Autocannibalistic and Anyspace Indexing Algorithms with Application to Sensor Data Mining.

摘要: Efficient indexing is at the heart of many data mining algorithms. A simple and extremely effective algorithm for under any metric space was introduced in 1991 by Orchard. Orchard’s has not received much attention database community because a fatal flaw; it requires quadratic space. In this work we show that can produce reduced version less space, but produces nearly identical speedup. We achieve casting an anyspace framework, allowing deployed applications to take as index their main memory/sensor afford. As shall demonstrate, ability create also allows us auto-cannibalistic Auto-cannibalistic algorithms are which initially require certain amount or classify data, if unexpected circumstances them store additional information, they dynamically delete parts themselves make room new data. demonstrate utility autocannibalistic fielded project on insect monitoring with low power sensors, autonomous robot application.