论文信息 - Running Multiple Neural Networks for Process Trend Interpretation

Running Multiple Neural Networks for Process Trend Interpretation

An architecture for running multiple ART2 neural networks in parallel is presented. Each of the networks can treat different sizes of clusters. As the result, these multiple networks can show higher performance for detecting new classes. A proposed method is demonstrated for the detection of model changes for time-series models.

Yoshiyuki Yamashita | Hiroshi Komori | Mutsumi Suzuki

[1] James Robert Whiteley. Knowledge-based interpretation of process sensor patterns / , 1991 .

[2] James R. Whiteley,et al. A similarity-based approach to interpretation of sensor data using adaptive resonance theory , 1994 .

[3] Yoshiyuki Yamashita. Qualitative interpretation of process trends by using neural networks , 1998, 1998 Second International Conference. Knowledge-Based Intelligent Electronic Systems. Proceedings KES'98 (Cat. No.98EX111).

[4] Amit Mehrotra,et al. Observations and problems applying ART2 for dynamic sensor pattern interpretation , 1996, IEEE Trans. Syst. Man Cybern. Part A.

[5] M Fernández-Delgado,et al. MART: a multichannel ART-based neural network. , 1998, IEEE transactions on neural networks.

[6] James F. Davis,et al. Qualitative interpretation of sensor patterns , 1993, IEEE Expert.

[7] J. R. Whiteley,et al. Knowledge-based interpretation of sensor patterns , 1992 .

[8] Stephen Grossberg,et al. Art 2: Self-Organization Of Stable Category Recognition Codes For Analog Input Patterns , 1988, Other Conferences.

[9] Stephen Grossberg,et al. ART 2-A: An adaptive resonance algorithm for rapid category learning and recognition , 1991, Neural Networks.