Marpe Sora

@rgu.ac.in

Associate Professor & Head, Department of Computer Science & Engineering
RAJIV GANDHI UNIVERSITY ARUNACHAL PRADESH

Marpe Sora


            Download Compact PDF  
https://researchid.co/marpesora

EDUCATION

PhD in Computer Science and Engineering

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Engineering, Artificial Intelligence, Multidisciplinary, Computer Vision and Pattern Recognition
11

Scopus Publications

Scopus Publications

  • An efficient algorithm for the removal of motion artifacts in wearable ECG technology
    Jagdeep Rahul, Marpe Sora
    Iran Journal of Computer Science, 2025
  • Unsupervised learning for financial statement fraud detection using manta ray foraging based convolutional neural network
    Ajit Kumar Singh Yadav, Marpe Sora
    Concurrency and Computation Practice and Experience, 2022
    SummaryFinancial statement fraud detection is a significant topic and a challenging task related to countless applications such as financial sectors, insurance, government agencies, law enforcement and so on. This article presents an unsupervised learning model to identify the financial fraud statements using the hybrid convolutional neural network. The fraud detection scheme can be performed using a few stages: data preprocessing, sampling, feature extraction, feature selection, clustering, and classification methods. The preprocessing and sampling processes are used to clean and recognize duplicate data. Then, the proposed model analyzes both text features and financial variables to provide the best classification of financial statement fraud. In this article, a new fuzzy red deer's algorithm is proposed for selecting the optimal feature set, which can improve the classification accuracy of the fraud prediction model. Subsequently, an adaptive density based clustering (ADBC) approach is introduced for labeling the selected features through the clustering process. Finally, the fraud statements are predicted by proposing a hybrid CNN‐MRFO model wherein the hyperparameters of the CNN model are optimized through the MRFO learning algorithm. The simulation results prove the effectiveness of the proposed model in terms of accuracy (98.85%), precision (98.62%) and F‐measure (99.32%).
  • A novel and lightweight P, QRS, and T peaks detector using adaptive thresholding and template waveform
    Jagdeep Rahul, Marpe Sora, Lakhan Dev Sharma
    Computers in Biology and Medicine, 2021
  • Dynamic thresholding based efficient QRS complex detection with low computational overhead
    Jagdeep Rahul, Marpe Sora, Lakhan Dev Sharma
    Biomedical Signal Processing and Control, 2021
  • An improved cardiac arrhythmia classification using an RR interval-based approach
    Jagdeep Rahul, Marpe Sora, Lakhan Dev Sharma, Vijay Kumar Bohat
    Biocybernetics and Biomedical Engineering, 2021
  • Fraud detection in financial statements using text mining methods: A review
    Ajit Kr. Singh Yadav, Marpe Sora
    Iop Conference Series Materials Science and Engineering, 2021
    In the financial industry, financial fraud is one of the ever-growing hazards with far concerns. Financial statements are the fundamental papers which replicate economic position of a corporation. Users of the financial information like public, creditors etc. are the major foundations of a decision-making process for financing stakeholders. Financial fraud has extremely damaged the sustainable growth of financial markets and enterprises. The amount of financial reporting fraud cases keeps on developing. Each incident is a thick hit to partners, banks, and financial specialists and it costs human progress significantly. One of the serious issues is to recognize the financial reporting fraud by utilizing formation of an active model. The aim of this paper is to identifying frauds using various text mining techniques and guard, the public’s investments. This investigation will benefit auditors and financial governors.
  • Premature Ventricular Contractions Classification using Machine Learning Approach
    Jagdeep Rahul, Marpe Sora
    Proceedings International Conference on Smart Electronics and Communication Icosec 2020, 2020
    In this paper, Premature Ventricular Contractions [PVCs] beat classification is proposed for detecting the ventricular arrhythmia. ECG arrhythmia records are considered from MITBIH AD and denoised by using the discrete wavelet transform (DWT). Thereafter, two stage median filter is used to eliminate the baseline wander to obtain the clean and smooth ECG signal. Proposed method has calculated the statistical features of extracted QRS complex of both PVCs and normal beats. KNN and SVM algorithms are used for performance evaluation of the proposed method. Overall SVM algorithm using Gaussian function with kernel scale =0.56 achieved the Sp=99.71 %, $S_{\\mathrm{e}}$=99.80 %, +P=99.71 % and Acc=99.75%. The results obtained have shown that the PVCs classification method is more accurate and reliable, and can be used for automatic classification of arrhythmia.
  • Exploratory data analysis based efficient QRS-complex detection technique with minimal computational load
    Jagdeep Rahul, Marpe Sora, Lakhan Dev Sharma
    Physical and Engineering Sciences in Medicine, 2020
  • A novel adaptive window based technique for T wave detection and delineation in the ECG
    Jagdeep Rahul, Marpe Sora
    Bio Algorithms and Med Systems, 2020
    The electrocardiogram (ECG) morphology determines the overall activity of the heart and is the most widely used tool in the diagnostic processes. T wave is a crucial wave component that reveals very useful information regarding various cardiac disorders. In this paper we have proposed a novel T wave detection technique based on adaptive window and simple decision rule. The proposed technique uses two-stage median filters followed by the Savitzky-Golay filter at the pre-processing stage to remove the noises in the ECG signal. The QRS complex is detected for locating the T wave as a reference in one ECG cycle. An R-R interval based window is considered for detecting the T wave, and decision logic depends on the iso-electric line value. The proposed technique is tested on the QT database and self-recorded dataset for its performance evaluation. In the present work, the results achieved for T wave detection sensitivity (Se), positive predictivity (+P), detection error rate (DER), and accuracy (Acc) on the QT database are Se = 97.57%, +P = 99.63%, DER = 2.78%, and Acc = 97.22% with an average time error of (3.468 ± 5.732) ms. The proposed technique shows Se = 99.94%, +P = 99.94%, DER = 0.01%, and Acc = 99.89% on the self-recorded dataset. The proposed technique is also capable of detecting both the upward and downward T wave efficiently in the ECG signal.
  • Baseline correction of ECG using regression estimation method
    Jagdeep Rahul, Marpe Sora, LakhanDev Sharma
    Proceedings 2019 4th International Conference on Internet of Things Smart Innovation and Usages Iot Siu 2019, 2019
    The presence of baseline wander in ECG signal severely affects the quality of ECG signal. Baseline wander is generally eliminated at the initial stage of preprocessing of ECG signal. In this paper a method using regression estimation is applied to ECG signal to remove the baseline wander. This method uses two stage median filter and regression estimation for smoothing and correction of baseline. The cross correlation is performed between the baseline wander containing ECG signal with corrected ECG signal. Then cross-correlation coefficient is calculated. The method evolves with the lowess and loess of local regression estimation, rlowess and rloess of robust local regression estimation. We found highest coefficient of 0.9914 for the baseline wander amplitude of 0.5mV using local regression method (lowess) and lowest coefficient of 0.9688 for the amplitude of 0.8mV using robust local regression method (rlowess). Further this method is validated on MIT-BIH, Noise Stress Test Database (NSTD).
  • An overview on biomedical signal analysis
    International Journal of Recent Technology and Engineering, 2019