Google distance analysis of the MIT curriculum
Author(s)
Dugan, Casey (Casey A.)
DownloadFull printable version (15.68Mb)
Other Contributors
Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Advisor
Hal Abelson.
Terms of use
Metadata
Show full item recordAbstract
The MIT curriculum has traditionally been organized into Departments and Schools, or groups of Departments. This influences degree titles and their requirements, subject listings seen by students and professors, and more. But subjects, Departments and Schools are not systematically analyzed or organized based on the knowledge they cover. The system built mined a subset of Institute subjects, taken from OpenCourseWare, for key topics, and performed Google searches on those terms. By composing keyword search result vectors for each subject, scores can be calculated between all pairs of subjects. These scores were used by an MDS layout algorithm and a Hierarchical Clustering algorithm to suggest two new organizations each for Department 6 (E.E. and C.S.) and for the MIT Departments into Schools. Convex hulls in MDS graphs of Department 6, colored based on classes a student has taken, are also used to predict new departments the student will take.
Description
Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2006. Includes bibliographical references (leaf 50).
Date issued
2006Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer SciencePublisher
Massachusetts Institute of Technology
Keywords
Electrical Engineering and Computer Science.