DCIT Lecturers' Papers

Permanent URI for this collection

https://uwispace.sta.uwi.edu/handle/2139/5462

Papers, abstracts and publications authored by staff of the Department of Computing and Information Technology, UWI St. Augustine.

Browse

Now showing 1 - 20 of 30

Heuristic Constraint-Path Routing Decision System
(IEEE Computer Society, 2005-05-18) Sivakumar, Shyamala; Phillips, William J.; Robertson, William; Goodridge, Wayne
Heuristic QoS algorithms under strict constraints perform poorly in terms of finding a path that is suitable for a user?s QoS needs - the multiple constraint path problem (MCP). Exact QoS algorithms, on the other hand, guarantee that a path satisfying user needs would be found and o..er a more realistic approach for solving the MCP problem in view of the fact that the NP-complete character of graphs are not common in real networks. This fact has driven approaches like the SAMCRA and A*prune algorithms. However, these algorithms still have very high running times relative to heuristic approaches. When QoS routing algorithms are used in online Tra..c Engineering (TE) environments it may be necessary to route thousands of traffic flows each minute. Exact algorithms simply cannot work in such environments. We propose a heuristic algorithm that is suitable for working in an online TE environment. Simulations show that this algorithm produce high success rates in terms of finding suitable constraint paths for user flows while at the same time having execution times comparable to another heuristic based algorithms.
Automated design and drawing of graphs
(2008-11-21T15:08:19Z) Bernard, Margaret
Error-correcting coders with variable length and non-uniform errors
(2008-11-21T15:08:36Z) Bernard, Margaret
Better Global Polynomial Approximation for Image Rectification
(2009-11-10T15:41:40Z) Ward, Christopher
When using images to locate objects, there is the problem of correcting for distortion and misalignment in the images. An elegant way of solving this problem is to generate an error correcting function that maps points in an image to their corrected locations. We generate such a function by fitting a polynomial to a set of sample points. The objective is to identify a polynomial that passes "sufficiently close" to these points with "good" approximation of intermediate points. In the past, it has been difficult to achieve good global polynomial approximation using only sample points. We report on the development of a global polynomial approximation algorithm for solving this problem.
Emotion-Oriented eCommerce Systems
(2010-06) LEON, SIMONE; NIKOV, ALEXANDER
During recent years, the use of intelligent systems in eCommerce has increased significantly, providing a new perspective on the overall online shopping experience. The rapid growth of eCommerce has motivated many studies on the relationship between website design, company reputation, and purchase intent. Understanding emotion is important in gaining insight on how to efficiently satisfy the needs of the eCommerce customer. Emotion recognition techniques based on modern mathematical models using computational intelligence are presented. Progression within this field of study is highlighted. The requirements for designing of intelligent emotion-oriented eCommerce systems are defined. A model for simulation of intelligent emotion-oriented eCommerce systems is proposed. It is an important tool supporting the experimental study and design of emotion-oriented eCommerce systems
Integrating Two Artificial Intelligence Theories in a Medical Diagnosis Application
(2010-11-03T14:27:45Z) Peter, Hadrain; Goodridge, Wayne
Reasoning Systems (Inference Mechanisms) and Neural Networks are two major areas of Artificial Intelligence(AI). The use of case-based reasoning in Artificial Intelligence systems is well known. Similarly, the AI literature is replete with papers on neural networks. However, there is relatively little research in which the theories of case-based reasoning and neural networks are combined.In this paper we integrate the two theories and show how the resulting model is used in a medical diagnosis application.An implementation of our model provides a valuable prototype for medical experts and medical students alike.
A Study of the Effectiveness of the Routing Decision Support Algorithm
(2010-11-03T15:11:32Z) Sahai, Ashok; Nikov, Alexander; Goodridge, Wayne
Multi criteria decision making (MCDM) methods like analytic hierarchy process, ELECTRE and multi-attribute utility theory are critically studied. They have irregularities in terms of the reliability of ranking of the best alternatives. The Routing Decision Support (RDS) algorithm is trying to improve some of their deficiencies. This paper gives a mathematical verification that the RDS algorithm conforms to the test criteria for an effective MCDM method when a linear preference function is considered
