The Semantic Web Lab (SWlab) at the University of Zakho has achieved a significant milestone with the publication of a groundbreaking research paper in the esteemed Mathematics journal, a leading publication in the field.
The paper, titled “A Semantics-Based Clustering Approach for Online Laboratories Using K-Means and HAC Algorithms,” authored by Saad Hikmat Haji, Karwan Jacksi, and Razwan Mohmed Salah, presents a novel approach to document clustering that addresses the limitations of traditional methods by incorporating semantic analysis.
The research focuses on clustering documents from online laboratory repositories, recognizing the growing importance of organizing and retrieving information from this valuable resource. The researchers developed a novel approach that:
- Collects and Preprocesses Data: Gathers short real-time descriptions of online laboratories from the web and applies various preprocessing techniques, including stemming and removing stop words.
- Creates a Vector Space Model: Utilizes the TF-IDF technique to represent the importance of terms within each document, creating a vector space model.
- Performs Semantic Clustering: Employs K-Means and Hierarchical Agglomerative Clustering (HAC) algorithms to group documents based on their semantic similarity, considering the underlying meaning of the text rather than just keyword occurrences.
The performance of the proposed approach was evaluated using a comprehensive set of metrics, including Silhouette average, purity, V-measure, F1-measure, accuracy score, homogeneity score, completeness, and NMI score. The results were compared across three scenarios: without preprocessing, preprocessing with stemming, and preprocessing without stemming, using five diverse datasets.
The findings of this research demonstrate the effectiveness of the proposed semantics-based clustering approach in improving the accuracy and quality of document clustering for online laboratory repositories. The results were also visualized in an interactive webpage for easy interpretation and exploration.
This publication in the Mathematics journal is a testament to the high-quality research conducted by the SWlab at the University of Zakho and its significant contributions to the field of document clustering and semantic web technologies.
For further details and access to the full paper, please refer to:
- Publication: Mathematics
- Volume: 11
- Issue: 3
- Pages: 548
- Authors: Saad Hikmat Haji, Karwan Jacksi, Razwan Mohmed Salah
- Publisher: MDPI