Logical design for temporal databases with multiple granularities
ACM Transactions on Database Systems (TODS)
Dynamic support vector machines for non-stationary time series forecasting
Intelligent Data Analysis
Hi-index | 0.00 |
This paper studies a multi-classification method based on support vector machine for temporal data. First, we give classic classification model of support vector machine. Then, we present a support vector machine model based on multi-weighted values, which is used to deal with multi-classification problems of temporal data. We define temporal type and prediction model for the temporal data. According to the temporal type model and the support vector machine model based on multi-weighted values, we propose a multi-classification method based on the support vector machine. Finally, experiments results show that our method can effectively solve the misclassification problems of temporal data.