Security management center needs to detect and delete many similar records of the device status information to reduce the data redundancy before analyzing the status of the supervised device. Most similarity record detection algorithms are based on the “sort-merge” model. Detection algorithms usually sort data set with keywords before detection of similar data. Existing methods of generating keywords tend to have the following problems: the keywords is not accurate, or multiple keywords are generated for sorting of multiple keywords. The paper proposes a method of synthesizing keywords by multiple encoding fields, and it is verified that this method can significantly optimize the performance of algorithm through experiment. We also compare the performance of each common detection algorithm through experiment.
CITATION STYLE
Liu, Z., Fang, L., Yin, L., Guo, Y., & Li, F. (2017). Research on similarity record detection of device status information based on multiple encoding field. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10658 LNCS, pp. 54–63). Springer Verlag. https://doi.org/10.1007/978-3-319-72395-2_6
Mendeley helps you to discover research relevant for your work.