Search

Association rule algorithm has always been a research hotspot in the field of data mining, in the context of today’s big data era, in order to efficiently obtain association rules and effectively update them, based on the original fast update pruning (FUP) algorithm, an association rule incremental update algorithm (FBSCM) based on sorting compression matrix is proposed to solve the shortcomings of frequent scanning of transaction datasets. Firstly, The algorithm maps the transaction dataset as a Boolean matrix, and changes the storage mode of the matrix(that is, adding two columns and a row vector); Secondly, the matrix is compressed many times during the generation of frequent k-itemset; After that, the items in the matrix are sorted incrementally according to the support degree of the itemset; Finally, the original string comparison operation is replaced by the vector product of each column of the matrix. Experimental results and analysis show that the FBSCM algorithm has higher temporal performance than the traditional FUP algorithm in different incremental dataset sizes, different minimum support thresholds and different feature datasets, especially when the incremental transaction volume is large or the minimum support degree is small.

The role of communication technologies has become increasingly vital in various fields such as industrial communication, surveillance and monitoring, healthcare, and data communication, leading to a surge in demand for these technologies in recent years. In this research paper, the LEMARS model is proposed as a novel and robust lightweight encryption assisted Man-in-The-Middle (MITM) attack-resilient steganography model for secure satellite imagery services. The proposed solution combines lightweight encryption and attack-resilient embedding to achieve higher attack-resilience with low computational cost and optimal reconstruction quality. The use of Feistel architecture-driven substitution and permutation-based block-cipher encryption and heuristic-driven pixel adjustment method for MITM-Adaptive B-Channel Embedding results in higher imperceptibility and superior attack-resilience under uncertain channels. The strategic amalgamation of FSPN-BCE, L2D-IWT, and HD-OMABE enabled attack-resilient steganography for secure satellite communication. The performance of LEMARS has been evaluated using Peak Signal-to-Noise Ratio (PSNR), entropy, Non-Parametric Change Rate (NPCR), Unified Average Change Intensity (UACI), and regular and singular block pattern changes. LEMARS achieved a PSNR value of 58.02 dB, an entropy value of 6.15 dB, an NPCR value of 99.84%, a UACI value of 33.70, and exhibited minimal block pattern changes. These results indicate that the proposed model achieved higher attack-resilience with low computational cost and optimal reconstruction quality. The research aimed to improve encryption and embedding methods to achieve an MITM-attack resilient embedding solution for secure satellite imagery services. The proposed LEMARS model exhibited superiority over other state-of-the-art methods, affirming the robustness of the proposed LEMARS model for real-time multimedia data security purposes, including remote sensing, satellite imagery services, telemedicine, and EHR. In conclusion, the LEMARS model offers an optimal solution for multimedia data security, combining lightweight encryption and attack-resilient embedding to achieve higher attack-resilience with low computational cost and optimal reconstruction quality.

Lung cancer is a dangerous tumor that requires accurate diagnosis for effective treatment. Traditional diagnosis involves invasive and time-consuming histologic examination, and radiologists face challenges in localizing lung tumors. Deep neural convolutional networks are frequently used to locate lung cancer, but this is still difficult when not accounting for surrounding lung tissue. Despite progress in research, healthcare still uses deep learning models to improve the precision and sensitivity of large datasets. CNN (Convolutional Neural Network) accuracy standards are adequate, but image properties such as flips, construction, and other uncommon alignments diminish its efficiency. CNN also does not store the geometric distribution between scanned picture features. CT (Computed Tomography) and PET (Positron Emission Tomography) scans require a method that takes into consideration the spatial information of picture characteristics, as they are vulnerable to alignment problems during the perusing process. To address these issues, the authors propose MCNet (MobileNetV2 with Capsule Network), a hybrid network that adopts feature extraction and categorization from MobileNetV2, and capsule network is used to overcome the limitations of convolutional neural networks (CNNs) when it comes to processing images with abnormal orientations, such as tilting or rotation. Although CNNs are effective in processing images presented in a standard orientation, they have difficulty handling variations in image orientation. In this work, MobileNetV2 serves as a backbone network for Capsule Networks in lung cancer diagnosis. The lung image collection dataset verifies the effectiveness of MCNet, and experimental results show that MCNet technology performs better than previous state-of-the-art techniques. The proposed hybrid MCNet architecture achieves the clinical goal of lung cancer diagnosis with a lower computational cost, reducing processing time complexity and false positive rates compared to current techniques.

The motivation for this paper is to consider that in recent years, the concept of metaverse, as the latest and most popular concept in the world, has been widely applied and studied in various industries, including economic management, art design, education and teaching. However, the academic and scientific circles have not reached a consensus on whether to define the metaverse as a technology or an intelligent scene. We believe that the metaverse should be a key concept and emerging theory for constructing the future wisdom field. Therefore, in this study, our research objective is to focus on the visual art evaluation in digital works, and propose a visual art quality evaluation method in future metaverse digital works. This method is based on the quality function deployment theory and fuzzy mathematics theory in marketing. The second core point of this study is to build a field framework for the visual art evaluation of future digital works based on the metaverse by combing the current international and domestic understanding of the concept of metaverse. In addition, taking visual art quality evaluation as the research object, we have constructed a visual art quality evaluation index system for digital works under the background of metaverse. The index system is composed of one first-class index, three second-class indexes and nine third-class indexes. At the same time, we proposed a new fuzzy mathematics evaluation method in the research, called G1 entropy method. This algorithm combines subjective weighting method: G1 method and objective weighting method: entropy method as an important method of quality evaluation, and carries out the final rating through the combination weight of G1 entropy method. This study makes up for the concept of the future metaverse, introduces the gaps in the theory of visual art evaluation of future digital works, innovates the analysis of new concepts and the improvement of old methods, builds a new scene of organic combination of new technology and traditional visual art, and provides new ideas for the improvement of art quality at home and abroad in the future. In general, we sorted out the contributions of this research, including the following three aspects: (1) we constructed the metaverse field structure of digital works. By analyzing the current international and domestic research literature on the application of metaverse technology, especially the concept of metaverse in art scenes, we proposed to construct the field structure of online visual art after introducing the concept of metaverse, including blockchain technology, artificial intelligence technology Interaction technology and Internet of things technology as the four characteristics; (2) Method theoretical contribution: we further take the visual art quality evaluation as the research object, construct the index system of visual art quality evaluation of digital works under the background of metaverse, and propose an evaluation method of G1 entropy method, which is actually a method of subjective weighting by experts; (3) We use the method proposed in (2) to complete the calculation and ranking of the importance of 9 indicators in a practical case, and give some countermeasures for the calculation results of the importance of indicators. In conclusion, this study has realized the construction of new application scenarios of concepts and the new improvement of methods, and can provide theoretical and practical case experience support for the quality improvement of international and domestic metaverse visual art.

