Embedded Files
Current projects
Current projects
Software Defect Prediction
Software Defect Prediction
A Novel Method for Software Defect prediction Using Bio-Inspired Algorithms
A Novel Method for Software Defect prediction Using Bio-Inspired Algorithms
Validation and verification of widely produced software products is a high priority in software engineering field. For this concern, efficiency of software defect prediction needs to be improved. This study focuses on using various bioinspired machine learning algorithms to predict defects.
Portfolio
Portfolio
Full stack Development
Full stack Development
VitalFlow: Streamlining Blood Bank Operations with Advanced Management
VitalFlow: Streamlining Blood Bank Operations with Advanced Management
During the Covid-19 pandemic, a concern regarding managing blood banks was raised due lack of donors and travel restrictions in many countries. Due to this, many patients' lives were at stake. A prototype of an online system to manage the donor details allowing the recipients and hospitals to reach in an emergency is proposed in this project.
NLP
NLP
Suicide Watch and Mental Health Detection Using NLP
Suicide Watch and Mental Health Detection Using NLP
Mental health deterioration is a global concern. It is proven that the risk of feeling suicidal correlates with mental health issues. Detection of such issues can be conducted based on textual contexts provided by the subjects. However the biggest issue of handling such medical data is the class imbalance. This project focuses on using ML, DL, LSTM and Transformer models and topic modeling to conduct multiclass text classification on mental health issues based on data extracted from subreddit posts.
Geospatial data analysis
Geospatial data analysis
Analysis of Sri Lankan Forest, Land sector coverage and Bio diversity Hotspots
Analysis of Sri Lankan Forest, Land sector coverage and Bio diversity Hotspots
World Heritage Sinharaja Rainforest in Sri Lanka is generating headlines about its forest cover. As a home to unique and endangered species, It is an excellent location for studying biological evolution. However, a data science based scientific analysis has yet to be conducted on issues regarding Sinharaja. The goal of this project is to visualize aspects of geospatial data in terms of biodiversity, climate, soil resources, and forest coverage of Sri lanka. Further improvements of this analysis can be used to make predictions based on trends in the geograpghy over the years.
Page updated
Google Sites
Report abuse