Search

Recently, the emergence of social media has opened the way for online harassment in the form of hate speech and offensive language. An automated approach is needed to detect hate and offensive content from social media, which is indispensable. This task is challenging in the case of social media posts or comments in low-resourced CodeMix languages. This paper investigates the efficacy of various multilingual transformer-based embedding models with machine learning classifiers for detecting hate speech and offensive language (HOS) content in social media posts in CodeMix Dravidian languages that belong to the low-resource language group. Experiments were conducted on six sets of openly available datasets in Kannada-English, Malayalam-English and Tamil-English languages. The objective is to identify a single pre-trained embedding model that commonly works well for HOS tasks in the above mentioned languages. For this, a comprehensive study of various multilingual transformer embedding models, such as BERT, DistilBERT, LaBSE, MuRIL, XLM, IndicBERT, and FNET for HOS detection was conducted. Our experiments revealed that MuRIL pre-trained embedding performed consistently well for all six datasets using Support Vector Machine (SVM) with Radial Basis Function (RBF) kernel. In a set of experiments conducted on six datasets, the highest accuracy results for each dataset are as follows: DravidianLangTech 2021 achieved 96% accuracy for Malayalam, 72% accuracy for Tamil, and 66% accuracy for Kannada. For HASOC 2021 Tamil, the accuracy reached 76%, and for HASOC 2021 Malayalam, it reached 68%. Additionally, HASOC 2020 demonstrated an accuracy of 92% for Malayalam. Moreover, we performed an in-depth error analysis and a comparative study, presenting a tabulated summary of our work compared to other top-performing studies. In addition, we employed a cost-sensitive learning approach to address the class imbalance problem in the dataset, in which minority classes get higher classification weights than the majority classes. The weights were initialized and fine-tuned to obtain the best balance between all the classes. The results showed that incorporating the cost-sensitive learning strategy avoided class bias in the trained model. In addition to the aforementioned points, a significant contribution of our research presented in this paper is introducing a novel annotated test set for Malayalam-English CodeMix. This new dataset serves as an extension to our existing data, known as the Hate Speech and Offensive Content Identification in English and Indo-Aryan Languages (HASOC) 2021 Malayalam-English dataset.

The voltage fluctuations are caused by variations in renewable energy source outputs, increased usage of nonlinear loads, high reactive power consumption of loads, etc. This results in damage to electric components and enormous economic losses. It is necessary to measure the parameters of voltage fluctuations and flicker levels to achieve a secure supply of power. However, higher-order harmonics and noises in voltage waveform make such assessment a burdensome task. The traditional techniques fail to provide exact monitoring due to the uncharacteristic variations in the input voltage signal. Nowadays, data-driven methods have become a prominent choice for non-stationary, nonlinear signal analysis as it efficiently tackles atypical changes by its capability to identify spatio-temporal information from given data. The present paper proposes a data-driven hybrid methodology for monitoring voltage fluctuations using the low-rank approximation of dynamic mode decomposition (DMD) and flat-top finite impulse response (FIR) filter. The low-rank approximation of DMD enables to identify the spatio-temporal coherent structures from the data by mapping the data into a lower-dimensional space. Adaptive flat-top FIR filter tracks the fundamental dynamic phasor by identifying the instantaneous frequency, magnitude and phase variations. The performance of the proposed hybrid methodology is assessed using different test cases, and is compared with various existing approaches. The promising results suggest that the proposed methodology can be used for accurate parameter identification and monitoring voltage fluctuations in smart grid scenarios. It can also be used as an accurate tool in distribution grids to detect higher-order harmonics.

Deep learning (DL) algorithms are widely used in object detection such as roads, vehicles, buildings, etc., in aerial images. However, the object detection task is still considered challenging for detecting complex structures, oil pads are one such example: due to its shape, orientation, and background reflection. A recent study used Faster Region-based Convolutional Neural Network (FR-CNN) to detect a single oil pad from the center of the image of size 256 × 256. However, for real-time applications, it is necessary to detect multiple oil pads from aerial images irrespective of their orientation. In this study, FR-CNN was trained to detect multiple oil pads. We cropped images from high spatial resolution images to train the model containing multiple oil pads. The network was trained for 100 epochs using 164 training images and tested with 50 images under 3 different categories. with images containing: single oil pad, multiple oil pad and no oil pad. The model performance was evaluated using standard metrics: precision, recall, F1-score. The final model trained for multiple oil pad detection achieved a weighted average for 50 images precision of 0.67, recall of 0.80, and f1 score of 0.73. The 0.80 recall score indicates that 80% of the oil pads were able to identify from the given test set. The presence of instances in test images like cleared areas, rock structures, and sand patterns having high visual similarity with the target resulted in a low precision score.

Dense Residual U-Net (DRU-Net) is a neural network used for image segmentation. It is based on the U-Net architecture and isa combination of modified ResNet as the encoder and modified DenseNet as the decoder blocks. DRU-Net captures both the local and contextual information. Previous studies on DRU-Net have not tested the influence of the spectral resolution of the images. In an earlier study, the DRU-Net was trained with grayscale images for epiphyte segmentation. The network trained and tested with grayscale images underperformed while varying the illumination and occupancy of the target in the frame. In this study, the same network was trained and tested with RGB images for assessing the increase in overall learning. The performance of the network in segmenting epiphytes under conditions such as good/poor illumination and high/low target occupancy was analyzed. Dice and Jaccard scores were used as evaluation metrics. The DRU-Net model trained with RGB images had an improvement of 20% over the grayscale model in both average Dice and average Jaccard scores of the target class. Based on the higher Dice and Jaccard scores, adding additional spectral information improves DRU-Net learning. The increased computation time required for training DRU-Net with RGB images will result in better output. This model could be further used for identifying multiple epiphytes in images with poor illumination and different occupancy conditions.

