Task-wise Split Gradient Boosting Trees for Multi-center Diabetes Prediction
This work addresses diabetes prediction for healthcare applications by improving accuracy in multi-center settings, though it appears incremental as it builds on existing gradient boosting and multi-task learning techniques.
The authors tackled the challenges of data heterogeneity and insufficiency in multi-center diabetes prediction by proposing Task-wise Split Gradient Boosting Trees (TSGB), which integrates gradient boosting decision trees with multi-task learning and introduces a novel split method based on task gain statistics. Experiments on real-world and benchmark datasets showed TSGB achieves superior performance against state-of-the-art methods, with deployment as an online risk assessment software.
Diabetes prediction is an important data science application in the social healthcare domain. There exist two main challenges in the diabetes prediction task: data heterogeneity since demographic and metabolic data are of different types, data insufficiency since the number of diabetes cases in a single medical center is usually limited. To tackle the above challenges, we employ gradient boosting decision trees (GBDT) to handle data heterogeneity and introduce multi-task learning (MTL) to solve data insufficiency. To this end, Task-wise Split Gradient Boosting Trees (TSGB) is proposed for the multi-center diabetes prediction task. Specifically, we firstly introduce task gain to evaluate each task separately during tree construction, with a theoretical analysis of GBDT's learning objective. Secondly, we reveal a problem when directly applying GBDT in MTL, i.e., the negative task gain problem. Finally, we propose a novel split method for GBDT in MTL based on the task gain statistics, named task-wise split, as an alternative to standard feature-wise split to overcome the mentioned negative task gain problem. Extensive experiments on a large-scale real-world diabetes dataset and a commonly used benchmark dataset demonstrate TSGB achieves superior performance against several state-of-the-art methods. Detailed case studies further support our analysis of negative task gain problems and provide insightful findings. The proposed TSGB method has been deployed as an online diabetes risk assessment software for early diagnosis.