Soft and Hard Clustering for Abstract Scientific Paper in Indonesian

For ease in grouping research papers is by doing clustering. Clustering is a method to classify the objects into subsets with similar attributes. Clustering method divided into two categories ie hard and soft clustering. Hard clustering is method to grouping the data items such that each item is only assigned to one cluster, K-Means is one of them. While Soft clustering is method to grouping the data items such that an item can exist in multiple clusters, Fuzzy C-Means (FCM) is an example. Most research papers are documented into groups that are associated with the area of expertise of the researcher, even though there is also research whose contents relate to other fields outside the area of expertise of the researcher, so it should also be documented in the group of other fields so that its contribution in other fields can be known. Here in this paper we analyse the abstract of papers written in Indonesian as data set. Data samples were taken from 3 fields, namely information technology, health and economics. Clustering process using k-means and FCM to find out whether scientific paper's abstracts from different fields of research can be in the same group / cluster as a whole, not whole or different groups. As unstructured data, abstracts must be processed through a text mining procedure first to become vector data.