A Semantics-Based Clustering Approach for Online Laboratories Using K-Means and HAC Algorithms

Due to the availability of a vast amount unstructured data in various forms (e.g., web, social networks, etc.), clustering text documents has become increasingly important. Traditional algorithms have not been able solve this problem because semantic relationships between words could accurately represent meaning documents. Thus, document extensively utilized enhance quality clustering. This method is called unsupervised learning and it involves grouping based on their meaning, common keywords. paper introduces new that groups from online laboratory repositories similarity approach. In work, dataset collected first by crawling short real-time descriptions laboratories’ Web. A vector space created using frequency-inverse frequency (TF-IDF) done K-Means Hierarchical Agglomerative Clustering (HAC) with different linkages. Three scenarios are considered: without preprocessing (WoPP); steaming (PPwS); (PPWoS). Several metrics used for evaluating experiments: Silhouette average, purity, V-measure, F1-measure, accuracy score, homogeneity completeness NMI score (consisting five datasets: labs, 20 NewsGroups, Txt_sentoken, NLTK_Brown NLTK_Reuters). Finally, creating an interactive webpage, results proposed work contrasted visualized.