Neural Classification Algorithm Research Articles

ORANGE FRUIT IMAGE QUALITY ASSESSMENT USING HYBRID MODEL

Since ripeness is regarded by consumers as the most important quality indication in the crops industry, maintaining and monitoring its ripeness has become a major issue in the business of growing crops. Additionally, as it significantly affects the product's quality and consumer preferences, the product's appearance is one of the makers' most pressing worries. However, the forecasting of storage life and the selection of the best harvest dates are still largely based on individual interpretation and practical experience. Threshold segmentation and certain morphological techniques are used in the early stage of the extraction of an area of interest. The segmented orange images are then divided into training and testing data sets, and the second stage entails extracting color-based traits from them. Choosing the classifier training settings is the focus of the third phase. The final phase, which makes use of the previously trained ANN, categorizes the data. One of the most crucial conclusions of this study is that creating a neural network is an empirical process that requires a lot of trials in order to identify the ideal variables that will give the neural classification algorithm the best performance. initialization of weights, the quantity of hidden neurons. Based on the proposed model the accuracy and efficiency is 1.3% which is more accurate result.

An Adaptive Deep Feature Neural Classification Algorithm for Efficient Cardiac Disease Early Risk Identification

Cloud environments can store data for the medical field with a large-scale system. Importantly, early detection of diseases using clinical data analysis is essential in medicine. Cardiovascular disease has emerged as the primary cause of sudden high-risk health-related fatalities in recent years. Analyzing time series data has become more complex and non-linear, making predictive risk analysis through feature analysis a crucial aspect of data analysis. Feature measures that are not feasible can hurt prediction accuracy and may result in misclassification. To overcome this problem, an improved clinical data analysis model using Adaptive Deep Feature Neural Classification (DFNC) method for cardiac data prediction can identify early risk stages. Time series data can be standardized from the CVD-DS dataset initially selected using a preprocessing model. Correlation with a subset of margins can be obtained using the Cardiac Deficiency Prediction Rate (CDPR). Then, the Support Frequent Scaling Feature Selection (SFSFS) model can extract the feature components based on the CDPR weights. The required features can be obtained using a deep neural classifier based on logistic neurons. The classifier is based on a Recurrent Neural Network (RNN) that deliberates each class category's feature values and Cardia Influence Rate (CIR). Classification, precision and recall can be implemented in the proposed method to provide high prediction accuracy. Additionally, early cardiovascular disease risk prediction accuracy may support early diagnosis management.

An automatic product recommendation system in e-commerce using Flamingo Search Optimizer and Fuzzy Temporal Multi Neural Classifier

<p>In this paper, a new automatic product recommendation system (APRS) is proposed to recommend the suitable products to the customer in e-commerce by analyzing the customers’ reviews. This recommendation system applies semantic aware data preprocessing, feature selection and extraction and classification. The initial level data preprocessing including blank space and stop word removal. Moreover, we use a Flamingo Search Optimizer (FSO) for optimizing the features that are extracted in the initial level data preprocessing. In addition, a new Fuzzy Temporal Multi Neural Classification Algorithm (FTMNCA) is proposed for performing effective classification that is helpful to make effective decision on prediction process. In addition, the proposed automatic product recommendation system recommends the suitable products to the customers according to the classification result. Finally, the proposed system is evaluated by conducting various experiments and proved as superior than the available systems in terms of prediction accuracy, precision, recall and f-measure.</p>

Анализ безопасности проекта национального стандарта «Нейросетевые алгоритмы в защищенном исполнении. Автоматическое обучение нейросетевых моделей на малых выборках в задачах классификации»

We propose a membership inference attack against the neural classification algorithm from the draft national standard developed by the Omsk State Technical University under the auspices of the Technical Committee on Standardization «Artificial Intelligence» (TC 164). The attack allows us to determine whether the data were used for neural network training, and aimed at violating the confidentiality property of the training set. The results show that the protection mechanism of neural network classifiers described by the draft national standard does not provide the declared properties. The results were previously announced at Ruscrypto’2023 conference.

