AutoML Strategy Based on Grammatical Evolution: A Case Study about Knowledge Discovery from Text

Abstract

The process of extracting knowledge from natural language text poses a complex problem that requires both a combination of machine learning techniques and proper feature selection. Recent advances in Automatic Machine Learning (AutoML) provide effective tools to explore large sets of algorithms, hyperparameters and features to find out the most suitable combination of them. This paper proposes a novel AutoML strategy based on probabilistic grammatical evolution, which is evaluated on the health domain by facing the knowledge discovery challenge in Spanish text documents. Our approach achieves state-ofthe-art results and provides interesting insights into the best combination of parameters and algorithms to use when dealing with this challenge. Source code is provided for the research community.

Tracing and Visualizing Human-ML/AI Collaborative Processes through Artifacts of Data Work

“…We also observed that additional datasets are recruited in the preparation stage to further benchmark model performance [30,119] or evaluate its robustness. Augmentations to the data can include human-supplied semantic annotations [25]. We observed that the preparation stage is still largely dominated by the activities of a single human or multiple humans working together.…”

“…We separate interactive processes into the artifacts of the graphical user interface and the user themselves. Elements of the user interface include bookmarked or saved insights [16,107], annotations [16,25,87]. Humans can also trigger or modify automated processes [13] across data science processes.…”

AutoSmart: An Efficient and Automatic Machine Learning Framework for Temporal Relational Data

“…It is widely believed that data and features determine the upper bound of machine learning, and models and algorithms can only approach this limit [10] -this suggests that the performance of a machine learning algorithm heavily depends on the quality of input features and the effectiveness of feature engineering. While recent developments in automatic processing of images [31], texts [7], and signals [14] by deep learning methods, feature engineering for relational data remains iterative, human-intuition driven, and hence, time-consuming. Generally, feature engineering for relational data includes feature generation and feature selection.…”

Recent Trends in Deep Learning Based Open-Domain Textual Question Answering Systems

“…Due to the complex and computationally expensive deep learning models, automated machine learning (AutoML) technologies have aroused widespread concern on hyperparameter optimization and neural architecture search methods [112]- [114]. However, there is little research about AutoML acceleration for the open-domain textual QA system.…”