Classification

The present study deals with imperative mood in Persian language based on prototype theory which is one of sub-theories of the cognitive linguistics approach. The research hypotheses are based on distinctions between the form and meaning components in the imperative mood and its distribution in the language corpus of contemporary literature. The aim of this paper is to describe and classify the criterion of the imperative mood in different language levels and beyond. According to the cognitive linguistics approach, language is part of a cognitive system and the syntactic structure of a language can't be studied separately, but all influential factors should be considered from different lingual aspects. The data analysis was carried out through descriptive and qualitative methods. It was found that the base of the verb has a high frequency in the corpus, but the form and meaning are different, with the meaning of imperative mood depending on the context. As a result, all language layers, forms and meanings in description and classification of the imperative mood should be taken into account to be analyzed cognitively.

Language learners have frequently been classified according to individual difference variables such as aptitude, personality, cognitive style, and motivation. However, a language teacher’s view seems to have been missing from such classifications. This exploratory research investigated whether and by which criteria Iranian EFL teachers classify their students. Based on preliminary interviews with 29 high-expertise Iranian EFL teachers, 21 criteria were identified and included in a questionnaire that was completed by 175 Iranian EFL teachers. The respondents almost unanimously agreed that they did classify their students according to their understanding of the character type, behavior patterns, and achievement patterns of their students. Then they rated the 21 criteria on a scale from 0 to 4 according to how important each classification criterion was for them. Factor analysis of questionnaire responses revealed six major classification criteria. Subsequently, in a case study, 26 EFL students in a typical Iranian high school class were asked to rate their classmates according to the six major criteria. Only five of the criteria were found to predict English achievement and Grade Point Average (GPA). A cluster analysis of the students’ peer ratings using the five criteria generated three clusters. An ANOVA revealed that the three clusters were accurately differentiated not only on the clustering criteria but also on the two non-clustering variables: EFL Achievement and GPA.

Due to enchantment in network technology, the worldwide numbers of internet users are growing rapidly. Most of the internet users are using online purchasing from various sites. Due to new online shopping trends over the internet, the seller needs to predict the online customer’s choice. This field is a new area of research for machine learning researchers. A random forest (RF) machine learning method is a widely used classification method. It is mainly based on an ensemble of a single decision tree. Online e-commerce websites accumulate a massive quantity of data in large dimensions. A Random Forest is an efficient filter in high-dimensional data to reliably classify consumer behaviour factors. This research article mainly proposed an extension of the Random Forest classifier named “Weighted Random Forests” (wRF), which incorporates tree-level weights to provide much more accurate trees throughout the calculation as well as an assessment of vector relevance. The weighted random forest algorithm incorporates the C4.5 method named a “Hybrid Weighted Random Forest” (HWRF) to forecast online consumer purchasing behaviour. The experimental results influence the quality of the proposed method in the prediction of the behaviour of online buying customers over existing methods.

One of the most important risks that the banks and financial institutes face, is credit risk which is related to not-paid instalments or the instalments paid with delay by borrowers. Banks use credit scoring models In order to prevent this type of risk. The goal of this research is to investigate the joint effect of economic time cycles and the industry sector on credit scoring models we are seeking to answer the key question: “when should bank change their credit scoring models based on economic time cycles and for which industry sectors?”. The dataset of the research involves all companies that were applied for a loan in one of the Iranian major banks during the years 2008-2011. The companies have been divided into four industry sectors including “Industry and Mine”, “Oil and chemical”, “Service and Infrastructure” and finally “Agriculture”. Based on the sector of the industry and year, 54 explanatory variables, both financial and non-financial, 12 distinct industry sectors and time-specific data sets are built then classification methods were used to classify customers into two groups of defaults and non-defaults. Finally, we compared the results by Wilcoxon Test. The results show that the companies that are in the groups of Industry and Mine and Agriculture, need their own special credit scoring based on industry type model and time but two other groups don’t need of course in the studies dataset duration. Finally, the study concluded by introducing the credit scoring strategies for different four-cycle of economies