As an extension of the concepts of fuzzy set and intuitionistic fuzzy set, the concept of Pythagorean fuzzy set better models some real life problems. Distance, entropy, and similarity measures between Pythagorean fuzzy sets play important roles in decision making. In this paper, we give a new entropy measure for Pythagorean fuzzy sets via the Sugeno integral that uses fuzzy measures to model the interaction between criteria. Moreover, we provide a theoretical approach to construct a similarity measure based on entropies. Combining this theoretical approach with the proposed entropy, we define a distance measure that considers the interaction between criteria. Finally, using the proposed distance measure, we provide an extended Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) for multi-criteria decision making and apply the proposed technique to a real life problem from the literature. Finally, a comparative analysis is conducted to compare the results of this paper with those of previous studies in the literature.

Due to the importance of the multi-level fully rough interval linear programming (MLFRILP) problem to address a wide range of management and optimization challenges in practical applications, such as policymaking, supply chain management, energy management, and so on, few researchers have specifically discussed this point. This paper presents an easy and systematic roadmap of studies of the currently available literature on rough multi-level programming problems and improvements related to group procedures in seven basic categories for future researchers and also introduces the concept of multi-level fully rough interval optimization. We start remodeling the problem into its sixteen crisp linear programming LP problems using the interval method and slice sum method. All crisp LPs can be reduced to four crisp LPs. In addition, three different optimization techniques were used to solve the complex multi-level linear programming issues. A numerical example is also provided to further clarify each strategy. Finally, we have a comparison of the methods used for solving the MLFRILP problem.

With the development of convolutional neural networks, many improved algorithms have been successively proposed to promote the accuracy of dense crowd counting. However, these algorithms are deployed with expensive computing resources, which is unbearable for small devices such as embedded systems with limited computing resources. To realize the real-time counting on the small devices, it is of great significance how to trade off the computation cost and processing accuracy of the dense crowd-counting algorithm. Thus, we propose a lightweight dense crowd-counting algorithm (LCNNet) to improve this issue. Specifically, the proposed LCNNet consists of two subnetworks, a feature extraction subnetwork, and a regression subnetwork, with a bottleneck depth-separable convolution with a residuals module as the basic module. The LCNNet effectively improves computational efficiency and reduces the computational cost, which can be performed on small devices. Extensive evaluations on four benchmark datasets well demonstrate the effectiveness of the proposed LCSNet for dense crowd-counting models. Meanwhile, the proposed LCNNet can maintain a comparable level of computational accuracy and computational cost on Vehicle counting datasets.

The picture fuzzy set (PFS) is a more frequent platform for describing the degree of positive, neutral, and negative membership functions that generalizes the concept of fuzzy sets (FSs) and intuitionistic fuzzy sets (IFSs). Neutrality is a crucial component of PFS, and the score function plays a crucial role in ranking the alternatives in decision-making situations. In the decision-making process, some researchers concentrate on the various aggregation operators’ development while ignoring the development of score functions. This factor causes several errors in the existing score function. If there are two separate picture fuzzy numbers (PFNs), there should be two different scores or accuracy values. Some researchers failed to rank the alternatives when the score and accuracy values for various PFNs were equal. To overcome the shortcomings, we proposed the perfect score function in this paper for ranking PFNs and introduced strong and weak PFSs. The shortcoming of the existing score function in PFNs has been highlighted in this paper. Furthermore, the decision-making approach has been depicted based on the proposed score function, and real-world applications have been shown by ranking the COVID-19 affected regions to demonstrate its efficacy.

The rough set has been proven to be a powerful tool for dealing with uncertainty and vagueness in different fields. The hesitant fuzzy rough set is a generalization of rough sets to solve more complex problems. Since the existing hesitant fuzzy rough sets don’t satisfy the inclusive property, a novel hesitant fuzzy rough set based on the dual score functions is proposed. Then four generalized hesitant fuzzy rough set about the novel one are presented. Next, the lower approximation distribution reductions can be obtained by the discernibility matrix. Moreover, it is discovered that finding the lower approximation distribution reductions of a hesitant fuzzy decision system is equivalent to finding the minimal transversals of its hypergraph. An improved algorithm for hesitant fuzzy decision systems based on hypergraphs is presented to accelerate the reduction process. Finally, The hybrid data of Hepatitis C Virus from UCI is used as an instance background for data analyses to reflect the feasibility of the proposed algorithm.

Convolutional neural networks (CNNs) have made significant progress in the field of cloud detection in remote sensing images thanks to their powerful feature representation capabilities. Existing methods typically aggregate low-level features containing details and high-level features containing semantics to make full use of both features to accurately detect cloud regions. However, CNNs are still limited in their ability to reason about the relationships between features, while not being able to model context well. To overcome this problem, this paper designs a novel feature interaction graph convolutional network model that extends the feature fusion process of convolutional neural networks from Euclidean space to non-Euclidean space. The algorithm consists of three main components: remote sensing image feature extraction, feature interaction graph reasoning, and high-resolution feature recovery. The algorithm constructs a feature interaction graph reasoning (FIGR) module to fully interact with low-level and high-level features and then uses a residual graph convolutional network to infer feature higher-order relationships. The network model effectively alleviates the problem of a semantic divide in the feature fusion process, allowing the aggregated features to fuse valuable details and semantic information. The algorithm is designed to better detect clouds with complex cloud layers in remote sensing images with complex cloud shape, size, thickness, and cloud-snow coexistence. Validated on publicly available 38-Cloud and SPARCS datasets and the paper’s own Landsat-8 cloud detection dataset with higher spatial resolution, the proposed method achieves competitive performance under different evaluation metrics. Code is available at https://github.com/HaiLei-Fly/CloudGraph.

Brain diseases is a wide range of disorders and diseases that affect the brain. They can change a person’s behavior, personality, and capacity for thought and function. CT images are more essential than conventional clinical tests for detecting brain hemorrhage accurately. MRI images of the brain can reveal even small abnormalities in the cranial region, helping providers diagnose a wide variety of conditions, ranging from brain stroke, cancers, aneurysms, and Alzheimer’s. This paper proposes a novel Fused dual neural (FDN) network for detecting brain cancer, stroke, aneurysms, and Alzheimer using Brain Medical Images (BMI) the combination of MRI and CT. In BMI, the adaptive bilateral filter reduces noise artifacts. Google Net is used to extract features from pre-processed MRI images, and Mobile Net is used to extract features from pre-processed CT images. The integration of extracted features from Google Net and Mobile Net is fused by the Wrapper method. Finally, the Deep Belief Network is employed for classifying brain stroke, cancer, Aneurysm, and Alzheimer’s diseases using BMI images. The quantitative analysis of the suggested method is determined using the parameters like specificity, recall, precision, F1 score, and accuracy. The proposed FDN achieves a high classification accuracy rate of 98.19%, 97.68%, 94.31%, and 93.82% for detecting stroke, cancer, Aneurysm, and Alzheimer respectively. The proposed FDN model improves the overall accuracy by 5.35%, 3.14%, 9.48%, 5.33%, and 0.55% better than Faster R-CNN, CNN, Inception-V3, DCNN, and Fine-tuning Network respectively.

