.png){kind=icon}
.png){kind=icon}
Return
Ensemble RBM-based classifier using fuzzy integral for big data classification
.png){kind=icon}
PRE
AI
DOI：10.1007/s13042-019-00960-3.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The restricted Boltzmann machine (RBM) is a primary building block of deep learning models. As an efficient representation learning approach, deep RBM can effectively extract sophisticated and informative features from raw data. Little research has been undertaken on using deep RBM to extract features from big data however. In this paper, we investigate this problem, and an ensemble approach for big data classification based on Hadoop MapReduce and fuzzy integral is proposed. The proposed method consists of two stages, map and reduce. In the map stage, multiple RBM-based classifiers used for ensemble are trained in parallel. In the reduce stage, the trained multiple RBM-based classifiers are integrated by fuzzy integral. Experiments on five big data sets show that the proposed approach can outperform other baseline methods to achieve state-of-the-art performance.
Keywords:
Deep learning
Ensemble learning
Restricted Boltzmann machine
Big data
Hadoop MapReduce
Fuzzy integral
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
2.7
Papers:
3.1K
Citations:
5.6K
Organization
View more organizations