SCImago Journal Rank (SJR) is one indicator of a journal's reputation.  The value is calculated based on several published journals, such as scholarly journals' scientific impact, representing the number of quotes sent to a journal and the relevance or reputation of journals from which the quotations originate. A high SJR value means that the corresponding journal has a high reputation. This study aims to approach the SJR classification by implementing a machine learning approach. A simple yet powerful method Naïve Bayes Classifier (NBC), is selected. NBC utilizes probability calculations based on Bayes' theorem. However, NBC has an assumption that the attribute values do not depend on each other. This method is optimized using Particle Swarm Optimization (PSO) to overcome this weakness. This study used SJR data of the computer science domain from 2014 to 2017. Publication without Q rank is filtered for better performance. As a result, the accuracy of the proposed method is higher than the baseline. The use of PSO  significantly improves the NBC performance based on the performed T-test. The PSO-NBC selects four of eight features: H index, Cites/ Doc (2 Years), and Ref. / Doc. Overall results show that using PSO-NBC is closer to SJR rather than using mere NBC.

“AppTree” is an intelligent platform to bring researchers, visitors, and all interested people closer to the oldest and most attractive botanical garden at the University of Tehran. AppTree can scan the QR-Barcode of each plant in person by smartphone or search various plants on the website and get all the useful knowledge about them. Also, the ability of AppTree is the recognition of different plants which don’t have labels. The plant recognition part is a machine learning module that can identify more than 100 different species of plants and give the user details about them. This novel platform is based on Android and Web-app and the identification of new plants type is done by machine learning approach. We utilized VGG19, a deep CNN, to classify images and to identify unlabeled plants. The classification accuracy, F1-score, recall, and precision were 98.25, 93.16%, 88.21%, and 94.85%, respectively, on the plant dataset of the University of Tehran. The proposed method was compared with other deep learning architectures such as AlexNet, AlexNetOWTBn, and GoogLeNet on the same dataset and obtained higher performance. Our AppTree platform has achieved considerable success and easily can be extended to use in other botanical gardens.

Cloud Computing has drastically simplified the management of IT resources by introducing the concept of resource pooling. It has led to a tremendous improvement in infrastructure planning. The major goals of cloud computing include maximization of computing resources with minimization of cost. But the truth is that everything has a price and cloud computing is no different. With Cloud computing there comes a number of security concerns which need to be addressed. Cloud forensics plays a vital role to address the security issues related to cloud computing by identifying, collecting and studying digital evidence in cloud environment. The aim of the research paper is to explore the concept of cloud forensic by applying optimization for feature selection before classification of data on cloud side. The data is classified as malicious and non-malicious using convolutional neural network. The proposed system makes a comparison of models with and without feature selection algorithms before applying the data to CNN. A comparison of different metaheuristics algorithms- Particle Swarm Optimization, Shuffled Frog Leap Optimization and Fire fly algorithm for feature optimization is done based on convergence rate and efficiency.

The widespread deployment of social media has helped researchers access an enormous amount of data in various domains, including the the COVID-19 pandemic. This study draws on a heuristic approach to classify Commercial Instagram Posts (CIPs) and explores how the businesses around the Holy Shrine were impacted by the pandemic. Two datasets of Instagram posts (one gathered data from March 14th to April 10th, 2020, when Holy Shrine and nearby shops were closed, and one extracted data from the same period in 2019), two word embedding models – aimed at vectorizing associated caption of each post, and two neural networks – multi-layer perceptron and convolutional neural network – were employed to classify CIPs in 2019. Among the scenarios defined for the 2019 CIPs classification, the results revealed that the combination of MLP and CBoW achieved the best performance, which was then used for the 2020 CIPs classification. It was found out that the fraction of CIPs to total Instagram posts has increased from 5.58% in 2019 to 8.08% in 2020, meaning that business owners were using Instagram to increase their sales and continue their commercial activities to compensate for the closure of their stores during the pandemic. Moreover, the portion of non-commercial Instagram posts (NCIPs) in total posts has decreased from 94.42% in 2019 to 91.92% in 2020, implying the fact that since the Holy Shrine was closed, Mashhad residents and tourists could not visit it and take photos to post on their Instagram accounts.

Diabetes is a type of metabolic disorder with a high level of blood glucose. Due to the high blood sugar, the risk of heart-related diseases like heart attack and stroke got increased. The number of diabetic patients worldwide has increased significantly, and it is considered to be a major life-threatening disease worldwide. The diabetic disease cannot be cured but it can be controlled and managed by timely detection. Artificial Intelligence (AI) with Machine Learning (ML) empowers automatic early diabetes detection which is found to be much better than a manual method of diagnosis. At present, there are many research papers available on diabetes detection using ML techniques. This article aims to outline most of the literature related to ML techniques applied for diabetes prediction and summarize the related challenges. It also talks about the conclusions of the existing model and the benefits of the AI model. After a thorough screening method, 74 articles from the Scopus and Web of Science databases are selected for this study. This review article presents a clear outlook of diabetes detection which helps the researchers work in the area of automated diabetes prediction.

