Download Annotated Bibliography Assignment for CMSI 698 Research Methods Seminar - Prof. Stephanie and more Exams Computer Science in PDF only on Docsity! CMSI 698 RESEARCH METHODS SEMINAR Fall 2008 Midterm Annotated Bibliography page 1 of 2 Assignment Problems Date Due Midterm Annotated Bibliography 14 OCT 2008 5 Identify and analyze references for classmate's project 21 OCT 2008 6 Proposal draft 4 NOV 2008 Final Research proposal and presentation 9 DEC 2008 Annotated Bibliography (due 14 October) A large part of this course consists of readings from current computer science literature. Such readings, regardless of subject matter, form the foundation of all successful research. An annotated bibliography documents these readings, and you shall write one as part of your course work. The annotated bibliography is equivalent to a midterm exam. Students will present their bibliographies-in-progress to the class over the next few weeks and incorporate class feedback into their bibliographies. You might think of this as analyzing various aspects of your site study, in anticipation to developing the various "elevations" and diagrams that an architectural firm will draw to convey their vision and design to the owner who has come to them to have a home built. This assignment includes both a paper containing the bibliography and a presentation of the related work contained in the bibliography. Bibliography The bibliography actually has two parts: an overview and the references. The overview describes the framework on which the proposal will be built. It needs to accomplish two tasks: identify the major issues with which the proposal must address, identify the significant work that has already been accomplished (and peer reviewed) in the area, and set the scope of the proposal (identify the ideas and references that are beyond the scope of or otherwise not related to the proposal effort). The citation section consists of reference citation/individual annotation pairs. Each citation should be written with a hanging indent; the following annotation should be an indented paragraph. Refer to figure 1 for an example. S.E. August and C.P. Dolan (1992). Hughes Research Laboratories: Trainable Text Skimmer for MUC-4, System Description. Proceedings of the Fourth Message Understanding Conference (MUC-4). Mc Lean, Virginia, 16-18 June, 1992. The Trainable Text Skimmer (TTS) is a statistical system which uses a set of Bayesian classifiers to extract data from free text input and to select and complete the slots of an appropriate event template. First, a set of frequently used phrases is identified in a training corpus. Next, a text is skimmed to determine which category of event template it most closely matches. Finally, relevant data is extracted from the text and used to complete the template. TTS is effective for the majority of slots, but it is thought that performance cannot be improved significantly without including language- or domain-specific processing, of which it currently contains very little. Figure 1. A sample citation/annotation pair.
