DCGAN DATA BALANCING TO IMPROVE ACCURACY OF HYBRID CNN-LSTM INTRUSION DETECTION FRAMEWORK IN SDN ENVIRONMENT

- Network management is improved by Software-Defined Networking (SDN), which offers centralised control and programmability. Nevertheless, this centralisation also makes it susceptible to a number of assaults, such as brute force, botnet, infiltration, online attacks, Distributed Denial of Service (DDoS), and Denial of Service (DoS). In order to successfully detect and categorise these attacks, this paper suggests a hybrid deep learning model that combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks with an attention mechanism. To improve detection accuracy, the CNN component extracts spatial characteristics, the LSTM records temporal dependencies, and the attention mechanism highlights important features. The model's effectiveness in recognising various attack types in SDN environments is demonstrated by experimental assessments. Key Words: SDN Security, Intrusion Detection, CNN-LSTM Hybrid, Attention Mechanism, Cyberattack Detection.

An analysis of machine learning models for sentiment analysis of Tamil code-mixed data

The integration of renewable energy sources, such as wind and solar power, into smart grids presents significant challenges due to their inherent variability and intermittency. Accurate forecasting of renewable energy generation is essential for maintaining grid stability, minimizing energy imbalance, and optimizing power distribution. This paper proposes a hybrid deep learning model that combines Convolutional Neural Networks (CNN) for spatial feature extraction and Long Short-Term Memory (LSTM) networks for temporal dependency modeling. The approach is applied to forecast wind and solar power generation using meteorological data from multiple geographical locations. The model's performance is compared to traditional forecasting methods such as ARIMA and standalone LSTM models. Experimental results show superior forecasting accuracy and improved error margins achieved by the hybrid CNN-LSTM model, offering an enhanced solution for real-time energy management and integration into smart grid operations. This research follows the methodology proposed by Fozlur Rayhan in “A Hybrid Deep Learning Model for Wind and Solar Power Forecasting in Smart Grids” and builds upon it to demonstrate the practical application and effectiveness of hybrid deep learning models in renewable energy forecasting.

With the development of information technology and mobile communication, people's usage of emoji is increasing. However, designing emojis by artists can be time-consuming and costly. Therefore, this study attempts to use the Deep Convolution Generative Adversarial Network (DCGAN) method in deep learning to automatically generate emojis. DCGAN is a combination of Convolutional Neural Network (CNN) and Generative Adversarial Network (GAN). It introduces convolutional networks into the generative model for unsupervised training, which can improve the learning effect of the generator network. DCGAN consists of two different models: a generator model and a discriminator model. The generator's role is to produce fake images that are similar to the training images, and the discriminator's role is to judge whether the fake image is the same as a real image. During training, the generator continuously tries to defeat the discriminator by producing better and better fake images, while the discriminator tries its best to distinguish between real and fake images. Through continuous optimization of the generator and discriminator, this study can finally obtain clear new emojis. Overall, the images generated by DCGAN can largely fulfill our purpose. Many novel emojis could be obtained to replace the work of human artists. However, the quality of the generated emojis has some defects, and it is still necessary to screen out images with strange shapes that are not suitable for daily life.

This study proposes a GIS and deep learning-based integrated simulation framework to enhance the prediction and management of coastal flood risks. The framework integrates spatial datasets—including digital elevation models (DEMs), storm surge data, and terrain characteristics—preprocessed through ArcGIS Pro and Jupyter-based pipelines to estimate flood heights. A hybrid deep learning model, combining convolutional neural networks (CNNs) and long short-term memory (LSTM) networks in a parallel structure, is designed to learn spatial features and temporal dynamics simultaneously from time-series flood maps. The fused output is used to generate a probabilistic flood risk index. Experimental results using coastal datasets from Busan demonstrated that the hybrid model outperformed standalone CNN and LSTM models, achieving superior macro and weighted averages across multiple performance metrics. Notably, CNN exhibited perfect recall for early warnings, while LSTM showed the highest precision for policy decision support. The hybrid model balanced both strengths, proving robust even under imbalanced data conditions. The integration with ArcGIS Pro allows for high-resolution flood visualization and intuitive risk mapping, supporting real-time alerts and disaster response planning. This approach offers strong scalability to other coastal regions and practical utility for real-world early flood warning systems and climate resilience strategies.

