https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/1776 Wed, 29 Apr 2026 15:50:42 GMT 2026-04-29T15:50:42Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/1815 Title: A Federated Approach on Heterogeneous NoSQL Databases Authors: Dharmasiri, H.M.L. Abstract: NoSQL, initially an industry buzz word and now it has become a topic of major interest. NoSQL addresses the problems of scalability and performance of the traditional RDBMS (Relational Database Management System) when it come to high volumes of data. Today there are over 100 NoSQL implementations mainly focused around four data models with a lot of heterogeneity between them. Therefore integrating several NoSQL data stores to work together in situations such as changing the underlying data store becomes a hectic task and migrating the data from one system to another is also not feasible due to the high volumes of data involved. This research suggests a database federation approach to cater these problems. Database federation has been around since the early 80 s which has proven to be successful in integrating heterogeneous data storages. The objective of this research is to address the heterogeneity of NoSQL data stores with database federation. In this research we have successfully implemented a NoSQL federation with Cassandra, MongoDB and CouchDB proving that NoSQL federation is feasible with a certain degree of overhead. Fri, 01 Jan 0012 00:00:00 GMT https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/1815 0012-01-01T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/1809 Title: Simulating Gait and Postural Effects of Aging for Improved Diversity in Virtual Crowds Authors: Gunaratne, C. Abstract: Autonomous character crowds have been effectively deployed in a variety of application domains including video games, films, psychology and operational research. Diversification of crowd characters has been proved to be a great contributor to realism. Diversification of individuals within real human crowds can be attributed to many factors such as gender, age, skin color, clothing, walking style, level of aggression etc., while most crowd simulation systems prefer to use clothing, skin color and other appearance related variables to make characters seem less similar to each other. This research aims to explore the possibility of increasing crowd diversity by simulating the biomechanical effects of aging in real humans within characters. Support vector regression has been used to train an aging parameter predictor on data from gerontology research recording spinal curvature and walking pattern deterioration caused by progression of age. The simulated characters have been reprogrammed to be able to predict their behavior through this predictor and reflect the results within the simulation. A spatial diversity algorithm has been proposed and implemented to distribute the resulting variants evenly among each other. Finally, three user evaluation experiments have been conducted to gauge the perceptibility and accuracy of simulating these biomechanical changes, to users and to evaluate the impact caused by biomechanical variety on the diversification of the crowd simulation as a whole. The results of these experiments prove that simulating posture and walking pattern deterioration in older characters does have a significant improvement when simulating old characters and that having this biomechanical diversity within a virtual human crowd does provide a considerable improvement in crowd diversity. Fri, 01 Jan 0012 00:00:00 GMT https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/1809 0012-01-01T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/1812 Title: Musical Genre Classification Using Ensemble of Classifiers Authors: Chathuranga, Y.M.D. Abstract: The increase of the music databases on the personal collection and the Internet have brought a great demand for music information retrieval, and especially automatic musical genre classi cation. Most automatic music genre classi cation researches have been focusing on combining information from di erent sources than the musical signal. This thesis presents a novel ensemble approach for the automatic music genre classi cation problem using low-level characteristics of audio signals into high-level hierarchically organized genre taxonomies. The proposed approach uses two feature vectors, Support vector machine classi- er with polynomial kernel function and a pattern recognition ensemble approach. The short term low-level audio features are derived from 30ms audio signal frames with a hop-size of 10ms and all the long term low-level audio features have a frame size of 10 seconds. More speci cally, two types of feature vectors for representing frequency domain, temporal domain and cepstral domain short term based audio features and modulation frequency domain long term based audio features are proposed for individual classi cation. For feature selection purposes, we used wrapper method for short term based feature vector and ltering method for long term based feature vector. The support vector machine classi er with polynomial kernel function is employed as the base classi ers for each individual feature vectors. Using our proposed features SVM act as a strong base learner in AdaBoost, so its performance of the SVM classi er cannot improve using boosting methods. The nal genre classi cation is obtained from the set of individual results according to a weighting combination late fusion method and it outperformed the trained fusion method. Music genre classi cation accuracy of 78% and 81% is reported on the GTZAN dataset over the ten musical genres and the ISMIR2004 genre dataset over the six musical genres, respectively. We observed higher classi cation accuracies with the ensembles, than with the individual classi ers and improvements of the performances on the GTZAN genre dataset are three percent on average. This ensemble approach shows that it is possible to improve the classi cation accuracy by using di erent types of domain based audio features. Fri, 01 Jan 0012 00:00:00 GMT https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/1812 0012-01-01T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/1810 Title: English to Sinhala Speech-to-Speech Phrase Translation for Mobile Applications Authors: Bulathsinhala, C.L. Abstract: English to Sinhala translation has been researched and experimented by many in the past few years. A fully functioning translating methodology for English to Sinhala translation has not yet been suggested due to many problems which have to be overcome. This paper describes a English-to-Sinhala speech-to-speech translator for mobiles. The study has revolved mainly around the tourist domain since this is where the language barrier comes into play almost all the time. The proposed system is broken down into three components which are speech recognition, translation and text-to-speech component. In this paper we hope to go in to detail on these three components designs, approaches and present their implementation details. The work proposed can be considered as a step forward in filling this gap. Fri, 01 Jan 0012 00:00:00 GMT https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/1810 0012-01-01T00:00:00Z