Efficient Similarity Query Processing in High Dimensional Databases (2006–2008)
One of the biggest challenges in modern data management is to deal with very large amount of highly complex and highly diversified data, to manage it and find required information efficiently. Similarity query processing is to search all data items similar to a query target (data retrieval), or similar to each other (data clustering). This type of query can be found, directly or after some transformation, from numerous application areas, such as spatiotemporal databases, multimedia information management, web search, bioinformatics, and data mining. In this project, we investigate this fundamental issue as a general problem, generalised from many related areas and aiming at applying our results to these areas.