,
abercrombie marseille
distance-based outlier metric space k-NN query incremental mining flexibility
Metric space in the fast distance-based outlier mining
Chinese Summary: The k-nearest neighbor data points from the index level in isolation can effectively find the data set of isolated points,
abercrombie paris, but the basic algorithm requires O (Nz) calculate the distance between the number of positions,
franklin marshall femme, does not apply to large data sets. This presents a metric space in the triangle inequality using the fast mining algorithm? Early pruning. ADVP using k-NN query for each neighbor point to be saved query to calculate the distance of neighboring points on the boundary of isolation. Degree of isolation have been found less than the weakest sector of the isolated outlier degree of data points can be trimmed without the need for k-NN query. Sampling method based on optimization of the search order to improve the pruning effect. While ADVP naturally extended to the incremental algorithm. Large data sets in the standard experimental results show that, ADVP and significant savings compared to existing algorithms to calculate overhead,
franklin marshall paris, better scalability; incremental ADVP to deal effectively with new data. More articles related to topics:
An improved uconc service grid authorization decision-making model
parameters of fuzzy self-tuning pid controller design and simulation
Compton scattering and the second generation two-photon Compton scattering -