The Delivery and Development of an online course using Open
(2010-11-10T15:49:01Z) Bernard, Margaret; Ramnanan, Anil
In September 2005, a Computer Science course was developed using almost exclusively a number of Open Source Learning Content Management Tools. This course was delivered twice over two semesters to students at St. George’s University, Grenada. The tools that were used to develop the course were the Burrokeet LCMS which was used to develop the content and ATutor which was used to deliver the course. Throughout the development of the course a number of observations were made as to the issues involved in developing a course for online delivery such as dealing with different file formats and arranging content in a manner that would be best suited to online delivery. This paper provides an overview of the Burrokeet System and the technologies used to develop this Open Source Learning Content Management System. It gives a practical view of how an online course using Open Source Learning Content Management tools was developed and delivered.
INTEGRATING WEB-BASED INSTRUCTION AND CLASSROOM TEACHING:CASE STUDIES FROM THE CARIBBEAN
(2010-11-19T16:41:06Z) Cummings, Thompson; Bernard, Margaret
This paper describes an integrated instructional model for integrating web-based instruction and classroom teaching in educational institutions in the Caribbean. Two studies are presented: one, a 1-semester Computer Science course for students at university level and the other, a 2-year course in Information Technology designed for secondary school students across the Caribbean. In each case, webbased and classroom teaching are woven into each other seamlessly throughout the period of instruction to form an integrated instructional system. The studies show that the integrated approach facilitated more effective learning. At university level, teaching emphasis changed from delivery of content to one of facilitating and guiding the learning process. The lecturer’s knowledge of the subject area remained of critical importance to the learning outcome. Students took more responsibility for their own learning and for monitoring their own progress. Teaching methods at secondary schools in the Caribbean are largely traditional, classroom teaching. With the integrated approach to teaching, the learning environment more readily caters for all levels of children within a class, allowing the teacher to provide more individual attention to students. The web course design has to be specially tailored to the student age group and the school environment.
A model for the Instructional Design, Development, Delivery
(2010-11-19T16:41:29Z) Bernard, Margaret; Cummings, Thompson
The steps in producing software are usually organized into a lifecycle consisting of several phases of development. The systems development lifecycle (SDLC) provides a comprehensive framework for describing the process of solving information system problems. Given the SDLC success in software production for many years, a variant of this organized approach is put forward for Web based course production. This variant is called EMBER and consists of five steps. Although each step is presented individually, the steps are interrelated and occur in an iterative manner. The EMBER model is a structured approach to Web based course development. This model focuses on the design, development, delivery, and evaluation of instruction for a Web based course for students at secondary school level in the English speaking Caribbean. A key design decision is that the Web based course would be used to complement classroom teaching. The teacher plays the role of facilitator rather than a dispenser of instruction. This role is not new for the teacher. However, the technology provides the means whereby he/she can perform that role with effectiveness.
A MODEL FOR MAINTAINING INTEROPERABILITY OF
(2010-11-19T16:41:54Z) Bernard, Margaret; Singh, Rajendra G