SVM Algorithm for Industrial Defect Detection and Classification

This article presents a new algorithm for recognizing defects and discontinuities. It is a neural classification algorithm of the SVM class used for the vision system in the technological sequence. At the basis of the used method of Support Vector Machines (SVM) lies the concept of decision-making space, which is divided by building boundaries separating objects with different class affiliation, that is, defects and discontinuities. The Support Vector Machines method is supposed to perform classification tasks by constructing in a multidimensional space hyperplane separating cases belonging to different classes.

Interval probabilistic neural network

Automated classification systems have allowed for the rapid development of exploratory data analysis. Such systems increase the independence of human intervention in obtaining the analysis results, especially when inaccurate information is under consideration. The aim of this paper is to present a novel approach, a neural networking, for use in classifying interval information. As presented, neural methodology is a generalization of probabilistic neural network for interval data processing. The simple structure of this neural classification algorithm makes it applicable for research purposes. The procedure is based on the Bayes approach, ensuring minimal potential losses with regard to that which comes about through classification errors. In this article, the topological structure of the network and the learning process are described in detail. Of note, the correctness of the procedure proposed here has been verified by way of numerical tests. These tests include examples of both synthetic data, as well as benchmark instances. The results of numerical verification, carried out for different shapes of data sets, as well as a comparative analysis with other methods of similar conditioning, have validated both the concept presented here and its positive features.

Twentieth century ENSO characteristics in the IPCC database

In this paper, we assess and compare to observations the spatial characteristics of the twentieth Century ENSO SST variability simulated by 23 models of the IPCC-AR4/CMIP3 database. The analysis is confined to the SST anomalies along the equatorial Pacific and is based on the use of a non-linear neural classification algorithm, the Self-Organizing Maps. Systematic biases include a larger than observed proportion for modelled ENSO maximum variability occurring in the Western Pacific. No clear relationship is found between this bias and the characteristics of the modelled mean state bias in the equatorial Pacific. This bias is mainly related to a misrepresentation of both El Nino and La Nina termination phases for most of the models. In contrast, the onset phase is quite well simulated. Modelled El Nino and La Nina peak phases display an asymmetric bias. Whereas the main bias of the modelled El Nino peak is to exhibit a maximum in the western Pacific, the simulated La Nina bias mainly occurs in the central Pacific. In addition, some models are able to capture the observed El Nino peak characteristics while none of them realistically simulate La Nina peaks. It also arises that the models closest to the observations score unevenly in reproducing the different phases, preventing an accurate classification of the models quality to reproduce the overall ENSO-like variability.

An efficient neural classification chain of SAR and optical urban images

In this paper a suitable neural classification algorithm, based on the use of Adaptive Resonance Theory (ART) networks, is applied to the fusion and classification of optical and SAR urban images. ART networks provide a flexible tool for classification, but are ruled by a large number of parameters. Therefore, the simplified ART2-A algorithm is used in this paper, and the neural approach is integrated into a classification chain where fuzzy clustering for merging of classes is also considered. The interaction between the two methods leads to encouraging results in less CPU time than classification with fuzzy clustering alone or other classical approaches (ISODATA). Examples of classification are provided using C-band total power AIRSAR data and optical images of Santa Monica, Los Angeles.

An efficient neural classification chain of SAR and optical urban images

In this paper a suitable neural classification algorithm, based on the use of Adaptive Resonance Theory (ART) networks, is applied to the fusion and classification of optical and SAR urban images. ART networks provide a flexible tool for classification, but are ruled by a large number of parameters. Therefore, the simplified ART2-A algorithm is used in this paper, and the neural approach is integrated into a classification chain where fuzzy clustering for merging of classes is also considered. The interaction between the two methods leads to encouraging results in less CPU time than classification with fuzzy clustering alone or other classical approaches (ISODATA). Examples of classification are provided using C-band total power AIRSAR data and optical images of Santa Monica, Los Angeles.