in this matter. The proposed systematic review will specify the methods, components, language of rules, and evaluation methods of current ontology-driven guideline-based clinical decision support systems. Methods: This review will identify literature through searching MEDLINE (via Ovid), PubMed, EMBASE, Cochrane Library, CINAHL, ScienceDirect, IEEEXplore, and ACM Digital Library. Gray literature, reference lists, and citing articles of the included studies will be searched. The quality of the included studies will be assessed by the mixed methods appraisal tool (MMAT-version 2018). At least two independent reviewers will perform the screening, quality assessment, and data extraction. A third reviewer will resolve any disagreements. Proper data analysis will be performed based on the type of system and ontology engineering evaluation data. Discussion: The study will provide evidence regarding applying ontologies in guideline-based clinical decision support systems. The findings of this systematic review will be a guide for decision support system designers and developers, technologists, system providers, policymakers, and stakeholders. Ontology builders can use the information in this review to build well-structured ontologies for personalized medicine. Systematic review registration: PROSPERO CRD42018106501 [ABSTRACT FROM AUTHOR]

Semi-supervised learning, which entails training a model with manually labeled images and pseudo-labels for unlabeled images, has garnered considerable attention for its potential to improve image classification performance. Nevertheless, incorrect decision boundaries of classifiers and wrong pseudo-labels for beneficial unlabeled images below the confidence threshold increase the generalization error in semi-supervised learning. This study proposes a novel framework for semi-supervised learning termed consistency-regularized bad generative adversarial network (CRBSGAN) through a new loss function. The proposed model comprises a discriminator, a bad generator, and a classifier that employs data augmentation and consistency regularization. Local augmentation is created to compensate for data scarcity and boost bad generators. Moreover, label consistency regularization is considered for bad fake images, real labeled images, unlabeled images, and latent space for the discriminator and bad generator. In the adversarial game between the discriminator and the bad generator, feature space is better captured under these conditions. Furthermore, local consistency regularization for good-augmented images applied to the classifier strengthens the bad generator in the generator–classifier adversarial game. The consistency-regularized bad generator produces informative fake images similar to the support vectors located near the correct classification boundary. In addition, the pseudo-label error is reduced for low-confidence unlabeled images used in training. The proposed method reduces the state-of-the-art error rate from 6.44 to 4.02 on CIFAR-10, 2.06 to 1.56 on MNIST, and 6.07 to 3.26 on SVHN using 4000, 3000, and 500 labeled training images, respectively. Furthermore, it achieves a reduction in the error rate on the CINIC-10 dataset from 19.38 to 15.32 and on the STL-10 dataset from 27 to 16.34 when utilizing 1000 and 500 labeled images per class, respectively. Experimental results and visual synthesis indicate that the CRBSGAN algorithm is more efficient than the methods proposed in previous works. The source code is available at https://github.com/ms-iraji/CRBSGAN ↗. [ABSTRACT FROM AUTHOR]

Background: Road traffic accidents have been one of the leading causes of death. Despite the increasing trend of road traffic apps, there is no comprehensive analysis of their features and no taxonomy for the apps based on traffic safety theories. This study aimed to explore the characteristics of available mobile apps on road traffic health/safety and classify them with emphasis on Haddon's matrix.Methods: The researchers examined the mobile applications related to road traffic health/safety using qualitative content analysis. Google Play was searched using a combination of the keywords. Haddon's matrix was applied to analyze and classify those mobile apps residing in the categories of Road Traffic health & Safety, and Road Traffic Training.Results: Overall, 913 mobile apps met the inclusion criteria and were included in the final analysis. Classification of the apps based on their features resulted in 4 categories and 21 subcategories. A total number of 657 mobile apps were classified based on Haddon's matrix. About 45.67% of these apps were categorized as the road traffic health & safety group.Conclusions: Haddon's matrix appears to have the potential to reveal the strengths and weaknesses of existing mobile apps in the road traffic accident domain. Future development of mobile apps in this domain should take into account the existing gap. [ABSTRACT FROM AUTHOR]

In image clustering, it is desired that pixels assigned in the same class must be the same or similar. In other words, the homogeneity of a cluster must be high. In gray scale image segmentation, the specified goal is achieved by increasing the number of thresholds. However, the determination of multiple thresholds is a typical issue. Moreover, the conventional thresholding algorithms could not be used in color image segmentation. In this study, a new color image clustering algorithm with multilevel thresholding has been presented and, it has been shown how to use the multilevel thresholding techniques for color image clustering. Thus, initially, threshold selection techniques such as the Otsu and Kapur methods were employed for each color channel separately. The objective functions of both approaches have been integrated with the forest optimization algorithm (FOA) and particle swarm optimization (PSO) algorithm. In the next stage, thresholds determined by optimization algorithms were used to divide color space into small cubes or prisms. Each sub-cube or prism created in the color space was evaluated as a cluster. As the volume of prisms affects the homogeneity of the clusters created, multiple thresholds were employed to reduce the sizes of the sub-cubes. The performance of the proposed method was tested with different images. It was observed that the results obtained were more efficient than conventional methods. [ABSTRACT FROM AUTHOR]

In this paper, a new algorithm for generating more-randomized keys for symmetrical cipher one-time pad (OTP) according to the linear congruential (LCG) method based on the idea of genetic algorithm is proposed. The method, genetic-based random key generator, is proposed for generating keys for the OTP method with a high degree of key randomness; this adds more strength to the OTP method against breaking this cryptosystem. This algorithm is composed of two parts. Initially, the first population is being generated by LCG method, and then, genetic operators for generating the next populations are being used. Generating random keys with the presented method requires seven-parameter key that increases the security of communication between the transceivers. [ABSTRACT FROM AUTHOR]