This study presents a novel approach in the field of renewable energy, focusing on the power generation capabilities of floating offshore wind turbines (FOWTs). The study addresses the challenges of designing and assessing the power generation of FOWTs due to their multidisciplinary nature involving aerodynamics, hydrodynamics, structural dynamics, and control systems. A hybrid deep learning model combining Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks is proposed to predict the performance of FOWTs accurately and more efficiently than traditional numerical models. This model addresses computational complexity and lengthy processing times of conventional models, offering adaptability, scalability, and efficient handling of nonlinear dynamics. The results for predicting the generator power of a spar-type floating offshore wind turbine (FOWT) in a multivariable parallel time-series dataset using the Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) model showed promising outcomes, offering valuable insights into the model’s performance and potential applications. Its ability to capture a comprehensive range of load case scenarios—from mild to severe—through the integration of multiple relevant features significantly enhances the model’s robustness and applicability in realistic offshore environments. The research demonstrates the potential of deep learning methods in advancing renewable energy technology, specifically in optimizing turbine efficiency, anticipating maintenance needs, and integrating wind power into energy grids.

Background: In the Industry 4.0 landscape, integrating artificial intelligence (AI) with smart manufacturing is essential for enhancing automated monitoring, predictive maintenance, and system optimization. However, traditional centralized AI model training poses critical risks to data privacy, security, and scalability, especially when sensitive operational data from factory machines is shared across platforms. Methods: This study proposes a decentralized, intelligent framework designed for real-time machine monitoring that enhances fault detection accuracy while safeguarding data privacy. The approach begins with real-time sensor data acquisition—capturing vibration, temperature, and acoustic signals from distributed factory units via edge devices. These signals undergo preprocessing and advanced feature extraction using Wavelet Transform and Empirical Mode Decomposition (EMD) to reveal critical fault characteristics. Results: A hybrid deep learning model that combines Convolutional Neural Networks (CNNs) with Long Short-Term Memory (LSTM) networks is used for classification. CNNs are responsible for extracting spatial features, whereas LSTMs identify temporal dependencies in time-series. With the federated learning (FL) framework, model training can be done collaboratively across edge devices without the need to transfer sensitive raw data. Conclusion: This ensures security and enhances model generalization. Results from experiments indicate that the suggested FL-based hybrid model exceeds centralized architectures regarding detection accuracy, computational efficiency, and adaptability. This research provides a scalable and secure solution that enhances intelligent monitoring for Industry 4.0 systems.

This work explores the classification of driving behaviors using a hybrid deep learning model that combines Convolutional Neural Networks (CNNs) with Long Short-Term Memory (LSTM) networks (ConvLSTM). Sensor data are collected from a smartphone application and undergo a preprocessing pipeline, including data normalization, labeling, and feature extraction, to enhance the model’s performance. By capturing temporal and spatial dependencies within driving patterns, the proposed ConvLSTM model effectively differentiates between normal and aggressive driving behaviors. The model is trained and evaluated against traditional stacked LSTM and Bidirectional LSTM (BiLSTM) architectures, demonstrating superior accuracy and robustness. Experimental results confirm that the preprocessing techniques improve classification performance, ensuring high reliability in driving behavior recognition. The novelty of this work lies in a simple data preprocessing methodology combined with the specific application scenario. By enhancing data quality before feeding it into the AI model, we improve classification accuracy and robustness. The proposed framework not only optimizes model performance but also demonstrates practical feasibility, making it a strong candidate for real-world deployment.

