Internet of Things (IoT) security is one of the prominent issues that has gained significant attention among researchers in recent times. The recent advancements in IoT introduce various critical security issues and increase the risk of privacy leakage of IoT data. The implementation of Blockchain (BC) can be a potential solution for security issues in the IoT. This paper presents a perceptible description of the security threats in IoT, as well as the security characteristics and challenges introduced during the integration of BC with IoT. An analysis of different consensus protocols and existing security techniques is discussed briefly. A comparative analysis of a number of Distributed Ledger Technology (DLT) platforms based on both quantitative and qualitative evaluation criteria is also presented. This paper explores the role Of Blockchain Technology (BCT) in improving security in Intrusion Detection Systems (IDS) and other applications in the IoT environment. In addition, the paper also outlines the open issues and highlights possible research opportunities that can be beneficial for future research.

Background: Assessment of root development and apical closure is critical in dental disciplines, including endodontics, trauma management, and age estimation. This study aims to leverage advances in deep learning Convolutional Neural Networks (CNNs) to automatically evaluate the apical region status of permanent first molars, highlighting a digital health application of AI in dentistry. Methods: In this retrospective study, 262 Cone Beam Computed Tomography (CBCT) scans were reviewed, and 147 anonymized dental images were cropped from pseudopanoramic radiographs, including standard measurements. Tooth regions were resized to 471 × 1075 pixels and split into training (80%) and test (20%) sets. CNN performance was assessed using accuracy, precision, recall, F1-score, and receiver operating characteristic (ROC) curves with area under the curve (AUC), demonstrating AI-based image analysis in a dental context. Results: Precision, recall, and F1-scores were 0.79 for open roots and 0.81 for closed roots, with a macro average of 0.80 across all metrics. The overall accuracy and AUC were also 0.80. Conclusions: These results suggest that CNNs can be effectively used to assess apical patency from ROI images derived from pseudopanoramic radiographs.

Identity and Access Management (IAM) that can be both secure and flexible has become critical as corporate networks continue expanding in size and complexity. The agility required to address changing cybersecurity threats like phishing attempts, insider attacks, and behavioral anomalies that circumvent rule-based controls is frequently lacking in traditional IAM systems. This article presents MyD, an AI-driven IAM framework that incorporates supervised machine learning and hybrid deep-learning approaches for realtime anomaly detection in order to overcome these constraints. Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) layers have been utilized in the system architecture to process real-world behavioral data that was obtained by enterprisewide monitoring agents and trained on Random Forest classifiers. This design supports the detection of minute departures from normal behavior by allowing the system to learn both temporal and spatial patterns in user activities. An analysis of a dataset with more than 8,900 labeled records generated an AUC value of 0.3135, an F1 score of 30.78%, and a recall rate of 78.14%. These results indicate the sensitivity of the framework to anomalous behavior and its ability to perform in situations with unbalanced data. The MyD system shows great promise for real-time, user-specific, scalable threat detection in business settings. All things considered, the experimental results confirm its significance as a cutting-edge solution for intelligent IAM infrastructure.

Generalized Zero-Shot Learning (GZSL) aims to recognize classes in the test dataset that do not have image samples in the training dataset. GZSL tasks typically place all classes into a semantic space using predefined semantic attributes for seen and unseen classes. Since there are no real images for unseen classes, classifying these classes correctly is a challenging task. To overcome this challenge, synthetic features for unseen classes are generated using generative networks. Based on this approach, we proposed a generator-based GZSL model that selects the best samples using machine learning methods for the generated synthetic features. In our proposed model, we preferred semantically rich representations instead of traditional semantic attributes for semantic information representation. Variational Autoencoder (VAE) and Generative Adversarial Network (GAN) were used together to generate synthetic features. We applied k-means and DBSCAN clustering algorithms to the generated synthetic features and then classified them. To evaluate our proposed model, we conducted experiments on well-known GZSL benchmark datasets AWA2, CUB, and FLO. We extended our experiments to include open-set classes. Comprehensive experiments showed GZSL classification performances of 67.8% on AWA2, 77.0% on CUB, and 92.4% on FLO. Additionally, we observed the improving effect of k-means and DBSCAN clustering algorithms on GZSL classification performance.

Video and image denoising techniques aim to eliminate noise while preserving image details. However, in the process of noise reduction, while some image texture may get lost, residual noise artifacts can persist. So far, no single model has achieved universal success across all types and levels of noise. In this study, we present a noise reduction model that combines deep learning methods using limited hardware resources. Specifically, we leverage a combination of Convolutional Neural Network (CNN) and Generative Adversarial Network (GAN) architecture, which has demonstrated effectiveness in preserving image structure as the depth of the model increases. Additionally, we enhance visual quality by incorporating GAN model after the CNN network. Our evaluation of the proposed approach reveals superior performance at low noise levels compared to previous neural network-based methods. By utilizing the findings of this study, we gain insights into the more efficient utilization of deep learning and traditional noise reduction methods, explore their behavior at different noise levels, and compare results obtained from diverse datasets. The proposed method outperforms other CNN-based methods at certain noise levels, thereby providing valuable prior knowledge to researchers in this field.

