Daftar Isi: PENGELOMPOKAN DOKUMEN BERITA MENGGUNAKAN FUZZY C-MEANS

Main Authors:	ASTUTI, DWI, Primartha, Rifkie
Format:	Thesis NonPeerReviewed Book
Bahasa:	ind
Terbitan:	, 2016
Subjects:	T58.5 General works Management information systems Cf. HD30.213 Industrial management Cf. HF5549.5.C6+ Communication in personnel management Cf. TS158.6 Automatic data collection systems (Production control)
Online Access:	http://repository.unsri.ac.id/23197/1/RAMA_55201_09101002015.pdf http://repository.unsri.ac.id/23197/2/RAMA_55201_09101002015_0001067709_01_front_ref.pdf http://repository.unsri.ac.id/23197/3/RAMA_55201_09101002015_0001067709_02.pdf http://repository.unsri.ac.id/23197/4/RAMA_55201_09101002015_0001067709_03.pdf http://repository.unsri.ac.id/23197/5/RAMA_55201_09101002015_0001067709_04.pdf http://repository.unsri.ac.id/23197/6/RAMA_55201_09101002015_0001067709_05.pdf http://repository.unsri.ac.id/23197/7/RAMA_55201_09101002015_0001067709_06-ref.pdf http://repository.unsri.ac.id/23197/8/RAMA_55201_09101002015_0001067709_07_lamp.pdf http://repository.unsri.ac.id/23197/

Daftar Isi:

News is presented in the amount and type of which many will be difficult grouped manually for news should be read carefully one by one. This then led to a search of news difficulty in finding the appropriate information. This study implements Fuzzy C-Means to classify documents in Indonesian language news by viewing common content words in a document. Fuzzy C-Means method of classifying the data into clusters based on the degree of membership that allows data to be able to enter more than one cluster. The principle of this grouping is to minimize the value of the objective function to determine the value of the degree of initial membership. The greater the degree of membership in a cluster of data, the greater the data into the cluster members. Results of tests on 200 datasets extracted from online media shows the level of accuracy highest fmeasure 0.98. accuracy values obtained during the experiment showed results obtained is highly dependent on the number of variations of the words used in weighting.