SLTC e-Repository
The SLTC e-Repository is a digital platform designed to store, manage, and provide access to academic and research outputs of the Sri Lanka Technological Campus. It serves as a central hub for theses, dissertations, research papers, and other scholarly content produced by students and faculty.
Communities in DSpace
Select a community to browse its collections.
Recent Submissions
Sentiment Analysis for Amazon Reviews to Identify Customer Interests.
(Sri Lanka Technology Campus, 2024-11-06) Abeysirigunawardana,K. Aloka; Rajapaksha,Dr. Rasika
In the twenty-first century, consumers now
frequently buy online and use reviews to judge the quality
of the products they purchase. These reviews are also
examined by businesses to enhance their offerings. In
order to glean business insights from massive datasets,
this study uses sentiment analysis on Amazon product
reviews. To determine the specific subjects upon which
the entire collection of reviews is predicated, an LDA
model is created for the Amazon review dataset. Word
frequencies for each topic are visualised by us. After
determining which machine learning model is most
appropriate for sentiment analysis, the analysis is carried
out on a topic-by-topic basis. Using logistic regression,
the themes derived from the LDA model are categorised
as either positive or negative product review subjects.
Analysis is done on both the positive and negative
subjects.
DreamyMirror A Machine Learning-Based Personalized Outfit Recommendation System Integrating Skin Tone Classification and Size
(Sri Lanka Technology Campus, 2024-11-06) Serasinghe,S. J. A. S. K.; Jayarathne,P. V. C. G.
Due to the increasing popularity of e-commerce, buying
has undergone a tremendous shift and belongs to a
different world. ” But there is still some difficulties with
personalization when it comes to the proposed size and
the choice of the color taking into consideration the skin
color of the person. In this paper, the authors introduce
DreamyMirror, a novel system for recommending outfits
based on the individual user’s skin tone and size, along
with allowing for virtual fitting. Based on a skin tone
classifier CNN model built from CelebA database and a
pre-trained Pose_iter_440000 of OpenPose. size
prediction caffemodel, DreamyMirror provides unique
suggestions of outfits according to the appearance of the
person. The system also contains the Virtual Try on
system wherein users can anticipate what the selected
apparels look like at a glance. The proposed system can
automatically recognize skin tone with an overall
classification accuracy of 92% and the size prediction
accuracy is found to be 95%. Moreover, the user
engagement metrics also show an 85% level of
satisfaction with the features – this is about the virtual
try-on. This work shows how the application of AI in
fashion retail can help better engage the customer
through a truly individualized shopping experience and,
at the same time, minimize the return rate because of the
wrong size or color choice.
Intelligent Road Safety Navigation in Sri Lanka: A Review of Machine Learning Techniques and Proposal of a Model for Predicting Accident Hotspots and Severity
(Sri Lanka Technology Campus, 2024-11-06) Jayamanna,JMCT; Kalansooriya,Pradeep
Both public safety and the stability of the economy
are seriously threatened by traffic accidents. Like many other
regions, Sri Lanka is faced with the challenge of road accidents,
which hinder the path to safer roads. One of the reasons that
accidents occur is that people are unaware of common accident
locations. The government has already enforced other tactics,
like traffic signals and fines, to reduce these incidents, but they
have been ineffective. In order to decrease road accidents, people
must change their driving patterns. While looking for a solution
to that problem, existing studies around the world have proposed
predictive machine learning models for accident-prone locations
known as hotspots and severity levels. But there were not any
existing studies that proposed a solution that was suitable for Sri
Lanka. This research seeks to address these challenges by
conducting an in-depth review of existing machine learning
techniques and proposing the most suitable model approaches
for prediction of accident hotspots and severity in Sri Lanka
based on the availability of the accident data. One of the main
objectives is to identify the correct machine learning techniques.
According to studies, 81% used the 'Random Forest' algorithm,
which is a supervised machine learning algorithm for the
prediction. And Random Forest performed better in
approximately 69% of the studies. And this research is not just
proposing suitable model approaches for predictions. It provides
the foundation to revolutionize road safety through the
development of an intelligent road safety mobile navigation
application
THE IMPACT OF DIFFERENT CORPORATE EVENT ANNOUNCEMENTS ON SHORT-RUN STOCK RETURNS; EVIDENCE FROM COLOMBO STOCK EXCHANGE
(Sri Lanka Technology Campus, 2024-11-06) Abeyrathna,Lakshmika.; Sudasinghe,Sandali.
Publicly traded markets function as the base of our
economic system. Efficiently valued financial markets power
up national development according to Malkiel (2010). Market
effectiveness shows precise security assessment results. When
information travels rapidly into stock prices in an efficient
market it won't help investors earn more than the market or
find ways to get abnormal returns from what they already
know.
This paper examines the efficiency of the market in relation
to four prominent corporate events: We examine the
information event impact of stock dividend announcements
alongside bonus issues, rights issues, and stock splits under Sri
Lanka's conditions. A sample of 13 stock dividend and 9 bonus
issue releases formed the base data alongside 31 rights issue
and 29 stock split announcements. This research uses market
model event studies to track price reactions through Average
Abnormal Returns and Cumulative Average Abnormal
Returns calculations. The research discoveries question what
current market efficiency stories tell us.
Investors in Sri Lanka currently lack complete data about
stock price reactions to corporate news. Our study examines
market reactions to corporate event announcements in CSE
using 82 unique events across 14 economic zones between 2019
and 2021.
A Novel Factorization Method Using Continued Fractions
(Sri Lanka Technology Campus, 2024-11-06) Vinodya,Malshi; Ranasinghe,Rajitha
The study of continued fractions is a significant area
of mathematics with diverse applications, particularly in the
field of factorization. Continued fractions can be used to
approximate irrational numbers and are integral to algorithms
for factoring integers. In this study, we present a novel method
for factoring large integers that utilize generalized continued
fractions to improve efficient factorization. Additionally, we
introduce several theoretical statements about generalized
continued fractions and demonstrate their application within
the proposed factorization algorithm. Using this algorithm, we
successfully factor a large integer into two prime numbers,
whose product constitutes the original large number. Our
findings suggest that this method is a highly effective tool in
number theory, cryptography, and computational mathematics.