Title | ||
---|---|---|
A Safety Evaluation Method Of Mine Pressure Based On Incomplete Labeled Data Stream Classification |
Abstract | ||
---|---|---|
Mine pressure monitoring data is essentially a data stream, and the safety label of mine pressure is difficult to obtain. In this case, the safety evaluation of mine pressure can be regarded as the incomplete labeled data stream classification, and classification labels are safety and unsafety. The safety evaluation method of mine pressure used in this paper is a incomplete labeled data stream classification algorithm based on K means clustering, which uses K means clustering to label unlabeled data, and uses support vector machine as the base classifier, and uses Bayesian classifier to filter noise data, and uses the double thresholds determined by Hoeffding Bounds inequality to detect concept drifts. Experimental results show the method can better label unlabeled data, and better detect concept drifts in data stream, and it has better classification accuracy for data stream, and it can be applied to the safety evaluation of mine pressure. |
Year | Venue | Keywords |
---|---|---|
2017 | 2017 13TH INTERNATIONAL CONFERENCE ON NATURAL COMPUTATION, FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY (ICNC-FSKD) | mine pressure, safety evaluation, incomplete label, data stream classification |
Field | DocType | Citations |
k-means clustering,Naive Bayes classifier,Pattern recognition,Data stream,Computer science,Coal mining,Support vector machine,Artificial intelligence,Statistical classification,Cluster analysis,Classifier (linguistics),Machine learning | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Gang Sun | 1 | 463 | 36.98 |
Jia Zhao | 2 | 36 | 8.80 |
Zhongxin Wang | 3 | 0 | 0.68 |
Hao Wang | 4 | 7 | 2.79 |
Chuanyun Ni | 5 | 0 | 0.68 |
Guobao Cai | 6 | 0 | 0.34 |
Xiaolian Chen | 7 | 0 | 0.34 |