Using distributed word representation and document distance for short text document clustering

Name: Supavit Kongwudhikunakorn

LCSH: Kasetsart University -- Theses. M.Eng. (Computer Engineering) 2018

Classification :.LCCS: QA278.55

LCSH: Kasetsart University. -- Department of Computer Engineering -- Theses

LCSH: Document clustering

LCSH: Cluster analysis

LCSH: Data mining

Abstract: As short text documents are becoming the main form of communication, the number of these documents are rapidly increase. Latent information often contained in these documents. The lack of suitable knowledge discovery process often leave this valuable information neglected. One of the knowledge discovery process is clustering. The objective of this thesis is to identify the best combination of document representation, document distance, and document clustering mechanism that yields the best clustering quality. Since short text representations are often sparse which lead to high running time, vector densification is needed. Document representations are expanded by external knowledge sources represented by a Distributed Representation and densified with the vector densification mechanism. To cluster documents, K-means++ partitioning-based clustering technique is applied, where the similarities of documents are measured by knowledge-based Word Mover’s Distance. To validate the effectiveness of the proposed method, experiments were conducted to compare the clustering quality against several leading methods. The proposed method produced clusters of documents that resulted in higher precision, recall, F1-score, and adjusted Rand index for both real-world and standard data sets. Furthermore, manual inspection of the clustering results was conducted to observe the efficacy of the proposed method. The topics of each document cluster are undoubtedly reflected by members in the cluster.

Kasetsart University. Office of the University Library

Address: Bangkok

Email: tdckulib@ku.ac.th

Name: Kitsana Waiyamai

Role: Thesis Advisor

Name: Thanawin Rakthanmanon

Role: Thesis CO-Advisor

Name: Hutchatai Chanlekha

Role: Thesis CO-Advisor

Created: 2018

Modified: 2025-07-09

Issued: 2025-07-09

วิทยานิพนธ์/Thesis

application/pdf

URL: https://www.lib.ku.ac.th/KUthesis/0error/2561/supavit-kon-all.pdf

CallNumber: QA278.55 .S87

eng

DegreeName: Master of Engineering

Level: Master's Degree

Descipline: Computer Engineering

Grantor: Kasetsart University

©copyrights Kasetsart University

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	supavit-kon-all.pdf	14.42 MB

ใช้เวลา

0.022444 วินาที