https://elibrary.khec.edu.np/handle/123456789/195 2026-07-17T04:31:48Z 2026-07-17T04:31:48Z Bhandari, Basant Babu Dhakal, Aakash Raj Maharjan, Laxman Karki, Asmin https://elibrary.khec.edu.np/handle/123456789/328 2024-08-10T10:38:06Z 2021-09-01T00:00:00Z

Title: Nepali Handwritten Letter Generation using GAN Authors: Bhandari, Basant Babu; Dhakal, Aakash Raj; Maharjan, Laxman; Karki, Asmin Abstract: The generative adversarial networks seem to work very effectively for training generative deep neural networks. The aim is to generate Nepali Handwritten letters using adversarial training in raster image format. Deep Convolutional generative network is used to generate Nepali handwritten letters. Proposed generative adversarial model that works on Devanagari 36 classes, each having 10,000 images, generates the Nepali Handwritten Letters that are similar to the real-life data-set of total size 360,000 images. The generated letters are obtained by simultaneously training the generator and discriminator of the network. Constructed discriminator networks and generator networks both have five convolution layers and the activation function is chosen such that generator networks generate the image and discriminator networks check if the generated image is similar to a real-life image dataset. To measure the quantitative performance, Frechet Inception Distance (FID) methodology is used. The FID value of 18 random samples, generated by networks constructed, is 38413677.145. For a qualitative measure of the model let the reader judge the quality of the image generated by the generator trained model. The Nepali letters were generated by the adversarial network as required. The evaluation helps the generative model to be better and further enables a better generation that humans have not thought of. Description: https://doi.org/10.3126/jsce.v9i9.46303

2021-09-01T00:00:00Z Sukamani, Umesh https://elibrary.khec.edu.np/handle/123456789/327 2024-08-10T10:38:06Z 2021-09-01T00:00:00Z

Title: Risk Management in Construction Firms in Nepal Authors: Sukamani, Umesh Abstract: The main purpose of this study is to identify and focus on major factors that affect risk management in construction sites in developing countries like Nepal. Relative Importance Index (RII) analysis is carried out to rank major factors that affect risk management. Besides, ANOVA analysis is carried out to test the hypothesis. Findings show that there is no difference between small and large-scale project groups in their perception of the significance level of factors affecting Project Risk Management. Moreover, “a) payment delay b) project funding problem, and c) defective design” were top three major factors that affect risk management in developing countries like Nepal. The managerial level of construction firms mainly in developing countries, should focus on the top ten critical factors for better improvement of risk management no matter its size- be it is small- or large scale construction firms.

2021-09-01T00:00:00Z Ghemosu, Dinesh Joshi, Shashidhar Ram https://elibrary.khec.edu.np/handle/123456789/200 2024-08-10T10:40:20Z 2021-01-01T00:00:00Z

Title: DETECTION AND CLASSIFICATION OF MRI-BASED BRAIN TUMOR VIA JAYA ALGORITHM AND TWIN SUPPORT VECTOR MACHINE Authors: Ghemosu, Dinesh; Joshi, Shashidhar Ram Abstract: Brain tumor detection and classification is one of the challenging tasks in the medical image application. Early detection of a brain tumor can help diagnosis and treatment of the patients. Magnetic Resonance Imaging (MRI) is widely used for the detection of brain tumor. Manual analysis of brain MRI, and classification of brain tumor is tedious and time-consuming job. This paper introduces the new novel approach of brain tumor segmentation and classification using BRATS 2015 datasets. Our system exploits the benefits of Jaya Algorithm (JA) as optimization technique for finding multi-level thresholds to segment the tumor part from the MRI. Feature extraction is implemented by Gray Level Co-occurrence Matrix (GLCM), followed by Principal Component Analysis (PCA) for feature reduction. Due to its inherent distinct feature and advantages, a machine-learning approach, Twin Support Vector Machine (TSVM) is used as a classifier. The prediction accuracy of proposed system yielded up to 97.89 % with sensitivity 96.48%, 98.97 precision, 97.91% F1 Score and 0.0798 MSE. The accuracy, sensitivity, F1 Score and MSE are found comparable to the other state-of-arts machine learning methods.

2021-01-01T00:00:00Z Ghimire, Santosh Mishra, Bimala https://elibrary.khec.edu.np/handle/123456789/199 2024-08-10T10:40:19Z 2021-01-01T00:00:00Z

Title: A REVIEW ON THE LEBESGUE SPACES Authors: Ghimire, Santosh; Mishra, Bimala Abstract: In this article, we begin with classical Lebesgue spaces Lp with p being constant and review the various properties such as completeness and duality of the space. To this end, we also discuss the boundedness of Hardy-Littlewood maximal function and interpolation on such spaces. Finally, we focus our attention on variable exponent Lebesgue spaces and review various results on it. Moreover, we also see the differences in between these Lebesgue spaces.

2021-01-01T00:00:00Z