Search

The persistence of hate speech continues to pose an obstacle in the realm of online social media. Despite the continuous evolution of advanced models for identifying hate speech, the critical dimensions of interpretability and explainability have not received proportional scholarly attention. In this article, we introduce the HateInsights dataset, a groundbreaking benchmark in the field of hate speech datasets, encompassing diverse aspects of this widespread issue. Within our dataset, each individual post undergoes thorough annotation from dual perspectives: firstly, conforming to the established 3-class classification paradigm that includes hate speech, offensive language, and normal discourse; secondly, incorporating rationales that outline specific segments of a post supporting the assigned label (categorized as hate speech, offensive language, or normal discourse). Our exploration yields a significant finding by harnessing cutting-edge state-of-the-art models: even models demonstrating exceptional proficiency in classification tasks yield suboptimal outcomes in crucial explainability metrics, such as model plausibility and faithfulness. Furthermore, our analysis underscores a promising revelation concerning models trained using human-annotated rationales. To facilitate scholarly progress in this realm, we have made both our dataset and codebase accessible to fellow researchers. This initiative aims to encourage collaborative involvement and inspire the advancement of the hate speech detection approach characterized by increased transparency, clarity, and fairness.

Urdu, being a common language in South Asia, has not received significant attention in terms of language processing compared to more advanced languages. In the field of Natural Language Processing (NLP), the task of text summarization holds great importance due to its ability to comprehend textual content and generate concise summaries. Text summarization can be either extractive or abstractive in nature. While considerable efforts have been made to advance extractive summarization techniques, the limitations associated with it have been extensively explored and explained in the paper. However, the domain of abstractive summarization for the Urdu language remains largely unexplored. The challenges and underlying factors that have impeded progress in this domain have also been addressed. This paper specifically focuses on abstractive summarization of the Urdu language using supervised learning. To accomplish this, a labeled dataset consisting of Urdu text and its abstractive summaries is required. A dataset of Urdu text and its corresponding abstractive summaries has been prepared for the purpose of supervised learning. Additionally, the paper presents the results of summary generation, measured in terms of a rough score. Transformer’s encoder-decoder network was employed to generate abstractive summaries in Urdu, yielding a ROUGE-1 score of 25.18 in Urdu text summarization. Moreover, a novel evaluation metric called the “disconnection rate” has been introduced as a context-aware evaluation metric to enhance the assessment of a summary, known as the Context Aware RoBERTa Score.

In this article, a tunable RF sensor is presented for the measurement of dielectric materials (liquids and solids) based on a metamaterial resonator. The proposed novel configuration sensor has a microstrip line-loaded metamaterial resonator with tunable characteristics by utilizing a single varactor diode in the series of the resonator. CST Microwave studio is employed for 3D simulations of the tunable sensor, and the desired performance is attained by optimizing various structural parameters to enhance the transmission coefficient (S21 magnitude) notch depth performance. The proposed RF sensor can be tuned in L and S-bands using the varactor diode biasing voltage range of 0–20 V. To validate the performance of the sensor, the proposed design has been simulated, fabricated, and tested for the dielectric characterization of different solid and liquid materials. Material testing is performed in the bandwidth of 1354 MHz by incorporating a single metamaterial resonator-based sensor. Agilent’s Network Analyzer is used for measuring the S-parameters of the proposed sensor topology under loaded and unloaded conditions. Simulated and measured S-parameter results correspond substantially in the 1.79 to 3.15 GHz frequency band during the testing of the fabricated sensor. This novel tunable resonator design has various applications in modulators, phase shifters, and filters as well as in biosensors for liquid materials.

Technological advancements have revolutionized the space industry, facilitating deep space exploration using CubeSats. One objective is to locate potential life-support elements, such as water, on extraterrestrial planets. Water possesses a distinct spectral signature at 183 GHz, useful in remote sensing and environmental monitoring applications. Detecting this signature provides crucial information about water and ice presence and distribution on celestial bodies, aiding future exploration and colonization efforts. Mostly in space remote sensing uses corrugated horn antennae due to high gain and radiation patterns but fabrication of corrugated antenna is very challenging or even impossible in some cases. To ease this challenge, in our research we propose ideas to transform a corrugated horn antenna into a smooth-walled design by using MATLAB Cubic smoothing Splines algorithms. We compare simulation results between smooth-walled and corrugated antennas, and we can see some improvements in insertion losses, Voltage Standing Wave ratio (VSWR), and gain. We also manufactured this 183 GHz antenna using a commercially available 3D printer by utilizing Acrylonitrile Butadiene Styrene (ABS) material. The antenna surface was then coated with a thin layer of copper using conductive paint. In the end, we practically evaluate smooth-walled antenna functionality and compare it with the theriacal results. Validating the antenna’s functionality proposes a cost-effective and accessible production method to be used in a CubeSat engineering model or university students’ project.

