Implementing agglomerative hierarchic clustering algorithms for use in document retrieval
Information Processing and Management: an International Journal
Information retrieval: data structures and algorithms
Information retrieval: data structures and algorithms
ACM Computing Surveys (CSUR)
Performance Evaluation of Some Clustering Algorithms and Validity Indices
IEEE Transactions on Pattern Analysis and Machine Intelligence
Ant colony system: a cooperative learning approach to the traveling salesman problem
IEEE Transactions on Evolutionary Computation
| Hi-index | 0.00 |
Traditional hierarchical clustering methods adopt a greedy strategy to merge objects progressively and construct a clustering dendrogram. However, their clustering quality might not be reliable because only local optimal information is referred during a dendrogram construction. To conquer the problem, this paper proposes a global optimal strategy to guide the dendrogram construction. The strategy aims to find an optimal circular traveling order that minimizes the total traveling distances for visiting all objects along the branches of the dendrogram, which is viewed as a traveling salesman problem (TSP). The TSP problem is solved using the variable neighborhood search (VNS) method because of its parameter-free advantage. Then, the clustering dendrogram is constructed based on the information provided by the order. Through our experiments, the clustering quality of our proposed method is superior to traditional hierarchical clustering methods.