National University

QueryAssist is a model designed to enhance communication with databases by transforming Telugu natural language queries, both text and speech, into SQL queries. Built for government schools of the Telugu-speaking states in India, this model utilizes Word Vectors for semantic analysis, ensuring accurate query generation. QueryAssist acts as an intuitive interface to interact with SQL databases, by addressing challenges faced by schools in accessing and utilizing data. Its standout features are its ability to comprehend Telugu queries and its error handling and refinement system. Through extensive experiments, QueryAssist has proven its effectiveness in transforming natural language queries into SQL commands. The model’s architecture, results, and its potential to improve the quality of decision-making processes within government schools are presented in this paper.

Wireless connections with high capacity, security, and affordability are becoming increasingly crucial for the growth of interactive multimedia and broadband services. A promising approach to meet this need is the use of radio frequency (RF) and optical fiber technology to distribute millimeter-wave signals. This article provides a summary of current research on the radio over fiber (RoF) technology and future uses for it in next-generation networks. Firstly, we introduce the basics of RoF technology, including its various components and architectures. We provided a comparative analysis between External Modulation Based and Direct Modulation based RoF link. On the basis of simulation analysis, the measured Q-Factor is 303.064 and 5.50 while using External and Direct modulation schemes respectively, for 1 dB/km optical fiber impairments. We also discuss the benefits and challenges of employing RoF technology in wireless access networks, highlighting the key issues that require attention for RoF technology to fully realize its potential. Afterward, we provide an extensive review of recent research on RoF technology. We examine the performance and limitations of used RoF link and identify the key research challenges in the associated field. Finally, we discuss the future directions and opportunities for research in RoF technology, our article aims to provide easy to understand of RoF technology and its impact on the next- generation networks.

Innumerable approaches of deep learning-based COVID-19 detection systems have been suggested by researchers in the recent past, due to their ability to process high-dimensional, complex data, leading to more accurate prediction of the COVID-19 infected patients. There is a visible dominance of Convolutional Neural Network (CNN) based models analysing chest images like X-rays and Computed Tomography (CT) scans for prediction, while the utilization of audio data for the same is less prevalent. Considering the respiratory system is one of the primary means by which the SARS-CoV-2 virus spreads, respiratory sounds are a potential biomarker for determining the presence of COVID-19. In this paper, we propose a novel approach for the detection of COVID-19 from amidst a dataset comprising of respiratory sound samples of healthy, COVID-19, and other lung diseases which are often misinterpreted as COVID-19. The approach employs an autoencoder for anomaly detection and a Long Short-Term Memory (LSTM) network for the detection of COVID-19 from amongst other lung diseases. The first stage of the model comprises an encoder-decoder-based autoencoder model with baseline reconstruction error, trained in an unsupervised environment, to reconstruct “healthy” audio signals. An LSTM based multi-class classifier is proposed for the second stage to classify the infected samples into the five classes: COVID-19, Bronchiolitis, COPD, Pneumonia and URTI. The experimental results demonstrate the efficacy of our proposed approach in detecting COVID-19 from a 5-class test set of audio samples of patients suffering from respiratory disease, with an accuracy of 98.7%, and an AUC of 1.

Deep learning techniques allow us to learn about a person’s behavior based on pictures and videos. Using digital cameras, the system can identify and classify a person’s behavior based on images and videos. This paper aims to present a method for detecting drivers’ drowsiness based on deep learning. To determine which transfer learning technique best suits this work, we used DenseNet169, MobileNetV2, ResNet50V2, VGG19, InceptionV3, and Xception on the dataset. The dataset used in this paper is the Driver Drowsiness Dataset (DDD), which is publicly available on Kaggle. This dataset consists of 41,790 RGB images, and each image has a size of 227 $$\times$$ 227, which has 2 classes: drowsy and not drowsy. The Drivers Drowsiness Dataset is based on the images extracted from the real-life Drowsiness dataset (RLDD). After comparing the results coming from all 6 models, the highest accuracy achieved was 100% by ResNet50V2, and various parameters are calculated like accuracy, F1 score, etc. Additionally, this work compared the results with existing methods to demonstrate its effectiveness.

In this study, two complexes [Co(C7H9N3S2)2Cl2]Cl2, and [Ni(C7H9N3S2)2Cl2]Cl2 were synthesized from the ligands known as 2-Acetyl thiophene thiosemicarbazone (C7H9N3S2), respectively. C7H9N3S2 was characterized using various characterization techniques. The study of magnetic moment values (4.92 B.M. for [Co(C7H9N3S2)2Cl2]Cl2 and (2.93 for [Ni(C7H9N3S2)2Cl2]Cl2) shows that the complexes are paramagnetic with octahedral geometry. The value of electrical conductance (184 Ohm−1 cm2 mole−1 for [Co(C7H9N3S2)2Cl2]Cl2 and (173 Ohm−1 cm2 mole−1 for [Ni(C7H9N3S2)2Cl2]Cl2) suggested that ligand to metal ratio is 2:1 in its structure. In addition, the electronic spectrum analysis (8000–8650, 20,800–21,580, and 16,200–17,500 cm−1 suggested that the [Co(C7H9N3S2)2Cl2]Cl2 is spin-free octahedral complex. Similar way, the electronic spectrum values (9500–10,415, 14,200–14,940, and 23,500–24,000 cm−1 suggested that the [Ni(C7H9N3S2)2Cl2]Cl2 is in octahedral geometry. An infrared spectroscopy study showed that each equivalent ligand was attached to the metal with C=N, and C=S moiety using nitrogen and sulfur atoms. The vibration bands of νM-Cl were also observed at 325 for [Co(C7H9N3S2)2Cl2]Cl2 and 350 cm−1 [Ni(C7H9N3S2)2Cl2]Cl2. This observation confirms that Cl− ion is also coordinated with the metal ion. The article shows the synthesis and structure of new metal complexes [Co(C7H9N3S2)2Cl2]Cl2 and [Ni(C7H9N3S2)2Cl2]Cl2 based on 2-Acetyl thiophene thiosemicarbazone ligands.

Plants are a major food source worldwide, and to provide a healthy crop yield, they must be protected from diseases. However, checking each plant to detect and classify every type of disease can be time-consuming and would require enormous expert manual labor. These difficulties can be solved using deep learning techniques and algorithms. It can check diseased crops and even categorize the type of disease at a very early stage to prevent its further spread to other crops. This paper proposed a deep-learning approach to detect and classify cauliflower diseases. Several deep learning architectures were experimented on our selected dataset VegNet, a novel dataset containing 656 cauliflower images categorized into four classes: downy mildew, black rot, bacterial spot rot, and healthy. We analyzed the results conducted, and the best test accuracy reached was 99.25% with an F1-Score of 0.993 by NASNetMobile architecture, outperforming many other neural networks and displaying the model’s efficiency for plant disease detection.