In this article, we present new methods for constructing uninorms on bounded lattices under the additional constraints and prove that some of these constraints are sufficient and necessary for the uninorms. Moreover, some illustrative examples for the construction of uninorms are provided. At last, we show that the additional constraints on t-norms (t-conorms) and t-subnorms (t-subconorms) of some uninorms in the literature are exactly sufficient and necessary.

In this study, a unique generative adversarial network (GAN) architectural variation was suggested, which engages in adversarial game serve by preserving an appropriate distance in the latent dimension of the network. This method overcomes the mode collapse problem with a small dataset. Extensive experiments are conducted using the segmented medical leaf dataset with various classes and the generator network is able to produce all the artificial image classes. This is accomplished by combining a unique training technique with a reasonably simple model design.

Interval type-2 trapezoidal fuzzy (IT2TrF) number is a powerful tool to depict fuzzy information. Information measures methods have received more and more attention in recent years as they play an important role in decision-making theory. A new multi-attribute decision-making (MADM) method supported by IT2TrF information measures is investigated in this paper under the IT2TrF information environment. Firstly, three axiomatic definitions of IT2TrF information measures are introduced, which include information entropy, similarity measure and cross-entropy. Secondly, with the help of the exponential function, we formulate some information measure formulas, which are followed by the proofs that the exponential entropy, exponential similarity measure and exponential cross-entropy fit the three axiomatic definitions. Subsequently, a novel IT2TrF MADM method is designed, in which the IT2TrF exponential entropy and cross-entropy are utilized to generate the attribute weights, the IT2TrF exponential similarity measure is employed to obtain the closeness degree of the ideal solution and derive the most satisfying solution. Lastly, we provide a numerical example of corporate investment to demonstrate the applicability and feasibility of the proposed MADM method. The robustness and merits of the developed MADM method are highlighted by the comparative analysis.

Aiming at the problem that manufacturing enterprises that rely more on asset projects currently lack effective means of divestiture of non performing assets, starting from incomplete information theory, this paper derives an optimal decision-making model for the divestiture path of non performing assets caused by the quality and safety crisis of manufacturing enterprises in an incomplete information environment, and uses the comprehensive method of SUMDII to establish a comprehensive integrated optimization model that reflects subjective evaluation and objective information. In addition, this study provides specific decision-making methods and implementation steps for optimizing the stripping path of non-performing assets. The empirical analysis results verify and demonstrate the feasibility, operability, accuracy, and applicability of the established model. The results show that the model designed in the study exhibits strong stability in sensitivity testing. When the parameter vectors are taken as (1,0,0,0), (1,1,0,0), (1,1,0,0), and (1,1,1,1), respectively, the ranking results corresponding to the first three parameter vectors are stable, all of which are A2 > A4 > A1 > A3. At the same time, the applied decision-making result of the model is A4 > A1 > A2 > A3, which is consistent with the best scheme evaluated by experts and superior to most comparative models. At the same time, in the analysis of decision-making characteristics, the research and design model has the most comprehensive review of decision-making elements, which is superior to other comparative models. It can be seen that the model designed by the research can lead to higher quality NPL divestiture schemes, which can help manufacturing enterprises improve asset quality and weaken the negative impact of the quality and safety crisis in manufacturing enterprises.

To address the problem that single-channel neural networks cannot fully extract text semantic features in traditional user portrait construction methods, this paper proposes a dual-channel user portrait model based on DPCNN-BIGRU and attention mechanism. The model first uses Bidirectional Encoder Representation from Transformers(Bert) and CK-means+ to obtain the fusion vector of semantic features and topic features, and then feeds the vector into Deep Pyramid Convolutional Neural Networks (DPCNN) and Bidirectional Gated Recurrent Unit (BiGRU). Deep features and global features of the text are obtained simultaneously, and then weights are assigned by the attention mechanism. Finally, the output features of the dual channels are fused and classified. It is tested on the Sogou user portrait datasets, and the experimental results prove that the dual-channel model outperforms the baseline model.

At present, there is generally little research on intangible assets of sports events, and most of the research content focuses on general research on the meaning and content of intangible assets of sports events, the operation status of regional sports intangible assets, and development strategies of sports intangible assets. From the perspective of research results, only attention has been paid to the research on process management and process control of intangible assets of sports events, However, there is a lack of systematic research on the performance evaluation of intangible assets operation and management. It is necessary to build a scientific and reasonable performance evaluation system for intangible assets operation and management of sports events based on the content of intangible assets in sports events. The performance evaluation of intangible assets operation and management in sports events is multiple attribute decision making (MADM). Recently, the TODIM and VIKOR method has been used to cope with MADM issues. The double-valued neutrosophic sets (DVNSs) are used as a tool for characterizing uncertain information during the performance evaluation of intangible assets operation and management in sports events. In this manuscript, the double-valued neutrosophic number TODIM-VIKOR (DVNN-TODIM-VIKOR) method is built to solve the MADM under DVNSs. In the end, a numerical case study for performance evaluation of intangible assets operation and management in sports events is given to validate the proposed method.

In the laser spiral welding (LSW) process, the welding parameters have a significant impact on the weld quality. In this paper, experiments were conducted and experimental data were collected on galvanized steel sheets using the LSW process, and mathematical models were developed using response surface methodology (RSM) and genetic algorithm (GA) to verify the specific effects of each process parameter on the weld and to perform process optimization. Laser power, welding speed, gap and focal length were selected as the influencing factors, and melt depth, melt width and concave as the output results. In the RSM model we found that the laser power was positively correlated with the weld depth and width, while the welding speed was inversely correlated with the weld depth and width, the gap was positively correlated with the amount of concave, and the focal length had no significant effect on the weld. In the GA model we use a large amount of experimental data for BP neural network training and iterative optimization using a genetic algorithm. Validation experiments were conducted on two models, and the results indicated that the two models had higher accuracy in predicting the welding depth and width compared to predicting the concave. The GA model had an 8% increase in tensile strength and a 25% increase in plasticity of the weld joint obtained from the optimal process compared to the RSM model. The GA model has higher accuracy in optimizing the LSW process.

