Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
191 results
Search Results
Item Semantic Similarity and Paraphrase Identification for Malayalam Using Deep Autoencoders(Springer Science and Business Media Deutschland GmbH, 2021) Praveena, R.; Anand Kumar, M.; Padannayil, K.P.In this chapter, we deal with the sentence-level paraphrase identification for the Malayalam language. We use recursive autoencoder architecture for the unsupervised learning of phrase representations to extract features for paraphrase identification. Sentence’s features of varying lengths are converted to fixed-size representation using the convolution method of dynamic pooling. Initially, the Malayalam paraphrase identification system was designed to identify paraphrases and non-paraphrases alone and later extended to identify semi-equivalent paraphrases. Along with semantic features, conventional statistical features are further taken into account, resulting in improved system performance. The proposed system was implemented using word2vec embedding and obtained 77.67% accuracy for the two-class system and 66.07% for the three-class system. This chapter also discusses different experiments done for choosing the best parameters and embedding models. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.Item Deep Learning for COVID-19(Springer Science and Business Media Deutschland GmbH, 2022) Bs, B.S.; Manoj Kumar, M.V.; Thomas, L.; Ajay Kumar, M.A.; Wu, D.; Annappa, B.; Hebbar, A.; Vishnu Srinivasa Murthy, Y.V.S.Ever since the outbreak in Wuhan, China, a variant of Coronavirus named “COVID 19” has taken human lives in millions all around the world. The detection of the infection is quite tedious since it takes 3–14 days for the symptoms to surface in patients. Early detection of the infection and prohibiting it would limit the spread to only to Local Transmission. Deep learning techniques can be used to gain insights on the early detection of infection on the medical image data such as Computed Tomography (CT images), Magnetic resonance Imaging (MRI images), and X-Ray images collected from the infected patients provided by the Medical institution or from the publicly available databases. The same techniques can be applied to do the analysis of infection rates and do predictions for the coming days. A wide range of open-source pre-trained models that are trained for general classification or segmentation is available for the proposed study. Using these models with the concept of transfer learning, obtained resultant models when applied to the medical image datasets would draw much more insights into the COVID-19 detection and prediction process. Innumerable works have been done by researchers all over the world on the publicly available COVID-19 datasets and were successful in deriving good results. Visualizing the results and presenting the summarized data of prediction in a cleaner, unambiguous way to the doctors would also facilitate the early detection and prevention of COVID-19 Infection. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.Item Convolutional Neural Network-Enabling Speech Command Recognition(Springer Science and Business Media Deutschland GmbH, 2023) Patra, A.; Pandey, C.; Palaniappan, K.; Sethy, P.K.The speech command recognition system based on deep image classification is the key that would tremendously promise to revolutionize research and development by overcoming the communication barrier between human and machine or computer. We are all aware of challenges in identifying the voice command in noise and variability in speed, pitch, and projection. This paper has developed an efficient and highly accurate speech command recognition for smart and effective speech processing applications like modern telecommunication. In particular, a novel convolutional neural network (CNN) is presented that works with a one-second audio clip consisting of one specific word including ten speech commands and other words labeled as “unknown,” and model implementations were operated in the noisy environment. The CNNs are structurally fully developed in such a way to recognize the speech commands with the utilization of deep learning (DL) for image classification concepts. Thus, this research used the concept of DL for image classification to translate the problem of speech command recognition into the image domain. