Visualisasi keterhubungan antaristilah subjek pada metadata tesis IPB menggunakan peluang bersyarat

Abstract

Bogor Agricultural University library has metadata for its collection. The metadata is used for searching and browsing. The activity of searching and browsing can be based on metadata fields, such as title, author, subject, etc. Searching on the subject or topic is an effective way if someone want to search document that fulfill his/her topic. When someone search on a subject, the result is only documents that has the subject. So, someone needs to pick an appropriate query to find the desirable result. The decision for picking a query is based on how good the knowledge of someone about the collection. It will be hard if someone does not know about the collection. In that case, we need an application that shows the relationship between the subject, so the related terms can be used as query suggestion for someone to decide his/her query. In order to make the view of relationship more interesting, understandable, and interactive, the relationship is visualized. The result of this research is an implementation of a Java desktop application that searches document based on subject and visualize subject terms relationship. The documents are from IPB thesis metadata. From 7445 documents, 7416 documents has subject field. Obtained 4496 unique terms from the 7416 document. From test result, the process to show document result and visualization is fast enough. The average time to show document result and visualization is one second.