Learning Objects are being packaged for interoperability using internationally agreed standards such as the Sharable Content Object Reference Model (SCORM). A coursedesigner aggregating heterogeneously created Sharable Learning Objects (SLOs) from a repository when challenged with coarse SLOs would find a standards-based editor to be a useful tool for re-authoring if the interoperability of an edited coarse SLO could remain intact after the edits. Further, since many authoring applications are beginning to use XML to define SLOs (XSLOs) it is inevitable that a repository will contain (solely) heterogeneously authored XSLOs and as a result, we focused our research on maintaining the interoperability of re-authored coarse XSLOs. This paper presents research conducted to determine if a coarse XSLO could be edited in a standards-based editor without affecting its interoperability. Initially a model for the XSLO was developed and titled SIM. We found that if the SIM is applied during the authoring process, it will afford protection to the interoperability of the XSLO when the XSLO is subsequently edited in a corresponding SIM-aware DOM editor. We describe how a DOM editor (based on the Document Object Model) can be transformed into a SIM-aware DOM editor such that it is still standards-based. In addition, we present the application eLearnPro, which was developed to test the interoperability of XSLOs that were authored using the SIM, then re-authored by a SIM-aware DOM editor.
Burrokeet
(2010-11-19T16:42:25Z) Gardler, Ross; Singh, Rajendra G.; Cummings, Thompson; Ramanan, Anil; Mohammed, Shareeda; Bernard, Margaret; Rudder, Andrew
This item relates to the development and deployment of an open-source Learning Content Management System called Burrokeet.
COMPARING A NOVEL QOS ROUTING ALGORITHM TO STANDARD
(2010-11-19T16:42:41Z) Sivakumar, Shyamala; Phillips, Bill; Robertson, William; Goodridge S., Wayne
The problem of finding QoS paths involving several combinations of network metrics is NP-complete. This motivates the use of heuristic approaches for finding feasible QoS paths. Many constraint based routing algorithms find QoSpaths by first pruning resources that do not satish the requirements of the trafic flow and then running a shortest path algorithm on the residual graph. This approach results in a QoS path that biases thefirst metric used in the search process. In addition, it can be shown that this approach may not alwaysfind the optimal path. Our research introduces a QoS routing algorithm that is based on a decision support system that is used to compute QoS paths. We demonstrate the feasibility of this approach by comparing it to standard pruning techniques.
Expert Systems with Applications
(2011-06-13) Kieu, Duc; Chang, Chin-Chen
Recently, Zhang and Wang proposed a steganographic scheme by exploiting modification direction (EMD) to embed one secret digit d in the base-(2 × n + 1) notational system into a group of n cover pixels at a time. Therefore, the hiding capacity of the EMD method is log2(2 × n + 1)/n bit per pixel (bpp). In addition, its visual quality is not optimal. To overcome the drawbacks of the EMD method, we propose a novel steganographic scheme by exploiting eight modification directions to hide several secret bits into a cover pixel pair at a time. By this way, the proposed method can achieve various hiding capacities of 1, 2, 3, 4, and 4.5 bpp and good visual qualities of 52.39, 46.75, 40.83, 34.83, and 31.70 dB, respectively. The experimental results show that the proposed method outperforms three recently published works, namely Mielikainen’s, Zhang and Wang’s, and Yang et al.’s methods.
A High Stego-image Quality Steganographic Scheme with Reversibility and High Payload Using Multiple Embedding Strategy
(2011-06-13) Kieu, Duc; Chang, Chin-Chen
Tian’s method is a breakthrough reversible data embedding scheme with high embedding capacity measured by bits per pixel (bpp) and good visual quality measured by peak signal-to-noise ratio (PSNR). However, the embedding capacity and visual quality of this method can be significantly improved. Thus, we propose a simple reversible steganographic scheme in spatial domain for digital images by using the multiple embedding strategy. The proposed method horizontally and vertically embeds one secret bit into one cover pixel pair. The experimental results show that the proposed reversible steganographic method achieves good visual quality and high embedding capacity. Specifically, with the one-layer embedding, the proposed method can obtain the embedding capacity of more than 0.5 bpp and the PSNR value greater than 54 dB for all test images. Especially, with the five-layer embedding, the proposed method has the embedding capacity of more than 2 bpp and the PSNR value higher than 52 dB for all test images. Therefore, the proposed method surpasses many existing reversible data embedding methods in terms of visual quality and embedding capacity
Better Global Polynomial Approximation for Image Rectification
(2011-06-14) Ward, Christopher
When using images to locate objects, there is the problem of correcting for distortion and misalignment in the images. An elegant way of solving this problem is to generate an error correcting function that maps points in an image to their corrected locations. We generate such a function by fitting a polynomial to a set of sample points. The objective is to identify a polynomial that passes "sufficiently close" to these points with "good" approximation of intermediate points. In the past, it has been difficult to achieve good global polynomial approximation using only sample points. We report on the development of a global polynomial approximation algorithm for solving this problem.