The deep learning (DL) models require timely updates to continue their reliability and robustness in prediction, classification, and segmentation tasks. When the deep learning models are tested with a limited test set, the model will not reveal the drawbacks. Every deep learning baseline model needs timely updates by incorporating more data, change in architecture, and hyper parameter tuning. This work focuses on updating the Conditional Generative Adversarial Network (C-GAN) based epiphyte identification deep learning model by incorporating 4 different generator architectures of GAN and two different loss functions. The four generator architectures used in this task are Resnet-6. Resnet-9, Resnet-50 and Resnet-101. A new annotation method called background removed annotation was tested to analyse the improvement in the epiphyte identification protocol. All the results obtained from the model by changing the above parameters are reported using two common evaluation metrics. Based on the parameter tuning experiment, Resnet-6, and Resnet- 9, with binary cross-entropy (BCE) as the loss function, attained higher scores also Resnet-6 with MSE as loss function performed well. The new annotation by removing the background had minimal effect on identifying the epiphytes.

Deep learning (DL) methods are used for identifying objects in aerial and ground-based images. Detecting vehicles, roads, buildings, and crops are examples of object identification applications using DL methods. Identifying complex natural and man-made features continues to be a challenge. Oil pads are an example of complex built features due to their shape, size, and presence of other structures like sheds. This work applies Faster Region-based Convolutional Neural Network (R-CNN), a DL-based object recognition method, for identifying oil pads in high spatial resolution (1m), true-color aerial images. Faster R-CNN is a region-based object identification method, consisting of Regional Proposal Network (RPN) that helps to find the area where the target can be possibly present in the images. If the target is present in the images, the Faster R-CNN algorithm will identify the area in an image as foreground and the rest as background. The algorithm was trained with oil pad locations that were manually annotated from orthorectified imagery acquired in 2017. Eighty percent of the annotated images were used for training and the number of epochs was increased from 100 to 1000 in increments of 100 with a fixed length of 1000. After determining the optimal number of epochs the performance of the algorithm was evaluated with an independent set of validation images consisting of frames with and without oil pads. Results indicate that the Faster R-CNN algorithm can be used for identifying oil pads in aerial images.

Unmanned Aerial Vehicle (UAV) missions often collect large volumes of imagery data. However, not all images will have useful information, or be of sufficient quality. Manually sorting these images and selecting useful data are both time consuming and prone to interpreter bias. Deep neural network algorithms are capable of processing large image datasets and can be trained to identify specific targets. Generative Adversarial Networks (GANs) consist of two competing networks, Generator and Discriminator that can analyze, capture, and copy the variations within a given dataset. In this study, we selected a variant of GAN called Conditional-GAN that incorporates an additional label parameter, for identifying epiphytes in photos acquired by a UAV in forests within Costa Rica. We trained the network with 70%, 80%, and 90% of 119 photos containing the target epiphyte, Werauhia kupperiana (Bromeliaceae) and validated the algorithm’s performance using a validation data that were not used for training. The accuracy of the output was measured using structural similarity index measure (SSIM) index and histogram correlation (HC) coefficient. Results obtained in this study indicated that the output images generated by C-GAN were similar (average SSIM = 0.89–0.91 and average HC 0.97–0.99) to the analyst annotated images. However, C-GAN had difficulty to identify when the target plant was away from the camera, was not well lit, or covered by other plants. Results obtained in this study demonstrate the potential of C-GAN to reduce the time spent by botanists to identity epiphytes in images acquired by UAVs.

Security breaches due to attacks by malicious software (malware) continue to escalate posing a major security concern in this digital age. With many computer users, corporations, and governments affected due to an exponential growth in malware attacks, malware detection continues to be a hot research topic. Current malware detection solutions that adopt the static and dynamic analysis of malware signatures and behavior patterns are time consuming and have proven to be ineffective in identifying unknown malwares in real-time. Recent malwares use polymorphic, metamorphic, and other evasive techniques to change the malware behaviors quickly and to generate a large number of new malwares. Such new malwares are predominantly variants of existing malwares, and machine learning algorithms (MLAs) are being employed recently to conduct an effective malware analysis. However, such approaches are time consuming as they require extensive feature engineering, feature learning, and feature representation. By using the advanced MLAs such as deep learning, the feature engineering phase can be completely avoided. Recently reported research studies in this direction show the performance of their algorithms with a biased training data, which limits their practical use in real-time situations. There is a compelling need to mitigate bias and evaluate these methods independently in order to arrive at a new enhanced method for effective zero-day malware detection. To fill the gap in the literature, this paper, first, evaluates the classical MLAs and deep learning architectures for malware detection, classification, and categorization using different public and private datasets. Second, we remove all the dataset bias removed in the experimental analysis by having different splits of the public and private datasets to train and test the model in a disjoint way using different timescales. Third, our major contribution is in proposing a novel image processing technique with optimal parameters for MLAs and deep learning architectures to arrive at an effective zero-day malware detection model. A comprehensive comparative study of our model demonstrates that our proposed deep learning architectures outperform classical MLAs. Our novelty in combining visualization and deep learning architectures for static, dynamic, and image processing-based hybrid approach applied in a big data environment is the first of its kind toward achieving robust intelligent zero-day malware detection. Overall, this paper paves way for an effective visual detection of malware using a scalable and hybrid deep learning framework for real-time deployments.