Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Filters

2022 - 20244

Settings

Subject: search.filters.subject.BiLSTM

Search Results

Now showing 1 - 4 of 4
  • No Thumbnail Available
    Item
    Handwritten Text Recognition from an Image with Android Application
    (Institute of Electrical and Electronics Engineers Inc., 2022) Mule, H.; Kadam, N.; Naik, D.
    Nowadays, Storing information from handwritten documents for future use is becoming necessary. An easy way to store information is to capture handwritten documents and save them in image format. Recognizing the text or characters present in the image is called Optical Character Recognition. Text extraction from the image in the recent research is challenging due to stroke variation, inconsistent writing style, Cursive handwriting, etc. We have proposed CNN and BiLSTM models for text recognition in this work. This model is evaluated on the IAM dataset and achieved 92% character recognition accuracy. This model is deployed to the Firebase as a custom model to increase usability. We have developed an android application that will allow the user to capture or browse the image and extract the text from the picture by calling the firebase model and saving text in the file. To store the text file user can browse for the appropriate location. The proposed model works on both printed and handwritten text. © 2022 IEEE.
  • No Thumbnail Available
    Item
    Video to Text Generation Using Sentence Vector and Skip Connections
    (Springer, 2023) Mule, H.; Naik, D.
    Nowadays, video data is increasing rapidly and the need of robust algorithms to process the interpretation of the video. A textual alternative will be more effective and save time. We aim to produce the caption for the video. The most famous architecture used for this is the encoder-decoder (E-D) model. Recent attempts have focused on improving performance by including 3D-CNN, transformers, or structural changes in the basic LSTM units used in E-D. Sentence vectors are used in this work, improving the E-D model’s performance. From the video file, a sentence vector is generated and used by the decoder to generate an accurate description by using previously generated words. Skip connection in the encoder part avoids the vanishing gradients problem. All of our studies use the MSVD and CHARADES datasets. Four famous metrics, BLEU@4, METEOR, ROUGE, and CIDER, are used for performance evaluation. We have compared the performance of BERT, ELMo, and GloVe word embeddings. On experimental analysis, BERT embedding outperformed the ELMo and GloVe embeddings. For feature extraction, pretrained CNNs, NASNet-Large, VGG-16, Inception-v4, and Resnet152 are used, and NASNet-Large outperformed other models. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
  • No Thumbnail Available
    Item
    An Efficient Rainfall Prediction Model Using Deep Learning Method
    (Institute of Electrical and Electronics Engineers Inc., 2023) Verma, V.K.; Janagama, H.S.; Patil, N.
    Rainfall is a crucial aspect of the Earth's natural cycle and it is necessary for various activities such as agriculture, water supply and hydroelectric power generation. However excessive rainfall can lead to floods, landslides and other destructive consequences, while insufficient rainfall can cause droughts and water shortages. Therefore accurate estimation of rainfall is essential to manage and mitigate the impacts of rainfall. In this study, the dataset is collected from the NASA Power database [22] to predict the annual rainfall in Mangalore(Karnataka), India. The data is collected from January 1, 2003 to February 04, 2023 using NASA POWER API. The study used four models MLP[15], LSTM, BiLSTM, CNN to predict the daily average precipitation that contributes to the annual rainfall. The input parameters considered for the prediction are maximum monthly temperature, minimum monthly temperature, humidity, atmospheric pressure and wind speed[9]. The model's performance is measured using mean squared error (MSE) and mean absolute error (MAE) of the predicted values on training and testing ratio 80:20. CNN(Convolutional Neural Network) model outperforms and gives the MSE and MAE for the CNN(Convolutional Neural Network) model are 0.0041 and 0.0456 respectively. © 2023 IEEE.
  • No Thumbnail Available
    Item
    Detecting Fake News: A Comparative Evaluation of Machine Learning Techniques
    (Institute of Electrical and Electronics Engineers Inc., 2024) Aishwarya, C.; Venkatesan, M.; Prabhavathy, P.; Shetty, A.S.
    Fake news is a significant and well-acknowledged problem in contemporary society due to its rapid spread via social media and various online networking platforms, thereby making it difficult to determine the validity of information. In this study, we examine literature for this issue, prevalent datasets like LIAR, Politifact, and COVID-19, as well as classical machine learning and deep learning models such as SVM, BiLSTM, and CNN- BiGRU for fake news detection, and analyze their effectiveness and scope of application for fake news detection. © 2024 IEEE.