The Rate of the Precision in the Audio Visual Retrieval Resource by K-Means Algorithm

The aim of this paper is to survey the rate of the precision in the audio and visual material retrieval by clustering techniques and k- means algorithm.
The type of the research method is experimental. For the purpose of this paper, twenty- three albums are studied directly through observation. Albums published in summer 1391(solar). Songs catalogs according to PBCore metadata and one of the elements in this metadata choose instead of variable. In the next step, k- means algorithm in SPSS software was performed. Finally, the distance between any records with cluster center was calculated.
Finding: After performing k- means algorithm, the cluster center for the four clusters of intellectual content, intellectual property, instantiation, and extensions were determined. Due to the distance between each record and the center of the cluster this value is measured as the precision of the resource retrieval records. The closer to the center of each cluster the shorter the distance would be. However, the record-with the highest similarity to the center of the cluster has the greatest distance. The precision is calculated at 99 percent.