发明名称 |
METHOD FOR DEVELOPING MACHINE OPERATION CLASSIFIER USING MACHINE LEARNING |
摘要 |
A method (70) for developing machine operation classifiers for a machine (11) is disclosed. The method (70) includes receiving training data (32) associated with the machine (11) from one or more on-board engineering channels (21) associated with the machine (11) and determining one or more training features (34) based on the training data values (32). The method (70) also includes determining one or more training labels (36) associated with the one or more training features (34) and building a predictive model (56) for determining machine operation classifiers using a computer (60). Building the predictive model (56) may include feeding the one or more training features (34) and the one or more training labels (36) associated with the one or more training features (34) to a machine learning algorithm (42) and determining a predictive model (56) from the machine learning algorithm (42). The predictive model (56) may be used for receiving new data (52) associated with the machine (11) and determining a predicted label (58) based on the new data (52). |
申请公布号 |
WO2016044263(A1) |
申请公布日期 |
2016.03.24 |
申请号 |
WO2015US50168 |
申请日期 |
2015.09.15 |
申请人 |
CATERPILLAR INC. |
发明人 |
HODEL, BENJAMIN;KIM, SANGKYUM;LEE, PAUL |
分类号 |
G06N99/00 |
主分类号 |
G06N99/00 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|