To improve the recognition accuracy of badminton players’ swing movements, this study proposes a single inertial sensor based method for badminton swing movement recognition. This article proposes a badminton racket-mounted data gathering system with a single inertial sensor and proposes a real-time motion data flow-based window segmentation technique to capture motion data. On this basis, a two-layer classifier recognition model based on C4.5 Decision Tree (C4.5 T) algorithm and Random Forest (RF) method is constructed to recognize swing technical actions. Using the C4.5 T to identify the swing style of athletes; The RF method is used to recognize the swing technical action. The final experiment showed that the method studied achieved a recognition accuracy of 95.36% for six common swing movements. The proposed model has good application prospects in the recognition of badminton swing movements. However, due to the limitations of the experimental conditions, the recognition effect of this method on more complex swing movements needs to be further verified.

Support vector machine (SVM) is a supervised binary classifier with good generalization ability and excellent computational properties. It has been widely used in many fields such as image recognition, bioinformatics and so on. However, the traditional SVM requires the input data to be clear and knowable, while in the actual application process, there will be many cases that the input data is uncertain. In order to solve this problem, a new SVM model is proposed in this paper by combining the uncertainty theory with the SVM theory. The uncertainty theory was proposed by Liu in 2007, and it is often used to describe the uncertainty of things. The uncertain set in uncertainty theory is often used to model unsharp concepts. Therefore, this paper regards each uncertain data as an uncertain set and establishes a SVM model with uncertain chance constraints. However, the uncertain chance constraints are non-convex. Therefore, this paper gives the equivalent transformation process of constraint conditions when the input data are triangular uncertain sets. Finally, the non-convex constraint conditions are transformed into the linear constraint conditions, so that the model is transformed into a nonlinear programming model. In the numerical experiment, the Particle Swarm Optimization (PSO) algorithm is used to solve the problem, which proves the feasibility of the model.

Low power wide area networks (LPWANs) are made to survive conditions of extensive installation. Technological innovations, including Global Network Operator, Long Range Wide Area Network (LoRaWAN), Narrowband Internet of Things (NB-IoT), Weightless, Sigfox, etc., have adopted LPWANs. LoRaWAN is currently regarded to be one of the most cutting-edge and intriguing technology for the widespread implementation of the IoT. LoRaWAN is currently regarded to be one of the most cutting-edge and intriguing technology for the widespread implementation of the IoT. Although LoRaWAN offers the best features that make it fit with Internet - of - things specifications, there are still certain technical issues to overcome, such as link coordination, resource allocation and reliable transmission. In LoRaWAN, End-devices transmit randomized uplink frames to the gateways using un-slotted random-access protocol. This randomness with the restrictions placed on the gateways is a reason that leads to a considerable decline in network performance, in particular downlink frames. In this paper, we propose a new approach to increase Acknowledgement (ACK) messages throughput. The suggested method takes advantage of both class A and class B features to enhance and assist LoRaWAN’s reliability by ensuring that an ACK message is sent for every confirmed uplink while retaining the minimum energy level that is utilized by nodes.

Regression analysis is a potent tool to explore the relationship of variables and widely used in many areas. Classical statistics assume that the residual of regression model should follow the Gauss-Markov hypothesis. However, in many cases, the data is not obeying this hypothesis particularly real-life data. Therefore, this paper explores the Von Bertalanffy regression model under the framework of the uncertainty theory, and employs the uncertain maximum likelihood estimation (MLE) to estimate the unknown parameters. Furthermore, the uncertain hypothesis test and an algorithm for data modification which aimed to find outliers and modify data are studied, then the forecast value and confidence interval be formulated. Finally, a real-life numerical example of applying the above theories be given, this example shows that the uncertain MLE has better performance compare with the uncertain least squares and the least absolute deviations methods. Consequently, the uncertain MLE is a better way to deal with the real-life data.

In the current era of economic and cultural globalization, the demand for packaging design is increasing, and the packaging design requirements for brands and products in the entire consumer market are becoming increasingly strict and refined. Designers are facing one greater challenge after another. Packaging design, as a comprehensive discipline that integrates commerce, culture, and art, diverse cultural forms and elements will run through the entire process of brand packaging design. Good products need to be sold through good channels in order to achieve greater value. As a bridge connecting products and the market, the role of packaging design is even more evident. It can not only improve the recognition of products, but also help enterprises stand out among many similar products, thereby increasing product value. The brand packaging design effect evaluation is a classical multiple attribute group decision making (MAGDM) problem. This paper extends the EDAS method to the 2-tuple linguistic Pythagorean fuzzy sets (2TLPFSs). On the basis of the original EDAS method, 2-tuple linguistic Pythagorean fuzzy EDAS (2TLPF-EDAS) method is built for MAGDM. Finally, a case study for brand packaging design effect evaluation show that the new method proposed in this paper is reasonable.

In present world the major cause of global warming and climate change are emission of carbon and greenhouse gas. Governments and policymakers around the world want to put their best efforts to control the pollution and climate change to save our environment. To reduce greenhouse gas emissions Government and policymakers takes carbon tax policy on carbon emission. Also in real world uncertainty is a pervasive phenomenon. Humans have a significant ability to make logical decisions based on uncertain information. For this purpose, we are developing a pollution control fuzzy production inventory model with imperfect and break-ability items under preservation technology investment and carbon tax policy. In this model, the break-ability rate is dependent on inventory level as the break-ability rate of breakable items depends on the collected stress of inventory stock level. Here the unit production cost is dependent on raw material cost, wear-tear cost and development cost. Carbon emission is controlled by investing in carbon reduction technology and a fraction of product items are imperfect. In this study demand of the product depends on selling price and inventory stock level of product. Finally, this optimal control problem solved by using Pontryagin Maximum principle and the optimal results are illustrated graphically and numerically using MATLAB software. Subsequently, some sensitivity analysis is investigated as the impact of parameters on total profit.

In the past decade, people’s life is getting better and better, and the attention to sports competition is also increasing. In the current information age, sports and athletes’ data are very important, especially team football. In college, football coaches can use the data to analyze the situation of college football players and opposing players to better specify the corresponding tactics to win the game. However, at present, most of the data results need to be manually recorded and counted on the spot or after the game. In the process of statistics, Zhou Jing will inevitably have omissions and other problems. For this problem, a method based on space-time graph convolution. In the process, machine vision and motion recognition methods are combined, and the joint movements of different football players are extracted through the pose estimation method to obtain motion recognition results. To ented the methods on the KTH dataset. The results showed that the football motion recognition using the research method reached 98% on the dataset, which significantly improved the accuracy of nearly 5% over the existing state-of-the-art methods. At the same time, the accuracy rate of football movements was less than 5%. This means that the research method can effectively identify football sports, and can be widely used in other fields, and promote the development of human movement recognition in human-computer interaction and smart city and other fields.

