Computer Science Undergraduate Honours Essays (Okanagan Campus)
http://hdl.handle.net/2429/42497
Wed, 27 May 2015 21:26:13 GMT2015-05-27T21:26:13ZSchedule Viewer: A Scheduling Tool for UBC Okanagan Administration (Issued:2009)
http://hdl.handle.net/2429/27052
Schedule Viewer is an add‐in for Microsoft Office Outlook 2007 developed to assist administrative staff and faculty members in viewing a selected subset of academic schedules, which has not been possible until now. Information on times, professors and locations for classes has always been readily available online, but to make productive use of this data staff members have been forced to manually sift through spreadsheets and compile the data themselves. The Schedule Viewer application allows users to enter search criteria in order to select a group of time blocks (classes, labs, etc.) and have these time blocks displayed in a useful graphical way. Time blocks are displayed as appointments in an Outlook calendar which allows users to easily view the layout of academic schedules in relation to their specific query. In addition to viewing existing schedule information, users can create their own Outlook appointments which are then saved in the database for others to work with.
The Schedule Viewer application runs on a three‐tier client‐server architecture. The client portion of the system involves the Outlook add‐in which runs locally on each user’s machine. There are two components to the server architecture. The server application is written in Java and runs on a campus server. The server application communicates directly with each client and provides the link from the client to the database. The database is located on another server on campus and stores time block
information including times, dates, and locations. All information stored in the database, which was created specifically for Schedule Viewer, is obtained from the UBC Student Services web site [4] using an HTML screen scraping process.
Thu, 01 Jan 2009 00:00:00 GMThttp://hdl.handle.net/2429/270522009-01-01T00:00:00ZThe Piecewise Linear-Quadratic Model for Convex Bivariate Functions (Issued:2008)
http://hdl.handle.net/2429/27051
A Piecewise Linear-Quadratic (PLQ) function is used to describe data that can be represented by continuous functions with a piecewise linear domain for which the function is either linear or quadratic on, each piece of its domain [11]. While extensive work has been done with PLQ models for convex univariate functions, my work investigates the development of a two dimensional model that allows the implementation of algorithms for computing fundamental convex transforms. PLQ functions have been described in the literature, the efficient implementation of the algorithms requires the careful selection of a data structure. Initial investigation examined using a Voronoi diagram to represent the projection of the bivariate function into R2, to take advantage of efficient algorithms for the point location problem [3]. While suitable for representing a single PLQ, with operations for building a zero order model from data and evaluating the function over an irregular grid, we are uncertain
if it can be extended to compute other fundamental convex transforms efficiently.
In examining the Voronoi model, we were able to extend the base concept to represent the bivariate PLQ using two different representations: a tessellation-based model and a linear inequality-based model. The tessellation model is restricted to representing a PLQ with a bounded domain. The model represents data through triangular faces in a tessellation in R2 where each face is defined by its vertices and the associated function value. This model allows for the efficient evaluation over an irregular grid, the addition of two PLQ functions with bounded domains and multiplication by a scalar, but does not allow for the representation of an unbounded domain. To allow for an unbounded domain, a dual model was developed using linear inequalities to represent each face in R2. Numerical results are presented for each model and computational complexity of model components are discussed.
Tue, 01 Jan 2008 00:00:00 GMThttp://hdl.handle.net/2429/270512008-01-01T00:00:00ZComputer Modeling of Molecular Genetic Events using Content Sensitive Analysis (Issued:2007)
http://hdl.handle.net/2429/27050
The problem of motif finding plays an important role in understanding the development, function and evolution of organisms. The Planted (l, d)-Motif Problem first introduced by Pevzner and Sze [13] is a variant model of motif finding that has been widely studied. Nevertheless, despite the many different algorithms constructed to solve this problem, it is far from being solved entirely [5]. We analyze a number of algorithms including: the Basic Voting Algorithm, the Winnower and the Closest String. One thing that has become ubiquitous among these algorithms is the tradeoff between efficiency and accuracy. We formulate the motif-finding problem as a constraint satisfaction problem and introduce a special form of constraint consistency. By using a fixed-parameter algorithm for the closest string problem [4], we develop a propagation algorithm that enforces the new constraint consistency with the same worst-case time complexity as that of the standard path consistency algorithm. Experiments on randomly generated sequences indicate that our approach is effective and efficient.
Mon, 01 Jan 2007 00:00:00 GMThttp://hdl.handle.net/2429/270502007-01-01T00:00:00ZSegmentation of the Radiation Treatment Field in Dual Portal Images (Issued:2007)
http://hdl.handle.net/2429/27049
Conformal radiotherapy involves many treatments and multiple beams of radiation applied to the patient. For conformal radiotherapy the successful detection of the radiation field resulting from a treatment is critical for quality assurance purposes. Errors over multiple
treatments can compound resulting in the death of healthy cells while allowing cancerous cells to survive. Often the verification is performed manually by a technician which, across multiple treatments, is very time consuming.
However, automated methods have been developed using gradient thresholds to segment the dual portal image for field detection. I present an improvement of the accuracy of the generic gradient threshold technique achieved by incorporating the gradient directional information. I have implemented other segmentations techniques, such as the Fast Marching Method, that propagate a curve based on the properties of the image with the final curve location representing the detected field edge. Though the directional gradient threshold method is often more accurate than previous methods, it is less easily automated; whereas the curve propagation methods are easy to automate, are less computationally intense, and produce more accurate segmentations.
Mon, 01 Jan 2007 00:00:00 GMThttp://hdl.handle.net/2429/270492007-01-01T00:00:00Z