Unmanned Aerial Vehicles (UAVs) are essential in numerous applications, such as surveillance and delivery services, where operational reliability and safety are paramount.This research aims to enhance fault detection techniques by utilizing deep learning models capable of interpreting complex sensor data and identifying potential issues. A hybrid deep learning model, combining Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and Gated Recurrent Units (GRU), was developed to analyze sequential sensor data. The hybrid model leverages CNN layers to extract spatial features, while LSTM and GRU layers capture temporal dependencies. Experimental results demonstrate that the hybrid CNN-LSTM-GRU model outperforms the individual models, achieving a test accuracy of 85.66% and a validation accuracy of 88.34%. These findings underscore the hybrid model's superior capability in fault detection for UAVs, highlighting its potential to improve the reliability and safety of UAV operations through advanced deep learning frameworks.

Urban rail transit systems are essential for efficient, reliable, and environmentally friendly transportation in modern cities. Correct and effective short-term passenger flow prediction is crucial for optimizing operational efficiency, enhancing service quality, and ensuring passenger safety. Traditional prediction methods often fail to capture the complex, non-linear, and dynamic patterns in urban rail transit systems. This study uses a hybrid deep learning model combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to predict short-term passenger flow in urban rail transit. By integrating the strengths of CNN in capturing spatial features and LSTM in modeling temporal dependencies, the hybrid model aims to improve prediction accuracy. Using historical data from Hangzhou Metro, the study demonstrates the model's effectiveness in predicting passenger flow, which reasonably has a good fit. Additionally, models with different convolutional layers show different performances. These improved predictions offer valuable insights for transit authorities, enabling them to make more informed decisions regarding train scheduling, resource allocation, and emergency response planning. By anticipating passenger demand more accurately, authorities can optimize the deployment of trains, reduce waiting times, enhance passenger comfort, and improve overall service reliability.

With the intensification of global climate change, accurate prediction of air quality indicators, especially PM2.5 concentration, has become increasingly important in fields such as environmental protection, public health, and urban management. To address this, we propose an air quality PM2.5 index prediction model based on a hybrid CNN-LSTM architecture. The model effectively combines Convolutional Neural Networks (CNN) for local spatial feature extraction and Long Short-Term Memory (LSTM) networks for modeling temporal dependencies in time series data. Using a multivariate dataset collected from an industrial area in Beijing between 2010 and 2015 'which includes hourly records of PM2.5 concentration, temperature, dew point, pressure, wind direction, wind speed, and precipitation' the model predicts the average PM2.5 concentration over 6-hour intervals. Experimental results show that the model achieves a root mean square error (RMSE) of 5.236, outperforming traditional time series models in both accuracy and generalization. This demonstrates its strong potential in real-world applications such as air pollution early warning systems. However, due to the complexity of multivariate inputs, the model demands high computational resources, and its ability to handle diverse atmospheric factors still requires optimization. Future work will focus on enhancing scalability and expanding support for more complex multivariate weather prediction tasks.

Image recognition technology is an important field of artificial intelligence. Combined with the development of machine learning technology in recent years, it has great researches value and commercial value. As a matter of fact, a single recognition function can no longer meet people’s needs, and accurate image prediction is the trend that people pursue. This paper is based on Long Short-Term Memory (LSTM) and Deep Convolution Generative Adversarial Networks (DCGAN), studies and implements a prediction model by using radar image data. We adopt a stack cascading strategy in designing network connection which can control of parameter convergence better. This new method enables effective learning of image features and makes predictive models to have greater generalization capabilities. Experiments demonstrate that our network model is more robust and efficient in terms of timing prediction than 3DCNN and traditional ConvLSTM. The sequential image prediction model architecture proposed in this paper is theoretically applicable to all sequential images.

