K-means clustering and visualizing of significant words prototype for crowdsourced relationship comments / Muhammad Syafiq Mastor

Mastor, Muhammad Syafiq (2015) K-means clustering and visualizing of significant words prototype for crowdsourced relationship comments / Muhammad Syafiq Mastor. Degree thesis, Universiti Teknologi MARA.

[img]
Preview
Text
TD_MUHAMMAD SYAFIQ MASTOR CS 15_5.pdf

Download (297kB) | Preview

Abstract

Having a love relationship is normal to the humankind. It is the most common emotion that human has. Nevertheless, conflicts often occur throughout the relationship. Conflicts occur due to lack of understanding between partners. The most common conflicts are the misunderstanding women‟s feeling from their partners. In spite of the conflicts, men would do anything that is possible to understand their partners in order to save their relationship, either going for relationship counselling, asking for opinion in social media, learn from other‟s experience or even seeking advice from family and friends. However, these conventional methods have their disadvantages. Therefore, this research proposed a K-means clustering and visualizing of significant words prototype for crowdsourced relationship comments. Crowdsourcing is the practice of obtaining needed content by asking contributions from a batch of people and typically from the online community. The comments are clustered using K-Means algorithm and visualized using scatter plot. Each plot represents a comment where each comment are coloured according to their cluster and significant words. Compared to the existing relationship website, most of the website did not give freedom for user to browse on comments that have same ideas as they are. Most of the website force user to follow the highly rated advice which may not applicable to the user. This project gives freedom for users to choose which comments or advice that is suitable for their situation as well as to give comments or advice anonymously. This project also identifies the common issues or topics has been discussed in the comments.

Item Type: Thesis (Degree)
Uncontrolled Keywords: K-Means algorithm; Clustering Analysis; Visualized
Subjects: Q Science > QA Mathematics > Real-time programming
Q Science > QA Mathematics > Evolutionary programming (Computer science). Genetic algorithms
Divisions: Universiti Teknologi MARA, Melaka > Jasin Campus > Faculty of Computer and Mathematical Sciences
Depositing User: Staf Pendigitalan 5
Date Deposited: 17 Aug 2016 09:28
Last Modified: 23 Aug 2016 03:28
URI: http://ir.uitm.edu.my/id/eprint/14569

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year