Glaucoma is a condition of the eye that is caused by an increase in the eye’s intraocular pressure that, when it reaches its advanced stage, causes the patient to lose all of their vision. Thus, glaucoma screening-based treatment administered in a timely manner has the potential to prevent the patient from losing all of their vision. However, because glaucoma screening is a complicated process and there is a shortage of human resources, we frequently experience delays, which can lead to an increase in the proportion of people who have lost their eyesight worldwide. In order to overcome the limitations of current manual approaches, there is a critical need to create a reliable automated framework for early detection of Optic Disc (OD) and Optic Cup (OC) lesions. In addition, the classification process is made more difficult by the high degree of overlap between the lesion and eye colour. In this paper, we proposed an automatic detection of Glaucoma disease. In this proposed model is consisting of two major stages. First approach is segmentation and other method is classification. The initial phase uses a Stacked Attention based U-Net architecture to identify the optic disc in a retinal fundus image and then extract it. MobileNet-V2 is used for classification of and glaucoma and non-glaucoma images. Experiment results show that the proposed method outperforms other methods with an accuracy, sensitivity and specificity of 98.9%, 95.2% and 97.5% respectively.

Machine learning (ML) has been applied in civil engineering to predict the compressive strength of concrete with high accuracy. In this paper, five boosting ensemble algorithms, i.e., XGBoost, AdaBoost, GBDT, LightGBM, and CatBoost, were used to predict the compressive strength of high-performance concrete (HPC). The models were evaluated using performance indicators such as R2, root mean square error (RMSE), and mean absolute error (MAE). The results showed that the CatBoost model had the highest accuracy with a R2 (0.970) and a RMSE (2.916). The prediction accuracy of the model was increased through hyperparameter optimization, which got a higher with a R2 (0.975) and a RMSE (2.863). Meanwhile, the SHapley Additive exPlanations (SHAP) method was used to explain the output results of the optimal model (CatBoost), which generated explainable insights that further revealed the complex relationship between the prediction model parameters. The results showed that AGE, W/B, and W/C had the most impact on high-performance concrete compressive strength (HPCCS) prediction, which was similar to the results of sensitivity analysis. This study provided a theoretical basis and technical guidance for developing the mix design of a new high-performance concrete (HPC) system. In the future, the interpretable results of the model output should be iteratively checked and validated in the actual laboratory in order to provide guidance for engineering practice.

In literature granular computing and formal concept analysis algorithm use only single-value attributes to knowledge discovery for the data of spatio-temporal aspects. However, most of the datasets like forest fires and tornado storms involve multiscale values for attributes. The limitation of single-value attributes of the existing approaches indicates only the data related to event occurrence which may be missing the elicitation of important knowledge related to severity of event occurrence. Motivated by these limitations, this research article proposes a novel and generalized method which uses ordinal semantic weighted multiscale values for attributes in formal concept analysis with granular computing measures especially when spatio-temporal attributes are not given. The originality of proposed methodology is using ordinal semantic weighted multiscale values for attributes that give complete information of event occurrences. Moreover, the use of ordinal semantic weighted multiscale values improves the results of granular computing measures. The significance of proposed approach is well explained by experimental evaluation performed on publicly available datasets on storm occurring in different States of America.

Many recent studies have shown that Euler’s elastica regularization performs better than the famous total variation (TV) regularization on keeping image features in smooth regions during the process of denoising. In addition, an adaptive weighted matrix combined with total variation is a key technique for well restoring local features of image. Considering these two factors, in this paper, we propose an adaptive Euler’s elastica model for Poisson image restoration so as to well preserve both image features in smooth regions and local features of image. To solve this non-smooth and non-convex model efficiently, we design an alternating direction method of multipliers. Experiments on both natural and synthetic images illustrate the effectiveness and efficiency of the proposed method over the state-of-the-art methods in Poisson restoration and denoising, respectively. In particular, for Poisson restoration, our proposed method improves the TV method up to 2.46 about PSNR for dealing with the Peppers image with Gaussian blur and noise level σ = 1. In addition, the proposed method for Poisson denoising gets higher PSNR and SSIM values than the TAC method, while costing less CPU time.

In this paper, we have defined some fuzzy topological invariants for particular types of uniform fuzzy graph. Some particular useful types of uniform fuzzy graphs are Uniform Edge Fuzzy Graph, Uniform Vertex Fuzzy Graph, Uniform Vertex-Edge Fuzzy Graph and Totally Uniform Fuzzy Graph. For each particular type we have defined different kinds of degrees in a graph in accordance with the unique nature of it. In the end, we have applied all our output results to a cellular neural fuzzy graph as an example, to verify the predicting ability of topological invariants. The aim of this paper is to define more significant fuzzy topological invariants in fuzzy graphs. Our ideas will help to create a link between fuzzy graph theory and simple (crisp) graph theory.

In the last two decades, regardless of epidemiological, and clinical studies, the incidence of breast cancer (BC) is still increasing. However, so far, a lot of research has been done in this field to diagnose BC, and some of them have been discussed in the literature section. But still, happening major issues while dealing with fault feature matrix, generated from traditional feature extraction methods. As a result, the complexity of fault classification has raised, which will negatively impact fault identification’s accuracy and effectiveness. Thus, in this research, a novel hybridized machine learning-fuzzy and dimension reduction (MLF-DR) model has been proposed to improve the decision capabilities and efficiency of an ML model. A feature-based class-togetherness fuzzification method has been used for every feature. The novelty of our research work is to find all possibilities between cancerous and non-cancerous cells by implementing a fuzzy inference system (FIS) in the data analysis phase, and DR techniques at preprocessing phase to select the best optimizing features. This research tries to reduce the incidence of BC and prevent needless deaths, thus will probably follow necessary action to perform i.e. (i) FIS to interpret input values; (ii) principal component analysis (PCA), and recursive feature elimination (RFE) to select best features, and (ii) logistic regression (LR) and random forest (RF) models to predict BC with these features. Furthermore, all the experiments have been done on Wisconsin Breast Cancer Dataset (WBCD), freely available on the Kaggle repository using Python programming on Jupyter Notebook version 6.4.3. The key findings of this research are that the LR-PCA (8 components) model can reliably and successfully obtain the defect diagnosis results with 99.1% accuracy, as compared to individual LR and RF models.