Due to obstruction in photosynthesis, the leaves of the plants get affected by the disease. Powdery mildew is the main disease in cucumber plants which generally occurs in the middle and late stages. Cucumber plant leaves are affected by various diseases, such as powdery mildew, downy mildew and Alternaria leaf spot, which ultimately affect the photosynthesis process; that’s why it is necessary to detect diseases at the right time to prevent the loss of plants. This paper aims to identify and classify diseases of cucumber leaves at the right time using a deep convolutional neural network (DCNN). In this work, the Deep-CNN model based on disease classification is used to enhance the performance of the ResNet50 model. The proposed model generates the most accurate results for cucumber disease detection using data enhancement based on a different data set. The data augmentation method plays an important role in enhancing the characteristics of cucumber leaves. Due to the requirements of the large number of parameters and the expensive computations required to modify standard CNNs, the pytorch library was used in this work which provides a wide range of deep learning algorithms. To assess the model accuracy large quantity of four types of healthy and diseased leaves and specific parameters such as batch size and epochs were compared with various machine learning algorithms such as support vector machine method, self-organizing map, convolutional neural network and proposed method in which the proposed DCNN model gave better results.

Predicting the development of cancer has always been a serious challenge for scientists and medical professionals. The prompt identification and prognosis of a disease is greatly aided by early-stage detection. Researchers have proposed a number of different strategies for early cancer detection. The purpose of this research is to use meta-learning techniques and several different kinds of convolutional-neural-networks(CNN) to create a model that can accurately and quickly categorize breast cancer(BC). There are many different kinds of breast lesions represented in the Breast Ultrasound Images (BUSI) dataset. It is essential for the early diagnosis and treatment of BC to determine if these tumors are benign or malignant. Several cutting-edge methods were included in this study to create the proposed model. These methods included meta-learning ensemble methodology, transfer-learning, and data-augmentation. With the help of meta-learning, the model will be able to swiftly learn from novel data sets. The feature extraction capability of the model can be improved with the help of pre-trained models through a process called transfer learning. In order to have a larger and more varied dataset, we will use data augmentation techniques to produce new training images. The classification accuracy of the model can be enhanced by using meta-ensemble learning techniques to aggregate the results of several CNNs. Ensemble-learning(EL) will be utilized to aggregate the results of various CNN, and a meta-learning strategy will be applied to optimize the learning process. The evaluation results further demonstrate the model's efficacy and precision. Finally, the suggested model's accuracy, precision, recall, and F1-score will be contrasted to those of conventional methods and other current systems.

Early cancer detection is crucial for improving patient survival rates, as timely intervention greatly enhances treatment efficacy. One promising method for early detection is identifying cancerous cells through the detection of protein-level modifications, which serve as early indicators of malignancy. These protein modifications often result from complex biochemical processes that occurs before visible cellular abnormalities, making them critical targets for diagnostic technologies. In recent years, wireless biomedical sensors have advanced significantly, enabling precisely detecting these protein-level changes. These sensors have the potential to detect cancer at its earliest stages by monitoring the subtle alterations in protein structures and functions that distinguish healthy cells from cancerous ones. As the costs of genetic analysis continue to decrease, the development of Medical Internet of Things (MIoT) devices has become increasingly feasible. These devices are designed to perform real-time analyses of biological specimens—such as blood and urine—by detecting protein-level changes indicative of cancer. In this paper, a new machine learning method based on Extreme Randomized Trees (ERT) is developed to increase the speed of classification of cancerous cells based on single-cell Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq). The proposed method enhances the classification speed of the limited and noisy ATAC-seq data as it requires less computation to determine the best splits at each node of the decision trees. This method can significantly improve near real-time cancer risk assessment using samples collected by MIoT. Our proposed method achieves classification accuracy comparable to state of the art single-cell ATAC-seq data analysis techniques while reducing processing time by 259%, challenged by various low-data scenarios. This approach presents an efficient solution for rapid cancer monitoring within the MIoT framework.