Research on incomplete data mining and filling algorithm during depth learning process

DOI

10.2991/meita-16.2017.67How to use a DOI?

Keywords

Depth learning; missing data filling; automatic coding

Abstract

In this paper, an incomplete data padding algorithm based on depth learning is proposed. The algorithm has a rich information dimension for large data. A depth-filling network is constructed to extract the depth features of large data, and then the missing values are restored. Experimental results showed that the algorithm proposed in this paper can effectively improve the accuracy of data filling. To solve this problem, this paper proposes an incomplete data filling algorithm based on depth learning. The algorithm is based on the automatic coding machine to establish the automatic filling machine. On this basis, a deep-filling network model is constructed to analyze the depth characteristics of incomplete data and calculate the network parameters according to the layer-by-layer training idea and back propagation algorithm. Finally, the incomplete data is restored by deep filling network, and the missing value is filled. In the next step, we explore how to improve the data filling accuracy in multi-miss mode.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Liping Wang
PY  - 2017/02
DA  - 2017/02
TI  - Research on incomplete data mining and filling algorithm during depth learning process
BT  - Proceedings of the 2016 2nd International Conference on Materials Engineering and Information Technology Applications (MEITA 2016)
PB  - Atlantis Press
SP  - 325
EP  - 329
SN  - 2352-5401
UR  - https://doi.org/10.2991/meita-16.2017.67
DO  - 10.2991/meita-16.2017.67
ID  - Wang2017/02
ER  -