A CMAC-based scheme for determining membership with classification of text strings

摘要: Membership determination of text strings has been an important procedure for analyzing textual data a tremendous amount, especially when time is crucial factor. Bloom filter well-known approach dealing with such problem because its succinct structure and simple procedure. As membership classification becoming increasingly desirable, parallel filters are often implemented facilitating the additional requirement. The filters, however, tend to produce false-positive errors since must be performed on each layers. We propose scheme based CMAC, neural network mapping, which only requires single-layer calculation simultaneously obtain information both classification. A hash function specifically designed also proposed. proposed could effectively reduce by converging range acceptance minimum class during mapping. Simulation results show that committed significantly less than benchmark, limited identical memory usage at different levels.