Towards Wide Learning: Experiments in Healthcare
This addresses the bottleneck of feature engineering for machine learning practitioners in healthcare, offering a more efficient approach, though it appears incremental as it builds on existing architectures.
The paper tackles the problem of automating feature engineering in machine learning, which is time-consuming and requires expert knowledge, by proposing a Wide Learning architecture and testing it on three healthcare datasets, achieving state-of-the-art results on two datasets and 94.38% of the winner's accuracy on the third, with effort reduced to a few days compared to manual methods.
In this paper, a Wide Learning architecture is proposed that attempts to automate the feature engineering portion of the machine learning (ML) pipeline. Feature engineering is widely considered as the most time consuming and expert knowledge demanding portion of any ML task. The proposed feature recommendation approach is tested on 3 healthcare datasets: a) PhysioNet Challenge 2016 dataset of phonocardiogram (PCG) signals, b) MIMIC II blood pressure classification dataset of photoplethysmogram (PPG) signals and c) an emotion classification dataset of PPG signals. While the proposed method beats the state of the art techniques for 2nd and 3rd dataset, it reaches 94.38% of the accuracy level of the winner of PhysioNet Challenge 2016. In all cases, the effort to reach a satisfactory performance was drastically less (a few days) than manual feature engineering.