NSPACE

Welcome to NSBM Green University e-Repository!

One Platform for NSBM Publications

Photo by @inspiredimages

Communities in DSpace

Select a community to browse its collections.

Now showing 1 - 5 of 5

Recent Submissions

Swimming Stroke Analysis and Feedback System using Machine Learning
(NSBM Green University, 2024-10-24) Jayawardene, A.; Kalansooriya, P.
Swimming is a sport which offers numerous health benefits which however improper and poor techniques can affect the swimmer’s health as well as their performance by leading to significant injuries particularly affecting knees, back and shoulders. The proposed system utilizes machine learning and computer vision techniques to initially analyse freestyle swimming stroke videos and categorizes swimming postures as “good” and “bad” poses and provide feedback to swimmers using a classification machine learning model. The development process includes the segmentation of swimmer images, pose estimation and feature extraction and the training the machine learning model. The model is trained on a large set of swimming footage with annotations. Pose estimation accuracy is increased by the segmentation procedure, which makes sure that only relevant body movements are examined. Once trained, the machine learning model uses a user-friendly web application to recognise poses and provide corrective feedback. The expected outcomes of the proposed system are accurate pose estimation with a mean absolute error of less than 5 pixels and the machine learning model achieving over 90% accuracy in pose classification. Variable video quality and the requirement for model changes throughout time are the challenges of the proposed system as future research will be focused on evaluating for the other strokes and offering more detailed feedback thorough analysis.
A Comprehensive and Comparative Analysis of Parallel Frameworks and their Applications to Big Data Clustering
(NSBM Green University, 2024-10-24) Nawfal; Srivastava
The K-Means clustering using Lloyd’s algorithm is an iterative approach to partition the given dataset into K different clusters. The serial algorithm involves iterative steps where we compute the distance of each datapoint from the centroids and assign the datapoint to the nearest centroid. This approach is essentially known as the expectation-maximization step. Clustering involves extensive computations to calculate distances at each iteration, which increases as the number of data points increases. This provides scope for parallelism. However, we must ensure that in a parallel process, each thread has access to the updated centroid value, and no racing condition exists on any centroid values. In this paper, we will compare two different approaches. The first approach is an OpenMP flat synchronous method where all processes are run in parallel, and we use synchronization to ensure safe updates of clusters. The second approach we adopt is a GPU- based parallelization approach using OpenACC, wherein we will try to make use of GPU architecture to parallelize chunks of the algorithm to observe decreased computation time. We will analyze metrics such as speed up, efficiency, time taken with varying data points, and number of processes to compare the two approaches and understand the relative performance improvement we can get.
Predictive Modeling of Tourist Arrivals in Sri Lanka Using Linear Regression
(NSBM Green University, 2024-10-24) Somasiri, N.; Karunarathne, L.; Ganesan, S.
One of the sectors of a developing nation's economy that directly generates income is tourism. For this reason, predicting the number of visitors is crucial when deciding on policies to upgrade facilities and other relevant aspects of this sector. The data for this paper were gathered from the Corporate website of the Sri Lanka tourism development authority(SLTDA)This paper attempts to forecast tourist arrivals in Sri Lanka using Linear Regression Model. The time span used for this study is from January 2021 to March 2024. The performance of the model was evaluated using metrics such as Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) and R2 score. This paper forecasts tourist arrivals in Sri Lanka using the Linear Regression Model. While Linear Regression is suitable due to its simplicity and the scope of the dataset, more advanced forecasting models such as ARIMA or deep learning approaches like LSTM could have been considered for a more robust comparison. The study uses data from January 2021 to March 2024, and the model's performance is evaluated using metrics such as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R² score. The findings suggest that machine learning models can improve the forecasting accuracy of visitor arrivals, providing valuable insights for stakeholders in the tourism industry.
Ethical Landscape of Artificial Intelligence: A Review
(NSBM Green University, 2024-10-24) Gunasekera, D.
Artificial intelligence (AI) offers great potential to enhance and improv many aspects of human life and society. However, to ensure its benefits are realized in a lawful, fair, and trustworthy manner, the ethical challenges arising from increasingly autonomous technologies must be addressed. This paper reviews the current state of AI ethics research and ongoing multidisciplinary discussions. Author identified obstacles inhibiting responsible progress and opportunities to foster innovation aligned with principles like transparency, fairness, privacy, and human rights. The literature review examined technical reports and peer-eviewed research on issues such as biases in training data that could lead to unfair outcomes, challenges in ensuring accountability and oversight of advanced autonomous systems, and the tensions between maximizing social benefits and mitigating risks to privacy, fairness, and other human rights. The author found that these are key obstacles to achieving trustworthy AI applications. Emphasizing data and algorithmic ransparency, inclusive governance through collaboration in technical and policy solutions, and user-centered approaches aligned with societal values offers a path towards safely and equitably realizing AI’s promise. This can be achieved through flexible yet robust frameworks for managing risks in dynamic human-machine systems.
Data Mining Based Study to Analyze Passenger Transportation in Uva Province
(NSBM Green University, 2024-10-24) Wijayakoon, W.B.M.S.C.; Kuthulwatta, W.M.C.J.T.; Samapth, J.K.H.; Wijesekara K.B.
Passenger transportation methods are a very important factor for sustainable development and sustainable supply chain management. According to the pandemic situation and to employ well established economic and social based development within the Uva Province, there is a need of identifying any data patterns, uncovered information, and obtaining other valuable information. The primary objective of this research is to uncover patterns, correlations, and trends that influence transportation operations. According to the several studies, data mining techniques were used for data analysis, pattern identification and decision making. Hence, this research analyses road passenger transportation within the Uva Province in Sri Lanka using data mining techniques. The dataset of 26,100 entries, managed by the Passenger Transport Service Authority of Uva Province from 1976 - 2022, was collected and explored to uncover patterns, correlations, and trends that influence transportation operations. This analysis reveals several significant findings. The findings contribute to evidence-based recommendations aligned with current transportation trends and market conditions. This paper underscores the significance of data mining in transforming raw data into actionable insights that facilitate informed decision-making in the transportation sector.