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.Item An Efficient Infectious Disease Detection in Plants Using Deep Learning(Springer Science and Business Media Deutschland GmbH, 2024) Sunil, C.K.; Jaidhar, C.D.Over the past decade, agriculture has suffered reduced productivity from climate change and improper water, fertilizer, and pesticide use, fueling plant diseases. Pathogens pose the main threat, impacting crop yield and quality. Early detection and targeted treatments are crucial to improve both yield and quality. To address this, we have carried out deep learning-based approaches and published ours works in conferences and journSal. Those works are briefly discussed in the paper as follows: (i) Empirical work on different plant datasets is conducted to analyze the hyperparameters of the neural network. (ii) The research minimizes misclassifications by leveraging an ensemble-based strategy with AlexNet, ResNet, and VGGNet across seven plant leaf image datasets. The complexity of plant disease diagnosis in diverse conditions is tackled through a hybrid deep learning strategy, exemplified in the cardamom plant disease detection approach. (iii) An innovative deep learning-based approach is introduced to precise plant disease detection, crucial in the face of similar symptoms and imbalanced data. The proposed Multilevel Feature Fusion Network (MFFN) incorporates adaptive attention mechanisms, enhancing robustness by considering diverse network features. (iv) With cardamom plant disease classification utilizing U2-Net for background removal and EfficientNetV2 for classification, the network excels the performance on images with complex background, with this generated benchmark dataset with a complex background. This research work produced good results by achieving 99% accuracy on the tomato plant and 98.28% accuracy on the cardamom leaf dataset. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.Item A Transfer Learning Approach for Diabetic Retinopathy Classification Using Deep Convolutional Neural Networks(Institute of Electrical and Electronics Engineers Inc., 2018) Krishnan, A.S.; Clive, D.R.; Bhat, V.; Ramteke, P.B.; Koolagudi, S.G.Diabetic Retinopathy is a disease in which the retina is damaged due to diabetes mellitus. It is a leading cause for blindness today. Detection and quantification of such mellitus from retinal images is tedious and requires expertise. In this paper, an automatic identification of severity of Diabetic Retinopathy using Convolutional Neural Networks (CNNs) with a transfer learning approach has been proposed to aid the diagnostic process. A comparison of different CNN architectures such as ResNet, Inception-ResNet-v2 etc. is done using the quadratic weighted kappa metric. The qualitative and quantitative evaluation of the proposed approach is carried out on the Diabetic Retinopathy detection dataset from Kaggle. From the results, we observe that the proposed model achieves a kappa score of 0.76. © 2018 IEEE.Item Stock price movements classification using machine and deep learning techniques-the case study of indian stock market(Springer Verlag service@springer.de, 2019) Naik, N.; Mohan, B.R.Stock price movements forecasting is an important topic for traders and stock analyst. Timely prediction in stock yields can get more profits and returns. The predicting stock price movement on a daily basis is a difficult task due to more ups and down in the financial market. Therefore, there is a need for a more powerful predictive model to predict the stock prices. Most of the existing work is based on machine learning techniques and considered very few technical indicators to predict the stock prices. In this paper, we have extracted 33 technical indicators based on daily stock price such as open, high, low and close price. This paper addresses the two problems, first is the technical indicator feature selection and identification of the relevant technical indicators by using Boruta feature selection technique. The second is an accurate prediction model for stock price movements. To predict stock price movements we have proposed machine learning techniques and deep learning based model. The performance of the deep learning model is better than the machine learning techniques. The experimental results are significant improves the classification accuracy rate by 5% to 6%. National Stock Exchange, India (NSE) stocks are considered for the experiment. © Springer Nature Switzerland AG 2019.Item TAGS: Towards Automated Classification of Unstructured Clinical Nursing Notes(Springer Verlag service@springer.de, 2019) Gangavarapu, T.; Jayasimha, A.; S. Krishnan, G.S.; Kamath S․, S.K.Accurate risk management and disease prediction are vital in intensive care units to channel prompt care to patients in critical conditions and aid medical personnel in effective decision making. Clinical nursing notes document subjective assessments and crucial information of a patient’s state, which is mostly lost when transcribed into Electronic Medical Records (EMRs). The Clinical Decision Support Systems (CDSSs) in the existing body of literature are heavily dependent on the structured nature of EMRs. Moreover, works which aim at benchmarking deep