This paper presents the concepts of a complex intuitionistic fuzzy subfield (CIFSF) and the direct product of a complex intuitionistic fuzzy subfield which is generalized from the concept of a complex fuzzy subfield by adding the notion of intuitionistic fuzzy into a complex fuzzy subfield. The main contribution and originality of this research are adding the non-membership term to the definition of a complex fuzzy subfield that assigns for any element a complex-valued grade. We expand the complex fuzzy subfield and obtain a new structure called CIFSF. This new concept is innovative in that it may attain a wider range of values for both membership and non-membership functions where these functions are expanded to the unit disc in the complex plane. Furthermore, we discuss that the direct product of two CIFSFs is CIFSF, and some related properties are investigated. In addition, we present the definition of necessity and possibility operators on the direct product of CIFSF, and some associated theorems are given. Finally, we propose the level subsets of the direct product of two complex intuitionistic fuzzy subsets of a field and prove that the level subset of the direct product of two CIFSFs is a subfield and discuss some related results.

Reducing the dimensions of the original data set while preserving the information as much as possible is conducive to improving the accuracy and efficiency of the model. To achieve this, this paper presents a multi-strategy African vulture optimization algorithm that is the chaotic and elite opposition-based African vulture optimization with the simplex method and differential evolution strategy(CESDAVO). Three main improvements are introduced into African vultures optimization(AVO) to improve its capabilities in this study. Firstly, the chaotic elite opposition-based learning strategy is used to initialize and diversify individual positions of vultures. Secondly, the simplex method is used to optimize those poor individuals so as to further improve the local exploitation ability of the algorithm. Thirdly, the differential evolution strategy is used to make the algorithm escape from the local optimum and improve the global optimization capability of the algorithm. The results of the ablation experiments show that mixing the three strategies greatly improves the optimization performance of the algorithm. In addition, Nine algorithms are compared with CESDAVO on 15 benchmark functions, and this experimental result shows that its optimization capability is superior to the others. Then, the proposed CESDAVO is employed for feature selection, and 12 standard datasets are used for experiments. According to the experimental results, CESDAVO obtained the highest average classification accuracy on 11 datasets and the highest feature selection rate on 8 datasets, which is significantly better than other algorithms. Finally, CESDAVO is also applied to feature reduction for essays, removing 24 features and significantly improving the classification accuracy on multiple classifiers.

Despite the studies probing the factors associated with the adoption of 5G technology products, the current state of knowledge about this new technology products is still fragmented. Previous research has mainly concentrated on the “cumulative impact” of factors rather than the complexities of behavior. Based on complexity theory, this article aims to explain the behavioral intention of using 5G technology products by evaluating the combination of functional (i.e. price and quality), social (i.e. environmental awareness and knowledge) and consumer personality (i.e. openness and consciousness) factors as causal configuration. A sample of 150 Chinese consumers was examined using the fuzzy set qualitative comparative analysis (fsQCA) technique. The fsQCA outcomes illustrate that five different and effective configurations of functionality, social, and consumer personality factors exist to attain apex level intention of using 5G technology products. This article extends the existing literature by espousing a new procedural attitude to divulge the intricacy of 5G products adoption. It also provides valuable suggestions for 5G technology service managers and manufacturers to enhance the technology, social, and consumer personality features combination to implement 5G technology products successfully.

As one of the fundamental tasks in natural language processing, Multi-Label Text Classification (MLTC) is used to mark one or more relevant labels for a given text from a large set of labels. Existing MLTC methods have increasingly focused on improving classification effectiveness by fusing the correlations of labels. Still, the research suffers from difficulties in comprehensively extracting text features and distinguishing similar labels. This paper proposes a multi-label text classification model based on keyword extraction and attention mechanism. The model proposes using keywords to represent labels, adopting both self-attention and interactive attention mechanisms (between labels and text) to extract text features and create text vectors. Finally, fusing text vectors as the classifier’s input. Experiments are conducted on two public datasets and a self-built dataset of illegal advertisements. The experimental results show that the keyword-based label representation approach proposed in this paper can better obtain label semantics, avoid noise and improve the performance of the multi-label text classification.

Fuzzy sets provide an effective method for dealing with uncertain and imprecise problems. For data of intermediate fuzzy distribution, membership degrees of objects whose attribute values are larger or smaller than the normal value would be the same and carried out the same decision. However, objects with different values mean that the information they contain is different for the decision-making problem. The decision process of calculating membership degrees in fuzzy set will lose the information of data itself. Therefore, bilateral fuzzy sets and their three-way decisions are proposed. First, the deviation degree is proposed in order to distinguish these objects. Compared with the membership degree, the deviation degree extends the mapping range from [0, 1] to [- 1, 1]. For six typical membership functions, their corresponding deviation functions are discussed and deduced. Second, the concept of bilateral fuzzy sets is proposed and the corresponding operation rules are analyzed and proved. Then, three-way decisions and approximations based on bilateral fuzzy sets are constructed. Next, for the optimization of threshold, principle of least cost is extended to the three-way decisions model based on bilateral fuzzy sets, and theoretical derivation is carried out. Finally, based on probability statistics, the principle based on confidence interval is proposed, which provides a new perspective for threshold calculation.

Here, we develop a fuzzy controller using a new online self-adapting design. The objective of this work is to control a nonlinear process by using a one-dimensional input rule variable, instead of error and error variation. The initial limits of the fuzzy logic membership functions are mostly depend on experiments and previous knowledge of the dynamic process behaviors. Generally, the membership function parameters have a significant impact on control signal amplitude and, consequently on the convergence and stability of the controller-plant system. The proposed technique determines the limits of the antecedent membership functions online using the k th and k - 1 th outputs of the controlled plant and reference model, respectively. Meanwhile, the limits of the consequent membership functions are calculated using error and error variation. This approach ensures: (i) that the input/output variables have the required fuzzy space, (ii) the controlled plant follows the desired reference model, and (iii) the control signal amplitude is within acceptable limits. Additionally, (iiii) it takes into account the dynamic variability of the process and the existence of an overshoot. The membership function parameters are updated continuously through a self-adapting procedure, ensuring improved control performance. Ultimately, the proposed approach is improved using two nonlinear systems.

