Lochana Hasith

I am an energetic team member and adaptable to challenging situations. I am a tech enthusiast who likes to share knowledge with others and always willing to learn new skills. I believe that the positive attitude is the key to success.

Email: lochanahm@gmail.com
Phone: 0775182055
Course: CS

Interests:Machine Learning | Data Analytics | Bioinformatics |HCI | Image Processing

Technical Skills: Java | Python | HTML | CSS | Javascript | Bootstrap |PHP |
Codeigniter | Angular | Node.JS |Laravel | Ext JS |Keras |MySQL | MongoDB

Project Experience

eBEYONDS Loyalty System (Internship Project)
Technologies used- Laravel 5.6, Ext JS, MySQL
News Together – Online News & Correspondent Managing System.
Technologies used- PHP, CodeIgniter Framework, HTML, CSS, Bootstrap, Javascript, MySQL
Hospital Management System
Technologies used – Java Swing, MySQL
Lab Reservation System for UCSC
Technologies used – MongoDB, Express.js, Angular 5, Node.js, HTML, CSS, Javascript, Bootstrap
Manudam Yathra – School volunteering System
Technologies used- PHP, CodeIgniter Framework, HTML, CSS, Bootstrap, Javascript,MySQL

Work Experience

• Software Engineer – Intern | eBEYONDS (Pvt) Ltd. (August 2018- February 2019)

Achievements | Awards

• MedHack (2017) organized by the Ministry of health, nutrition and indigenous medicine – Top 5 Products
• HackLN(2017) Organized by the Department of Computer Science and Statistics, University of Kelaniya – Finalists
• Hack it (2017) organized by the Garners Labs -Finalists
• Ihack 4.0 (2018) organized by the UCSC ISACA Student Group – Participation
• IEEEXtreme 10.0 (2016) – Team – bitHackersPluz Country rank – 80 Region rank – 202

Final Year Project

Dimensionality Reduction for Meta-analysis of Cancer Genomic Data using Deep Learning.

• Cancer is a disease in the genetic level. It cannot be treated through basic principles of medicine. In order to solve this issue, personalized medicine is very helpful. To make personalized medicine, we need to identify the subtypes of cancers. The high dimensionality of genomic data makes the subtypes identification task more difficult. The existing state of the art dimensionality reduction methods sometimes fails when applied to genomic data. In this research, we try to create a novel deep learning model which will reduce the high dimensionality of genomic data and help to make the cancer subtype identification task more accurate.