In this paper, we propose a novel hybrid deep learning model designed for predicting the contribution of SMEs to the economy. Our hybrid model aims to address the challenge of accurately forecasting SMEs' economic contribution by leveraging the strengths of CNN in capturing spatial relationships and patterns and LSTM's ability to capture sequential dependencies. We present the architecture of the hybrid model and describe the data preprocessing steps, feature extraction, and model training. To evaluate the effectiveness of the proposed model, we conducted experiments on a comprehensive dataset of SMEs' economic indicators in Türkiye. The results demonstrate that the hybrid deep learning model outperforms traditional forecasting methods and standalone neural network models. The incorporation of CNN and LSTM enables the model to capture intricate relationships within the data, leading to more accurate and robust predictions.

In recent years, the use of artificial intelligence in medicine, as in many other fields, has begun to increase considerably. Creating magnetic resonance (MR) reports manually by medical doctors is a very difficult, time-consuming, and potentially error-prone process. In order to address these problems, a deep learning-based image captioning model is proposed in this study to automatically generate reports from brain MRIs. In the developed model, image processing, natural language processing, and deep learning methods are used together to produce text for the content and diagnoses in the medical image. First, pre-processing, such as rotating at random angles, changing size, cropping, changing brightness and contrast, adding shadows, and mirroring, were performed for MR images. Then, a model that generates reports was developed by utilizing the Bootstrapping Language Image Pre-Training (BLIP) model and the transformer architecture of the model. The experimental studies showed that the proposed model had successful results; the produced reports were highly similar to the original reports and could be used as a supplementary tool in medicine.

Electricity has been considered one of the most important resources in the world since its invention. Due to its nature, it is not possible to store it in large quantities after production. Due to this structure, it is extremely important to produce electricity at the right time and in the right quantities. The day-ahead electricity market, which is one of the electricity markets in Turkey, is a market where agreements are made one day before physical electricity trading. Participants in the market bid for the electricity they can produce the next day for a certain time period. The submitted bids are compared with the market clearing price (PTF) and the bids below the PTF are accepted. Since the PTF is not known from the previous day in this structure, it creates uncertainty. Correct PTF estimation helps market participants to make strategic decisions by minimizing their risks.

A deep learning approach based on Recurrent Neural Networks (RNN) was applied in this study due to the suitability of the data for time series analysis for the estimation of PTF. The data sets created in the study were created and compared by analyzing different sizes and unusual situations such as holidays and weekends. The performance of the model was evaluated with common error metrics such as Mean Absolute Deviation (MAD), Mean Square Error (MSE) and Mean Percentage Error (MAPE). In the model training, optimization studies were performed on hyperparameters such as epoch, batch size, activation function, optimizer selection, time step value and the success of the model was compared on different data sets. When the model results were examined, it was shown that the accuracy rates increased significantly especially in the data set where the data pre-processing step was performed by removing weekend and official holiday data. According to the results obtained as a result of hyperparameter optimizations, it is seen that the performance of the proposed RNN model is at a level that can compete with the results obtained in previous studies conducted with similar methods when compared with the literature.

As a result, this study shows that the RNN-based deep learning approach can be an alternative method for PTF estimation in the day-ahead electricity market and reveals that it can be an effective estimation tool that market participants can use in their decision-making processes. In future studies, it is aimed to work with larger data sets of the model and to further increase the model performance with different deep learning architectures.

In recent years, high bandwidth and cost-efficient wireless network technologies have emerged as a competition factor in process of constituting own infrastructure. On the other hand, today's challenges of designing an effective layout have become an important problem for public institutions and private companies with increasing requests. There are some methods to solve this problem. In this study, a new approach based on a genetic algorithm is proposed to solve the mentioned problem. A simulation is developed to test the success of the algorithm. The most effective layout design of access points is constituted by the distance between access points and the communication channels used in the developed simulation. The obtained experimental results showed that the proposed algorithm successfully achieved the challenge of designing an access point layout in terms of total coverage area and average bandwidth per user.

Although Knowledge Graphs (KGs) are widely used, they suffer from hosting false information. In the literature, many studies have been carried out to eliminate this deficiency. These studies correct triples, relations, relation types, and literal values or enrich the KG by generating new triples and relations. The proposed methods can be grouped as closed-world approaches that take into account the KG itself or open-world approaches using external resources. The recent studies also considered the confidence of triples in the refinement process. The confidence values calculated in these studies affect either the triple itself or the ground rule for rule-based models. In this study, a propagation approach based on the confidence of triples has been proposed for the refinement process. This method ensures that the effect of confidence spreads over the KG without being limited to a single triple. This makes the KG continuously more stable by strengthening strong relationships and eliminating weak ones. Another limitation of the existing studies is that they handle refinement as a one-time operation and do not give due importance to process performance. However, real-world KGs are live, dynamic, and constantly evolving systems. Therefore, the proposed approach should support continuous refinement. To measure this, experiments were carried out with varying data sizes and rates of false triples. The experiments have been performed using the FB15K, NELL, WN18, and YAGO3-10 datasets, which are commonly used in refinement studies. Despite the increase in data size and false information rate, an average accuracy of 90% and an average precision of 98% have been achieved across all datasets.