Permutation flow-shop scheduling is the strategy that ensures the processing of jobs on each subsequent machine in the exact same order while optimizing an objective, which generally is the minimization of makespan. Because of its NP-Complete nature, a substantial portion of the literature has mainly focused on computational efficiency and the development of different AI-based hybrid techniques. Particle Swarm Optimization (PSO) has also been frequently used for this purpose in the recent past. Following the trend and to further explore the optimizing capabilities of PSO, first, a standard PSO was developed during this research, then the same PSO was hybridized with Variable Neighborhood Search (PSO-VNS) and later on with Simulated Annealing (PSO-VNS-SA) to handle Permutation Flow-Shop Scheduling Problems (PFSP). The effect of hybridization was validated through an internal comparison based on the results of 120 different instances devised by Taillard with variable problem sizes. Moreover, further comparison with other reported hybrid metaheuristics has proved that the hybrid PSO (HPSO) developed during this research performed exceedingly well. A smaller value of 0.48 of ARPD (Average Relative Performance Difference) for the algorithm is evidence of its robust nature and significantly improved performance in optimizing the makespan as compared to other algorithms.

Due to the drastic increase in the research publications, numerous research articles are available electronically on different online digital libraries. Some research articles or papers are not retrieved during online searches due to their classification issues. The adequately structured research articles are relatively easily approachable as compared to semi-structured and unstructured research articles, and sometimes the reader does not get accurate results on different digital libraries as the research articles are not classified properly. Neglecting the semi-structured and unstructured published research not only causes gap deficiency but also affects the results of the proposed techniques and citations for other articles. Usually, researchers missed semi-structured and unstructured research articles during their online search. Classification techniques have been applied to structured articles and no significant work has been performed towards the classification of semi-structured and unstructured research articles. Therefore, this research focuses on the classification of semi-structured research articles using different supervised classification techniques so that the most accurate and large amount of relevant research results will be achieved. For experimentation, a labeled dataset was used for the classification of semi-structured papers. The dataset we used for experimentation is comprised of manually gathered research articles from Santos repository dataset and labeling them accordingly. The current study used four different supervised classification techniques such as Support Vector Machine (SVM) classifier, Naïve Based classifier, K Nearest Neighbor classifier, and Decision Tree classifier. The comparison was performed between these supervised classification techniques to see which classifier gives better accuracy. The unit of measures or parameters selected to compare these classifiers are: accuracy, recall, precision, and f-score. The evaluation was performed on the basis of results and comparison in the experimentation. Experimental results of classifiers K-neighbor are better than other classifiers SVM, Decision Trees and Naive Bayes

Novel graphitic carbon nitride (g-C3N4) microrods were prepared by chemical method via pre-treatment of melamine and molar solution of HNO3 at a lower temperature. The phase and morphological structure was examined by XRD and SEM, while elemental composition was found by performing EDX spectroscopy. SEM results revealed that the as-prepared g-C3N4 materials has rods like morphology with average length 3–5 µm. Furthermore, the vibrational modes were studied by FTIR, and optical properties were analyzed by UV–Vis and PL spectroscopy respectively. The optical measurements showed that the band gap of g-C3N4 microrods was 2.70 eV which make them useful for photocatalytic process under visible light. The blue PL emission at 455 nm was ascribed to band-to-band transition which also confirms the activation of g-C3N4 microrods in visible spectrum range. Finally, g-C3N4 microrods were utilized as photocatalyst for the decomposition of hazardous Methyl Orange (MO) dye, and the results showed that they exhibited excellent performance (k = 0.026 min−1) visible light irradiation. Keywords: Graphitic carbon nitride, Melamine, Microrods, Photocatalyst

Labeled data is the main ingredient for classification tasks. Labeled data is not always available and free. Semi-supervised learning solves the problem of labeling the unlabeled instances through heuristics. Self-training is one of the most widely-used comprehensible approaches for labeling data. Traditional self-training approaches tend to show low classification accuracy when the majority of the data is unlabeled. A novel approach named Self-Training using Associative Classification using Ant Colony Optimization (ST-AC-ACO) has been proposed in this article to label and classify the unlabeled data instances to improve self-training classification accuracy by exploiting the association among attribute values (terms) and between a set of terms and class labels of the labeled instances. Ant Colony Optimization (ACO) has been employed to construct associative classification rules based on labeled and pseudo-labeled instances. Experiments demonstrate the superiority of the proposed associative self-training approach to its competing traditional self-training approaches.

