Robust Classification with Noisy Labels for Manufacturing Applications: A Hybrid Approach Based on Active Learning and Data Cleaning.
Shuo ZhaoXin LiYing-Chi ChenPublished in: IECON (2021)
Keyphrases
- data cleaning
- robust classification
- active learning
- data integration
- data quality
- text classification
- record linkage
- training set
- outlier detection
- database
- missing values
- data processing
- web usage mining
- data warehouse
- fraud detection
- multi label
- learning algorithm
- data warehousing
- semi supervised
- databases
- cost sensitive
- pairwise
- missing data
- data sources
- pattern recognition
- training data
- website
- data extraction
- subspace methods
- web pages
- data model