learning models are limited. In this paper, we aim at leveraging the underutilized treasure-trove of patient-specific information present in the unstructured clinical nursing notes towards the development of CDSSs. We present a fuzzy token-based similarity approach to aggregate voluminous clinical documentations of a patient. To structure the free-text in the unstructured notes, vector space and coherence-based topic modeling approaches that capture the syntactic and latent semantic information are presented. Furthermore, we utilize the predictive capabilities of deep neural architectures for disease prediction as ICD-9 code group. Experimental validation revealed that the proposed Term weighting of nursing notes AGgregated using Similarity (TAGS) model outperformed the state-of-the-art model by 5% in AUPRC and 1.55% in AUROC. © 2019, Springer Nature Switzerland AG.Item Audio Replay Attack Detection for Speaker Verification System Using Convolutional Neural Networks(Springer, 2019) Kemanth, P.J.; Supanekar, S.; Koolagudi, G.K.An audio replay attack is one of the most popular spoofing attacks on speaker verification systems because it is very economical and does not require much knowledge of signal processing. In this paper, we investigate the significance of non-voiced audio segments and deep learning models like Convolutional Neural Networks (CNN) for audio replay attack detection. The non-voiced segments of the audio can be used to detect reverberation and channel noise. FFT spectrograms are generated and given as input to CNN to classify the audio as genuine or replay. The advantage of the proposed approach is, because of the removal of the voiced speech, the feature vector size is reduced without compromising the necessary features. This leads to significant amount of reduction on training time of the networks. The ASVspoof 2017 dataset is used to train and evaluate the model. The Equal Error Rate (EER) is computed and used as a metric to evaluate model performance. The proposed system has achieved an EER of 5.62% on the development dataset and 12.47% on the evaluation dataset. © 2019, Springer Nature Switzerland AG.Item Performance evaluation of deep learning frameworks on computer vision problems(Institute of Electrical and Electronics Engineers Inc., 2019) Nara, M.; Mukesh, B.R.; Padala, P.; Kinnal, B.Deep Learning (DL) applications have skyrocketed in recent years and are being applied in various domains. There has been a tremendous surge in the development of DL frameworks to make implementation easier. In this paper, we aim to make a comparative study of GPU-accelerated deep learning software frameworks such as Torch and TenserFlow (with Keras API). We attempt to benchmark the performance of these frameworks by implementing three different neural networks, each designed for a popular Computer Vision problem (MNIST, CIFAR10, Fashion MNIST). We performed this experiment on both CPU and GPU(Nvidia GeForce GTX 960M) settings. The performance metrics used here include evaluation time, training time, and accuracy. This paper aims to act as a guide to selecting the most suitable framework for a particular problem. The special interest of the paper is to evaluate the performance lost due to the utility of an API like Keras and a comparative study of the performance over a user-defined neural network and a standard network. Our interest also lies in their performance when subjected to networks of different sizes. ©2019 IEEE.Item Depthwise Separable Convolutional Neural Network Model for Intra-Retinal Cyst Segmentation(Institute of Electrical and Electronics Engineers Inc., 2019) Girish, G.N.; Saikumar, B.; Roychowdhury, S.; Kothari, A.R.; Rajan, J.Intra-retinal cysts (IRCs) are significant in detecting several ocular and retinal pathologies. Segmentation and quantification of IRCs from optical coherence tomography (OCT) scans is a challenging task due to present of speckle noise and scan intensity variations across the vendors. This work proposes a convolutional neural network (CNN) model with an encoder-decoder pair architecture for IRC segmentation across different cross-vendor OCT scans. Since deep CNN models have high computational complexity due to a large number of parameters, the proposed method of depthwise separable convolutional filters aids model generalizability and prevents model over-fitting. Also, the swish activation function is employed to prevent the vanishing gradient problem. The optima cyst segmentation challenge (OCSC) dataset with four different vendor OCT device scans is used to evaluate the proposed model. Our model achieves a mean Dice score of 0.74 and mean recall/precision rate of 0.72/0.82 across different imaging vendors and it outperforms existing algorithms on the OCSC dataset. © 2019 IEEE.