<![CDATA[A minor thesis is a document of a scientific work compiled by a student at the level of stratum 1 which discusses a particular topic or field of research or development results that the student has undertaken in order to take the final examination to obtain a degree. In the Reading Room of the Faculty of Computer Science and the Central Library of Brawijaya University there is a problem that arises that there is no categorization of all minor thesis documents stored. Hierarchical Agglomerative Clustering (HAC) method is implemented for clustering minor thesis documents based on minor thesis title. HAC classifies iterative documents from the smallest cluster to the largest 1 cluster. Input data that is in the form of title of minor thesis document of Informatics Engineering Brawijaya University. The preprocessing stage is performed on the minor thesis title data to get the term feature. All the terms obtained are processed to get the weight of TF-IDF. The value of similarity between documents obtained from the value of cosine distance. The clustering process uses 3 distance options as the single linkage, complete linkage and average linkage parameters. The clustering results of each distance parameter are displayed on the label of each cluster generated and each cluster generated is evaluated using silhouette coefficient. From the test result on 100 minor thesis documents obtained the value of Silhouette Coefficient from single linkage is 0,10125, complete linkage is 0,155733 and average linkage is 0,160428. Average linkage is better in grouping documents than single linkage and complete linkage.]]>
Copyrights © 2018
