Search for Articles:

MECS Press Menu

Aleta C. Fabregas

Work place: Graduate Programs, Technological Institute of the Philippines, Quezon City, Philippines

E-mail: alet_fabregas@yahoo.com

Website:

Research Interests: Programming Language Theory, Comparative Programming Language Analysis

Biography

Aleta C. Fabregas took her Bachelor of Computer Data Processing Management (BCDPM) with honors (cum laude) in Polytechnic University of the Philippines (PUP), Manila in 1986. She obtained her Master’s Degree in Master in Information Technology (MIT) from PUP Manila in 2006. She is currently working at PUP Manila as a full-time faculty with the rank of Associate Professor in the College of Computer and Information Sciences (CCIS), Associate Lecturer in Master of Science in Information Technology(MSIT), Graduate School and Course Facilitator at the post-baccalaureate program, master program and Expanded Tertiary Education Equivalency and Accreditation Program (ETEEAP) at the PUP Open University System. She is also a module writer in Java programming language and other programming languages.

Prof. Fabregas is a senior member of Royal Institute of Information Technology, Singapore, Philippine Schools, Universities and Colleges Computer Education and Systems Society, Inc.(PSUCCESS), and Philippine Computer Society (PCS).

Author Articles
Enhanced Initial Centroids for K-means Algorithm

By Aleta C. Fabregas Bobby D. Gerardo Bartolome T. Tanguilig III

DOI: https://doi.org/10.5815/ijitcs.2017.01.04, Pub. Date: 8 Jan. 2017

This paper focuses on the enhanced initial centroids for the K-means algorithm. The original k-means is using the random choice of initial seeds which is a major limitation of the original K-means algorithm because it produces less reliable result of clustering the data. The enhanced method of the k-means algorithm includes the computation of the weighted mean to improve the centroids initialization. This paper shows the comparison between K-Means and the enhanced K-Means algorithm, and it proves that the new method of selecting initial seeds is better in terms of mathematical computation and reliability.

[...] Read more.
Other Articles

Subscribe to receive issue release notifications and newsletters from MECS Press journals