In order to improve the optimal setting temperature problem to achieve the global optimum of product performance, costs and benefits. In this article, a hierarchical structure optimal setting approach of production indexes for the RHFTF is proposed. It is composed of three layers with different functions to obtain the temperature control setting model of the RHFTF. In the first layer, the bi-feature Gaussian mixture model clustering (BFGMMC) algorithm of loading plan is proposed to optimize the setting of a limited number of slabs. In the second layer, the type-2 fuzzy rule interpolation (T2FRI) setting method is developed to obtain the optimal setting curve. Meanwhile, an improved KH (Kóczy-Hirota) α-cut distance (IKHCD) algorithm is proposed to get the miss information between any two adjacent interpolation points. In the third layer, knowledge feedforward compensation of rule matrices (KFCRM) algorithm is presented to improve the anti-interference ability of the setting model. The results of the study can demonstrate that the proposed method improves the accuracy of the model and optimizes the control strategy. Furthermore, the experimental results show that the proposed method meets the process technical requirements.

A λ-subset, or a [0,λ]-valued fuzzy subset, is a mapping from a nonempty set to the interval [0,λ]. In this paper, we use the notion of hemimetrics, a kind of distance functions, as the basic structure to define and study fuzzy rough set model of λ-subsets by using the usual addition and subtraction of real numbers. We define a pair of fuzzy upper/lower approximation operators and investigate their properties and interrelations. These two operators have nice logical descriptions by using the related Lukasiewicz logical systems. We show that upper definable sets, lower definable sets and definable sets are equivalent, and they form an Alexandrov fuzzy topology. A processing of a λ-subset via fuzzy upper/lower approximation operators can actually considered as a processing of the related image, and thus has potential applications in image processing.

We studied China’s Common Prosperity process by assessing and comparing the level of Common Prosperity in different regions of China and made some beneficial recommendations to government departments. The research data comes from the China Statistical Yearbook, which includes data from 31 provinces and cities from 2015 to 2020. According to the relevant research, eleven evaluation indicators were selected. We combined GRA with the TOPSIS method for scoring and the K-means clustering algorithm for dividing the GRA-TOPSIS scoring results into three evaluation levels. Then, the convolutional neural network model was used to predict and simulate the level of common prosperity. Taking 2020 as an example, the results show: (1) From 2015 to 2020, China’s Common Prosperity level reached its highest point in 2020. Due to the impact of COVID-19 in 2019, the scores of 31 regions are generally lower than in the previous four years. The situation changed in 2020; (2) In terms of regional distribution, the economic development of Beijing, Shanghai, and other eastern regions is relatively good, with a higher degree of Common Prosperity than that of other regions; (3) The average prediction accuracy is high in our model. It can be close to 100%, indicating that the model has a good prediction effect. In addition, we made recommendations based on the research results, which have good references for actively promoting common prosperity.

Hydropower station is vital for the stable growth of the national economy. How to timely warn the possible faults of hydropower stations has become an increasingly popular research topic. The traditional detection model is difficult to detect the small abnormal changes in the data, and these abnormal changes are often the precursor of faults. To improve the sensitivity of the traditional detection model, this study introduced a weight factor into the traditional LSTM detection model. By using the correction mechanism, the LSTM correction model makes the prediction model never deviate from the normal track following the appearance of abnormal data. This ensures that the model can generate large residuals after abnormal data occur so that we can detect these abnormal data in time. Finally, this paper puts forward two factors related to equipment health and integrates these two factors to form a health index. The results show that the LSTM correction model based on the health index can not only detect small changes that cannot be detected by traditional detection models but also knows the wear and tear of equipment during operation based on the changes in health indicators.

In order to improve the accuracy and security of encrypted holographic 3D geographic information data acquisition and improve the actual resolution of geographic information files, a blind watermarking algorithm for encrypted holographic 3D geographic information data based on mapping mechanism is proposed. According to the characteristics of the mapping mechanism, a mapping mechanism structure diagram is constructed; Under the mapping mechanism technology, blind watermark data is preprocessed. Then, a watermark embedding operation is performed to obtain the watermark information image, and then a blind watermark that encrypts the holographic three-dimensional geographic information data is extracted. Finally, using the blind watermark signal as input, the blind watermark information is obtained by using the watermark strength, and the holographic 3D geographic data information is segmented and encrypted to complete blind watermark detection. The blind watermark algorithm for encrypting the holographic 3D geographic information data is studied. The results show that the maximum difference between the correlation coefficient of the algorithm in this paper and the correlation coefficient of the unaffected algorithm is only 0.04, which has better anti attack performance, high security, good terrain information collection ability, high data accuracy, and can achieve curvature repair of information data.

Background: Image semantic segmentation can be understood as the allocation of a predefined category label to each pixel in the image to achieve the region segmentation of the image. Different categories in the image are identified with different colors. While achieving pixel classification, the position information of pixel points of different categories in the image is retained. Purpose: Due to the influence of background and complex environment, the traditional semantic segmentation methods have low accuracy. To alleviate the above problems, this paper proposes a new real-time image semantic segmentation framework based on a lightweight deep convolutional encoder-decoder architecture for robotic environment sensing. Methodology: This new framework is divided into three stages: encoding stage, decoding stage and dimension reduction stage. In the coding stage, a cross-layer feature map fusion (CLFMF) method is proposed to improve the effect of feature extraction. In the decoding stage, a new lightweight decoder (LD) structure is designed to reduce the number of convolutional layers to speed up model training and prediction. In the dimension reduction stage, the convolution dimension reduction method (CDR) is presented to connect the encoder and decoder layer by layer to enhance the decoder effect. Results: Compared with other state-of-the-art image semantic segmentation methods, we conduct comparison experiments on datasets Cityscapes, SUN RGB-D, CamVid, KITTI. The Category iIoU combined with the proposed method is more than 70%, and the Category IoU is as high as 89.7%. Conclusion: The results reflect that the new method can achieve the better semantic segmentation effect.

The point features of low-texture images are insufficient and unreliable, so it is difficult to achieve good alignment and easy to damage the image structure. To solve these problems, in this paper, we propose a new image stitching method by using the sigmoid function to create perception mask. Firstly, the point features and line features are used to improve the accuracy of image registration and the naturalness of distortion. Secondly, an energy function is used to optimize the alignment model. Finally, we propose to use sigmoid function to create perception mask image to reduce artifacts and retain image structure. The gradient domain fusion algorithm is combined to achieve image fusion. Experimental results are provided to demonstrate that the proposed method is superior to some previous methods in reducing artifacts and maintaining image structure.

