We present an access method for set-valued attributes that is based on a multi-level extendible hashing scheme. This scheme avoids exponential directory growth for skewed data and thus generates a much smaller number of subqueries for query sets (so far fast-growing directories have prohibited hash-based index structures for set-valued retrieval). We demonstrate the advantages of our scheme over regular extendible hashing both analytically and experimentally. We also implemented a prototype and briefly summarize the results of our experimental evaluation. © Springer-Verlag Berlin Heidelberg 2007.
CITATION STYLE
Helmer, S., Aly, R., Neumann, T., & Moerkotte, G. (2007). Indexing set-valued attributes with a multi-level extendible hashing scheme. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4653 LNCS, pp. 98–108). Springer Verlag. https://doi.org/10.1007/978-3-540-74469-6_11
Mendeley helps you to discover research relevant for your work.