Nanocomposites based on iron oxide/titanium oxide nanoparticles were prepared by employing green synthesis, which involved phytochemical-mediated reduction using ginger extract. XRD confirmed the composite formation, while scanning electron microscopy (SEM), dynamic light scattering (DLS), and energy-dispersive X-ray spectroscopy (EDX) was employed to investigate the particle size, particle morphology, and elemental analysis. SEM indicated the formation of particles with non-uniform shape and size distribution, while EDX confirmed the presence of Fe, Ti and oxygen in their elemental state. The surface effects were investigated by Fourier transform infrared radiation (FTIR) and impedance spectroscopy (IS) at room temperature. IS confirmed the co-existence of grains and grain boundaries. Thus, FTIR and IS analysis helped establish a correlation between enhanced surface activity and the synthesis route adopted. It was established that the surface activity was sensitive to the synthesis route adopted. The sample density, variation in grain size, and electrical resistivity were linked with surface defects, and these defects were related to temperature. The disorder and defects created trap centers at the sample’s surface, leading to adsorption of CO2 from the environment.

Evolutionary Algorithms are bio-inspired optimization problem-solving approaches that exploit principles of biological evolution. , such as natural selection and genetic inheritance. This review paper provides the application of evolutionary and swarms intelligence based query optimization strategies in Distributed Database Systems. The query optimization in a distributed environment is challenging task and hard problem. However, Evolutionary approaches are promising for the optimization problems. The problem of query optimization in a distributed database environment is one of the complex problems. There are several techniques which exist and are being used for query optimization in a distributed database. The intention of this research is to focus on how bio-inspired computational algorithms are used in a distributed database environment for query optimization. This paper provides working of bio-inspired computational algorithms in distributed database query optimization which includes genetic algorithms, ant colony algorithm, particle swarm optimization and Memetic Algorithms.

This paper presents a metamaterial sensor using a double slit complementary square ring resonator (DS-CSRR) that has been utilized for the measurement of dielectric materials, especially coal powder. The design is optimized for best performance of deep notch depth in transmission coefficient (Magnitude of S21). Sensitivity analysis of transmission coefficient with respect to structure dimensions has been carried out. Metamaterial properties of double negative permitivity and permeability were extracted from the S–parameters of this sensor. The optimized structure is fabricated using low cost FR-4 PCB board. Measured result shows resonance frequency of 4.75 GHz with a deep notch up to −41 dB. Simulated and measured results show good agreement in desired frequency band. For material characterization, first, two known materials are characterized using this metamaterial sensor. Their respective resonances and dielectric constants are known, so the transcendental equation of the sensor is formulated. Afterwards, the proposed sensor is used for dielectric measurement of two types of coal powder, i.e., Anthracite and Bituminous. The measured value of dielectric constant of Anthracite coal is 3.5 and of Bituminous coal is 2.52. This is a simple and effective nondestructive measurement technique for material testing applications.

ABSTRACT Congenital anomalies affect approximately 2 to 3% of all live births every year.1 Sonographic antenatal detection of anomalies has become a new goal of modern obstetrics management. We report a case of omphalocole with associated anomalies, its sonographic detail and clinical implications, managed at our institution.

Reversible watermarking has gained importance due to increased involvement of digital media in sensitive fields, such as medical and law enforcement. We propose a prediction error expansionbased watermarking scheme that allows embedding reversible watermark in the image with low distortion. Research work proposes four-phase representation of image which allows exploitation of larger prediction context. We have also proposed a hybrid predictor that helps enhance the prediction accuracy. To reduce image distortion at lower capacity payloads, we use sorting of estimated prediction errors through sorting of prediction context variances. For improvement at higher capacity payloads, adaptive embedding is used to determine whether to embed single or two bits in a given prediction error. The results are compared against some state-of-the-art techniques in the field and show promising results.

