Sushila Sipai

Hi there, I'm Sushila

Welcome to my world! I'm Sushila Sipai , a CS PhD student at University of Texas at Dallas. I thrive on Natural Language Processing and Information Retrieval. Let's connect and create something amazing!

Linkedin iconGithub iconGoogle scholar iconGmail icon
Avatar image
Education
  • University of Texas at Dallas

    PhD in Computer Science (Intelligent Systems | HLTRI | Supervisor: Dr. Sanda Harabagiu) (2023-Current)


    Natural Language Processing, Machine Learning, Information Retrieval, Database Design

  • Kantipur Engineering College (Tribhuvan University)

    Bachelor in Computer Engineering


    Ranked: 1/120

Recent Projects
Project Web Design

Developed a CNN based model for translation of Nepali Sign language to English to reduce communication gap with deaf and/or mute people along with the publication of the dataset for public use.

Project Fraud
Fraud Transaction and Anomaly Detection
KNN
Random Forest
XGBoost
SVM
Bagging
Isolation Forest
Encoder Decoder

Implemented fraud detection and anomaly detection system using machine learning techniques on the IEEE-CIS Fraud Detection dataset from Kaggle, enhancing online transaction security.

Project Fire
Parkinson's Disease Data Analysis and Prediction Using Ensemble Machine Learning Techniques
Ensemble Learning
GBDT
Random Forest
Voting Classifier
Stacking Classifier

Built a model using ensemble learning techniques like GBDT, random forest, voting classifier, and stacking classifier to detect Parkinson's disease.

Project Fraud

Developed a string matching model using Fuzzy Logic in Python to generate compliance and audit reports to identify fraudulent transactions.

Recent Publications
Image post 2

Parkinson's disease data analysis and prediction using ensemble machine learning techniques

Jul 22, 2021
Mobile Computing and Sustainable Informatics- Proceedings of ICMCSI 2021
Image post

Two Dimensional (2D) Convolutional Neural Network for Nepali Sign Language Recognition

Dec 2, 2018
2018 12th International Conference on Software, Knowledge, Information Management & Applications (SKIMA)