Forecasting of energy consumption in Smart Buildings (SB) and using the extracted information to plan and operate power generation are crucial elements of the Smart Grid (SG) energy management. Prediction of electrical loads and scheduling the generation resources to match the demand enable the utility to mitigate the energy generation cost. Different methodologies have been employed to predict energy consumption at different levels of distribution and transmission systems. In this paper, a novel hybrid deep learning model is proposed to predict energy consumption in smart buildings. The proposed framework consists of two stages, namely, data cleaning, and model building. The data cleaning phase applies pre-processing techniques to the raw data and adds additional features of lag values. In the model-building phase, the hybrid model is trained on the processed data. The hybrid deep learning (DL) model is based on the stacking of fully connected layers, and unidirectional Long Short Term Memory (LSTMs) on bi-directional LSTMs. The proposed model is designed to capture the temporal dependencies of energy consumption on dependent features and to be effective in terms of computational complexity, training time, and forecasting accuracy. The proposed model is evaluated on two benchmark energy consumption datasets yielding superior performance in terms of accuracy when compared with widely used hybrid models such as Convolutional (Conv) Neural Network-LSTM, ConvLSTM, LSTM encoder-decoder model, stacking models, etc. A mean absolute percentage error (MAPE) of 2.00% for case study 1 and a MAPE of 3.71% for case study 2 is obtained for the proposed forecasting DL model in comparison with LSTM-based models that yielded 7.80% MAPE and 5.099% MAPE for two datasets respectively. The proposed model has also been applied for multi-step week-ahead daily forecasting with an improvement of 8.368% and 20.99% in MAPE against the LSTM-based model for the utilized energy consumption datasets respectively.

Software Defined Networking (SDN) has emerged as a technology which can replace the prevalent vendor based proprietary CLI networking devices. SDN has introduced applications based network control and provided various opportunities and challenges for research and innovation in these networks. Despite many advantages and opportunities in SDN, security is a matter of concern for developers who want to invest in SDN. In this paper we are analyzing the SDN security issues with their countermeasures. We have generalized four use cases threat model that should cover security requirements of SDN. These use cases are: (I) protect controllers from applications, (II) inter-controller protection, (III) protecting data plane or switches from controller, (IV) protecting controllers from malicious switches. We found that these SDN components are inter-related if one is secure another one is already secure. We also compared the SDN and traditional network security in terms of these four use cases and provide the insights for protection mechanism and security enhancements. A framework for the development of a SDN security application has been presented based on ryu controller. We believe that our threat model will help various researchers and developers to understand current security requirements and provide a ready reference to tackle vulnerabilities and threats in this area. Finally, we identify some open research problems and future research directions with a proposed security architecture.

Harnessing the full potential of the metal-based Laser Powder Bed Fusion process (LPBF) relies heavily on how effectively the overall reliability and stability of the manufactured part can be ensured. To this aim, the recent advances in sensorization and processing of the associated signals using Machine Learning (ML) techniques have made in situ monitoring a viable alternative to post-mortem techniques such as X-ray tomography or ultrasounds for the assessment of parts. Indeed, the primary advantage of in situ monitoring over post-mortem analysis is that the process can be stopped in case of discrepancies, saving resources. Additionally, mitigations to repair the discrepancies can also be performed. However, the in situ monitoring strategies based on classifying processing regimes reported in the literature so far operate on signals of fixed length in time, constraining the generalization of the trained ML model by not allowing monitoring processes with heterogeneous laser scanning strategies. As a part of this work, we try to bridge this gap by developing a hybrid Deep Learning (DL) model by combining Convolutional Neural Networks (CNNs) with Long-Short Term Memory (LSTM) that can operate over variable time-scales. The proposed hybrid DL model was trained on signals obtained from a heterogeneous time-synced sensing system consisting of four sensors, namely back reflection (BR), Visible, Infra-Red (IR), and structure-borne Acoustic Emission (AE). The signals captured different phenomena related to the LPBF process zone and were used to classify three regimes: Lack of Fusion (LoF), conduction mode and Keyhole. Specifically, these three regimes were induced by printing cubes out of austenitic Stainless steel (316 L) on a mini-LPBF device with operando high-speed synchrotron X-ray imaging and signal acquisition with the developed heterogeneous sensing system. The operando X-ray imaging analysis ensured that the regimes correlated with the defined process parameters. During the validation procedure of the trained hybrid DL model, the model predicted three regimes with an accuracy of about 98% across various time scales, ranging from 0.5 ms to 4 ms. In addition to tracking the model performance, a sensitivity analysis of the trained hybrid model was conducted, which showed that the BR and AE sensors carried more relevant information to guide the decision-making process than the other two sensors used in this work.