Perbandingan akurasi k-means dan fuzzy c-means untuk menentukan profesi berdasarkan data twitter

Elisabet, Sihite (2018) Perbandingan akurasi k-means dan fuzzy c-means untuk menentukan profesi berdasarkan data twitter. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

Preview	Text COVER.pdf - Accepted Version Download (1MB) | Preview
Preview	Text Abstract.pdf - Accepted Version Download (9kB) | Preview
Preview	Text Abstrak.pdf - Accepted Version Download (110kB) | Preview
Preview	Text BAB I.pdf - Accepted Version Download (112kB) | Preview
Preview	Text BAB II.pdf - Accepted Version Download (588kB) | Preview
	Text BAB III.pdf - Accepted Version Download (370kB)
	Text BAB IV.pdf - Accepted Version Restricted to Registered users only Download (854kB)
Preview	Text BAB V.pdf - Accepted Version Download (10kB) | Preview
Preview	Text Daftar Pustaka.pdf - Accepted Version Download (155kB) | Preview

Abstract

At this time due to the amount of data is so large, Twitter is not only used for writing a message and microblogging becomes the means of knowledge regarding things that are new and may investigate a phenomenon happening in people's lives. One of the phenomena that occur in the social life of the community is the selection of jobs that do not match the capabilities and expertise. One of the techniques used to investigate phenomena that occur by using the method of clustering. Clustering methods are the most commonly used is the K-Means method to perform a comparison of the accuracy of the clustering method with Fuzzy C-Means which is a modification of the K-Means clustering method. The first stage at the time of entering data clustering Twitter then grouped with K-Means method into 4 clusters i.e. cluter entrepreneurs (472 words), educators (668 words), Office employees (140 words) and cluster of artists (599 words). A method of Fuzzy C-Means cluster group members produce entrepreneurs (527 words), educators (1012), Office employees (193 words) and cluster of artists (778 words). The method used to perform the calculation accuracy by using the value of the internal and external validity, where the results of K-Means method and Fuzzy C-Means showing that the value of the validity of the resulting K-Means more high-value the external validity of the statistics of 0.86 and Rand to the value of the coefficient of 0.71. Therefore it can be concluded that K-Means method was more accurate in grouping the profession based on Twitter data. Keywords - Twitter, Clustering, Comparative Accuracy, K-Means, Fuzzy C-Means, The Validity.

Actions (login required)

View Item