With the widespread use of social networks, blogs, forums and e-commerce web sites, the volume of user generated textual data is growing exponentially. User opinions in product reviews or in other textual data are crucial for manufacturers, retailers and providers of the products and services. Therefore, sentiment analysis and opinion mining have become important research areas. In user reviews mining, topic modeling based approaches and Latent Dirichlet Allocation (LDA) are significant methods that are used in extracting product aspects in aspect based sentiment analysis. However, LDA cannot be directly applied on user reviews and on other short texts because of data sparsity problem and lack of co-occurrence patterns. Several studies have been published for the adaptation of LDA for short texts. In this study, a novel method for aspect based sentiment analysis, Sentence Segment LDA (SS-LDA) is proposed. SS-LDA is a novel adaptation of LDA algorithm for product aspect extraction. The experimental results reveal that SS-LDA is quite competitive in extracting products aspects.

Day by day the usage of infrared cameras has been increasing in the world. With the increasing use of thermal infrared cameras and images, especially in military, security and medicine, the need for coloring thermal infrared images to visible spectrum has arisen. In this study, a deep based model has been developed to generate visible spectrum images (RGB - Red Green Blue) from thermal infrared (TIR) images. In the proposed model, an encoder-decoder architecture with skip connections has been used to generate RGB images. KAIST-MS (Korea Advanced Institute of Science and Technology-Multispectral) dataset used for training and test the developed model. The experimental results extensively tested using Least Absolute Deviations (L1), Root Mean Squared Error (RMSE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index Measure (SSIM).

A new weighted multi-attribute based recommender system (WMARS) has been developed using extended user behavior analysis. WMARS obtained data from number of clicked items in the recommendation list, sequence of the clicked items in recommendation the list, duration of tracking, number of tracking same item, likes/dislikes, association rules of clicked items, remarks for items. WMARS has been applied to a movie web site. The experimental results have been obtained from a total of 567 heterogeneous users, including employers in different sectors, different demographic groups, and undergraduate and graduate students. Using different weighted sets of the attributes’ parameters, WMARS has been tested and compared extensively with collaborative filtering. The experimental results show that WMARS is more successful than collaborative filtering for the data set that was used.

This paper proposes an artificial neural network (ANN) to obtain the electron energy distribution functions (EEDFs) in SF₆ and argon from the following: 1) mean energies; 2) the drift velocities; and 3) other related swarm data. In order to obtain the required swarm data, the electron swarm behavior in SF₆ and argon is analyzed over the range of the density-reduced electric field strength E/N from 50 to 800 Td from a Boltzmann equation analysis based on the finite difference method under a steady-state Townsend condition. A comparison between the EEDFs calculated by the Boltzmann equation and by ANN for various values of E/N suggests that the proposed ANN yields good agreement of EEDFs with those of the Boltzmann equation solution results.

Fuzzy logic controllers (FLC) have gained popularity in the past few decades with successful implementation in many areas, including electrical machines’ drive control. Many colleges are now offering fuzzy logic courses due to successful applications of FLCs in nonlinear systems. However, teaching students a fuzzy logic controlled drive system in a laboratory, or training technical staff, is time consuming and may be an expensive task. This paper presents an educational tool for fuzzy logic controlled brushless direct current motor (BDCM), which is a part of a virtual electrical machinery laboratory project. The tool has flexible structure and graphical interface. Motor and controller parameters of the drive system can be changed easily under different operating conditions.

In this paper, an adaptive neuro-fuzzy inference system (ANFIS) has been presented to speed control of a switched reluctance motor (SRM). SRMs have become an attractive alternative in variable speed drives due to their advantages such as structural simplicity, high reliability, high efficiency and low cost. But, the SRM performance often degrades for the machine parameter variations. The SRM converter is difficult to control due to its nonlinearities and parameter variations. In this study, to tackle these problems, an adaptive neurofuzzy controller is proposed. Heuristic rules are derived with the membership functions then the parameters of membership functions are tuned by ANFIS. The algorithm has been implemented on a digital signal processor (TMS320F240) allowing great flexibility for various real time applications. Experimental results demonstrate the effectiveness of the proposed ANFIS controller under different operating conditions of the SRM.

Switched reluctance motor (SRM) is increasingly employed in industrial applications where variable speed is required because of their simple construction, ease of maintenance, low cost and high efficiency. However, the SRM performance often degrades for the machine parameter variations. The SRM converter is difficult to control due to its nonlinearities and parameter uncertainties. In this paper, to overcome this problem, a neuro fuzzy controller (NFC) is proposed. Heuristic rules are derived with the membership functions of the fuzzy variables tuned by a neural network (NN). The algorithm is implemented on a digital signal processor (TMS320F240) allowing great flexibility for various real time applications. Experimental results demonstrate the effectiveness of the NFC with various working conditions of the SRM.