Control of active prosthetic hands using surface electromyography (sEMG) signals is an active research area; despite the advances in sEMG pattern recognition and classification techniques, none of the commercially available prosthetic hands provide the user with an intuitive control. One of the major reasons for this disparity between academia and industry is the variation of sEMG signals in a dynamic environment as opposed to the controlled laboratory conditions. This research investigated the effects of sEMG signal variation on the performance of a hand motion classifier due to arm position variation and also explored the effect of static position and dynamic movement strategies for classifier training. A wearable system is used to measure the electrical activity of the muscles and the position of the forearm while performing six classes of hand motions. The system is made position aware (POS) using inertial measurement units (IMUs) for different arm movement gestures. The hand gestures are decoded under both static and dynamic forearm movements. Four time domain (TD) features are extracted from the sEMG signals along with IMU-based arm position information. The features are trained and tested using linear discriminant analysis (LDA) and support vector machine (SVM) for both TD and TD-POS features. The results for the SVM show a significant difference between the static and dynamic approaches, while the TD-POS features show enhanced classification performance in comparison to the TD-based classification. Results have shown the effectiveness of the dynamic training approach and sensor fusion techniques to improve the performance of existing stand-alone sEMG-based prosthetic control systems.

This mixed-methods study explores academic engagement among full time working scholars in Libya's universities, comparing how the patterns of returning foreign-educated and domestic-educated scholars' academic engagement differs. Survey analysis indicates that where scholars are educated influences their academic engagement, with individuals educated in developed countries having a greater potential to engage (e.g., through consultancy activities, employee training, and holding temporary posts) with non-academic organisations than those educated within Libya and other developing countries. We also find that these differences in academic engagement of returned scholars are attenuated by gender, scientific discipline, and academic rank. Semi-structured interview identifies several barriers to academic engagement that affect scholars. These include sociocultural perspective on gender roles, institutional barriers, and differences in culture between academic and non-academic actors. The evidence in this study offers an empirical foundation for the necessary policy reforms required in the Higher Education sector in Libya to enhance university-industry collaboration.

A study was conducted to determine the prevalence and molecular diagnosis of fasciolosis in Lohi sheep and Beetal goats in Okara, Sahiwal and Lahore districts of Punjab province. A total of 800 samples including 200 fecal and 200 bile, each from Lohi sheep and Beetal goats were collected and processed for microscopic examination of Fasciola eggs as well as its confirmation through polymerase chain reaction (PCR). Of the fecal samples, 3.5 and 2% samples whereas from bile samples, 6.5 and 4% bile samples from sheep and goats were found microscopically positive for Fasciola hepatica, respectively. By PCR, 4.5 and 3.5% fecal samples and 8.5 and 5% bile samples from sheep and goats, respectively were confirmed for F. hepatica. Statistically sheep and goats species were found equally likely acquiring the F. hepatica diagnosed by either source of material (fecal and bile samples) similarly fecal and bile samples were equally likely showing fasciolosis in either of sheep and goats.

Background: Babesia ovis and Theileria ovis are among the important and main etiological agents causing ovine babesiosis and ovine theileriosis, causing severe economic losses among sheep and goats. The aim of the present study was to determine the prevalence and molecular diagnosis of B. ovis and T. ovis in Lohi sheep at Livestock Experiment Station Bahadurnagar, Okara, Pakistan. Methods: The prevalence of B. ovis and T. ovis was investigated in 200 Lohi sheep of mixed age and sex by PCR during 2011. The assay was employed us-ing primers Bbo-F & Bbo-R, specific for a 549-bp fragment in B. ovis genomic DNA and primers TSsr 170F & TSsr 670R, specific for a 520-bp fragment in T. ovis genomic DNA. The animals were also screened for both haemoparasites through stained thin blood smears. Results: Thirty two (16%), 48 (24%) and 26 (13%) were the number of ani-mals found positive for B. ovis, T. ovis and for mixed infection with both para-sites respectively, through microscopy. Sixty eight (34%), 73 (37%) and 42 (21%) were the number of animals found positive for B. ovis, T. ovis and for mixed infection with both parasites respectively, through PCR test. Conclusion: The results indicate the high sensitivity of PCR for surveying babesiosis and theileriosis and there is noteworthy prevalence of these diseases in sheep at an experimental station where environmental conditions are rela-tively controlled as compared to field conditions.