Multi-modal information outbreak is consistently increasing in social media. Classification of tweet sentiments using various information modalities will help the recommender systems to achieve success in digital marketing. Moreover, aspect-level sentiment analysis categorizes a target’s sentiment polarity in a specific environment. Using topic modelling in aspect-level sentiment analysis enables the identification of more accurate aspect-based tweet sentiments. The existing sentiment classification techniques used for the development of recommendation systems do not focus on the aspect-based approach modelled using deep learning classifier with temporal analysis on the social media data. Hence, this paper proposes an efficient sentiment classification model that highlights the impact of topic modelling-based word feature embedding for improvising the classification of Twitter sentiments and product reviews based on temporal reasoning and analysis for performing predictive analysis. For tweets context analysis, Latent Dirichlet Allocation based topic modelling is used in this work which generates the topics. For each topic, the sentiment is calculated separately and the topic guided feature expansion is done using Senti-wordnet. Moreover, an extended deep learning classification algorithm called Long Short-Term Memory (LSTM) with word embedding and temporal reasoning(LSTMWTR) is proposed in this paper for improving the classification accuracy. Finally, the labelled data are classified using the existing machine learning algorithms namely Naïve Bayes, Support Vector Machines and also using the deep learning models such as Convolution Neural Network(CNN),LSTM, Recurrent Neural Networks (RNN) and the transformer model namelyBi-directional Encoder Representation from Transformers (BERT),Convolution Bi-directional Recurrent Neural Network (CBRNN) and the proposed deep learning algorithm namelyLSTMWTR. These sentiment classification algorithms have been evaluated with word embedding for tweet sentiment classification and product review classification. The results obtained from this work show that the proposed LSTMWTR algorithm emerges as the highly accurate model for tweet sentiment and product review classification.

Multi-granulation decision-theoretic rough set effectively combines Bayesian decision approaches with multi-granulation rough set theory, and provides an important theoretical framework for studying rough set. In this paper, we explore several extensional models of multi-granulation decision-theoretic rough sets under the normal distribution of the decision loss function. Using the 3σ rule of normal distribution, we transform the decision loss of the multi-granulation decision-theoretic rough set into a set of interval values. We construct the upper and lower approximations from the optimistic, weakly optimistic, pessimistic, weakly pessimistic, optimistic-pessimistic, weakly optimistic-pessimistic, pessimistic-optimistic, and weakly pessimistic-optimistic viewpoints, and provide the decision rules of the proposed rough set models. The work in this paper brings the decision behavior based on a multi-granulation decision-theoretic rough set closer to the actual situation.

Recently, the algorithmic trading of financial assets is rapidly developing with the rise of deep learning. In particular, deep reinforcement learning, as a combination of deep learning and reinforcement learning, stands out among many approaches in the field of decision-making because of its high performance, strong generalization, and high fitting ability. In this paper, we attempt to propose a hybrid method of recurrent reinforcement learning (RRL) and deep learning to figure out the algorithmic trading problem of determining the optimal trading position in the daily trading activities of the stock market. We adopt deep neural network (DNN), long short-term memory neural network (LSTM), and bidirectional long short-term memory neural network (BiLSTM) to automatically extract higher-level abstract feature information from sequential trading data, respectively, and then generate optimal trading strategies by interacting with the environment in a reinforcement learning framework. In particular, the BiLSTM consisting of two LSTM models with opposite directions is able to make full use of the information from both directions in attempting to capture more effective information. In experiments, the daily data of Dow Jones, S&P500, and NASDAQ (from Jan-01, 2005 to Dec-31, 2020) are applied to verify the performance of the newly proposed DNN-RL, LSTM-RL, and BiLSTM-RL trading systems. Experimental results show that the proposed methods significantly outperform the benchmark methods, such as RRL and Buy and Hold, with higher scalability and better robustness. Especially, BiLSTM-RL performs better than other methods.

Since prostate cancer is one of the most important causes of death in today’s society, the investigation of why and how to diagnose and predict it has received much attention from researchers. The cooperation of computer and medical experts provides a new solution in analyzing these data and obtaining useful and practical models, which is deep learning. In fact, deep learning as one of the most important tools for analyzing data and discovering relationships between them and predicting the occurrence of events is one of the practical tools of researchers in this way. This study segments and classifies prostate cancer using a deep learning approach and architectures tested in the ImageNet dataset and based on a method to identify factors affecting this disease. In the proposed method, after increasing the number of data based on removing dominant noises in MRI images, image segmentation using a network based on deep learning called faster R-CNN, and then feature extraction and classification with architecture Various deep learning networks have reached the appropriate accuracy and speed in detection and classification. The aim of this study is to reduce unnecessary biopsies and to choose and plan treatment to help the doctor and the patient. Achieving the minimum error in the diagnosis of malignant lesion with a criterion called Sensitivity of 93.54% and AUC equal to 95% with the ResNet50 architecture has achieved the goal of this research.

Brain tumor is an anomalous growth of brain cells. Segmentation of brain tumors is currently the most important surgical and pharmaceutical procedure. However, manually segmenting the brain tumor is a challenging task due to the complex structure of brain. In recent years, artificial intelligence techniques with the fuzzy logic have shown better results in the field of medicine. In this work, a novel deep learning classification network with fuzzy hexagonal membership function (DLC-FHMF) model has been proposed for accurately segmenting brain tumors. The different MRI modalities namely T1, T1-c, T2 and Flair images are preprocessed using a fuzzy hexagonal trilateral and median filter to eliminate the Rician noise. Afterwards, the DLC-FHMF model is used for segmenting the tumor portion by using the multimodal composition of MRI as input. The fuzzy weights are determined with hexagonal membership functions and convoluted with the corresponding MRI images. The quantitative examination is carried out using the performance metrics namely accuracy, specificity, precision, sensitivity, incorrect segmentation, under-segmentation, and over-segmentation. In addition to the above metrics, the pre-processing metrics include PSNR, RMSE, and SSIM. The experimental fallout portrayals that the proposed DLC-FHMF approach attains a better accuracy range of 99% for detecting brain tumors using the BRATS 2013 dataset. The proposed DLC-FHMF model improves the overall accuracy by 15.1%, 11.1%, 3.0%, 21.2% and 0.5% better than ANN, SVM, NB, DNN and DAE respectively.

Attribute reduction is a widely used technique in data preprocessing, aiming to remove redundant and irrelevant attributes. However, most attribute reduction models only consider the importance of attributes as an important basis for reduction, without considering the relationship between attributes and the impact on classification results. In order to overcome this shortcoming, this article firstly defines the distance between samples based on the number of combinations formed by comparing the samples in the same sub-division. Secondly, from the point of view of clustering, according to the principle that the distance between each point in the cluster should be as small as possible, and the sample distance between different clusters should be as large as possible, the combined distance is used to define the importance of attributes. Finally, according to the importance of attributes, a new attribute reduction mechanism is proposed. Furthermore, plenty of experiments are done to verify the performance of the proposed reduction algorithm. The results show that the data sets reduced by our algorithm has a prominent advantage in classification accuracy, which can effectively reduce the dimensionality of high-dimensional data, and at the same time provide new methods for the study of attribute reduction models.