Education in Geographic information systems usability and user-adaptivity
(2011-06-14) McAdams, Michael Andrew; Nikov, Alexander
The objective of this paper is to advance the usability and adaption for users of geographic information systems(GIS) by developing suitably trained professionals via a tailored graduate program. Human-computer interaction models are useful in analyzing the use of GIS in specific task situations.Such analysis provides a sound basis for GIS use optiomization.GIS applications,i.e., in environmental protection or urban and regional planning, require entirely different user interfaces that for those where users are experts. Building such interfaces taking into account GIS usability and adaptation to the user appears to be a promising approach.This could be achieved by careful analysis of GIS utilization and dynamic adaptation to user preferences and interests, the given task,goals and actual work context.The students should learn the usablity principles and how to apply them for designing GIS user interfaces.In addition, they should know to analyze GIS user tasks,interests, and preferences and adapt GIS user interfaces to them.This paper integrates these two complementary disciplines.Examples using ArcView illustrating GIS usability and user adaptivity are given.A MS GIS program with specification in usability and user-adaptive systems is proposed.
A Sudoku Based Wet Paper Hiding Scheme
(2011-06-14) Kieu, Duc; Wang, Zhi-Hui; Chang, Chin-Chen; Li, Ming-Chu
Good image quality and high hiding capacity are two basic requirements of information hiding systems. Technically, it is very challenging to achieve these two factors simultaneously. The purpose of obtaining either high hiding capacity or good image quality are various from application to application. Inspired from the wet paper codes proposed by Fridrich et al., we propose an information hiding scheme for grayscale images. The proposed scheme first uses a secret key to randomly select a subset of pixels from a cover image as dry pixels. Next, the toral automorphism is applied to the cover image to maximize the number of dry pixel pairs. Then, each secret digit in the base-9 numeral system is embedded into one dry pixel pair. The experimental results show that the proposed scheme can achieve good image quality (i.e., PSNR > 46 dB) and flexible hiding capacity. In addition, unauthorized users without knowing the secret key and the secret parameters used for the toral automorphism can not extract the embedded message
A Novel Information Concealing Method Based on
(2011-06-14) Wang, Zhi-Hui; Kieu, Duc; Chang, Chin-Chen; Li, Ming-Chu
A steganographic scheme with a very good visual quality of stego images was proposed by Zhang and Wang. However, the maximum hiding capacity of this method is 1 bit per pixel (bpp). To improve the hiding capacity of Zhang and Wang's method, we propose a steganographic scheme. The proposed method embeds 2K secret digits in the 5-ary notational systems into each group of (2K + 1) cover pixels, where K is a positive integer. Thus, the maximum hiding capacity of the proposed method can approach 2 bpp. The experimental results show that the PSNR values of the proposed method are always higher than 45 dB at the hiding capacity of 1.99 bpp for all test images.
Computational intelligence-based personalization of interactive web systems
(2011-06-14) Rambharose, Tricia; Nikov, Alexander
The main Computational Intelligence (CI) models for personalization of interactive web systems are identified as Fuzzy Systems, Genetic Algorithms, Neural Networks, Artificial Immune Systems and Swarm Intelligence which includes Particle Swarm Optimization, Ant Colony Optimization, Bee Colony Optimization and Wasp Colony Optimization. These models are reviewed and compared regarding their inception, functions, performance and application to personalization of interactive web systems. A taxonomy for personalization of interactive web systems based on CI methods is proposed. It uses two approaches to personalize web-based systems as profile generation and profile exploitation. Based on this taxonomy a general procedure and recommendations for personalization of interactive web systems are suggested. Future directions for application of CI modelling for personalization are discussed.

Browse

Browsing DCIT Lecturers' Papers by Issue Date

Results Per Page

Sort Options