EXplainable Artificial Intelligence (XAI)—From Theory to Methods and Applications

Intelligent applications supported by Machine Learning have achieved remarkable performance rates for a wide range of tasks in many domains. However, understanding why a trained algorithm makes a particular decision remains problematic. Given the growing interest in the application of learning-based models, some concerns arise in the dealing with sensible environments, which may impact users’ lives. The complex nature of those models’ decision mechanisms makes them the so-called “black boxes,” in which the understanding of the logic behind automated decision-making processes by humans is not trivial. Furthermore, the reasoning that leads a model to provide a specific prediction can be more important than performance metrics, which introduces a trade-off between interpretability and model accuracy. Explaining intelligent computer decisions can be regarded as a way to justify their reliability and establish trust. In this sense, explanations are critical tools that verify predictions to discover errors and biases previously hidden within the models’ complex structures, opening up vast possibilities for more responsible applications. In this review, we provide theoretical foundations of Explainable Artificial Intelligence (XAI), clarifying diffuse definitions and identifying research objectives, challenges, and future research lines related to turning opaque machine learning outputs into more transparent decisions. We also present a careful overview of the state-of-the-art explainability approaches, with a particular analysis of methods based on feature importance, such as the well-known LIME and SHAP. As a result, we highlight practical applications of the successful use of XAI.

View this article on IEEE Xplore

Effect of Data Characteristics Inconsistency on Medium and Long-Term Runoff Forecasting by Machine Learning

In the application of medium and long-term runoff forecasting, machine learning has some problems, such as high learning cost, limited computing cost, and difficulty in satisfying statistical data assumptions in some regions, leading to difficulty in popularization in the hydrology industry. In the case of a few data, it is one of the ways to solve the problem to analyze the data characteristics consistency. This paper analyzes the statistical hypothesis of machine learning and runoff data characteristics such as periodicity and mutation. Aiming at the effect of data characteristics inconsistency on three representative machine learning models (multiple linear regression, random forest, back propagation neural network), a simple correction/improvement method suitable for engineering was proposed. The model results were verified in the Danjiangkou area, China. The results show that the errors of the three models have the same distribution as the periodic characteristics of the runoff periods, and the correction/improvement based on periodicity and mutation characteristics can improve the forecasting accuracy of the three models. The back propagation neural network model is most sensitive to the data characteristics consistency.

View this article on IEEE Xplore

An Intelligent IoT Sensing System for Rail Vehicle Running States Based on TinyML

Real-time identification of the running state is one of the key technologies for a smart rail vehicle. However, it is a challenge to accurately real-time sense the complex running states of the rail vehicle on an Internet-of-Things (IoT) edge device. Traditional systems usually upload a large amount of real-time data from the vehicle to the cloud for identification, which is laborious and inefficient. In this paper, an intelligent identification method for rail vehicle running state is proposed based on Tiny Machine Learning (TinyML) technology, and an IoT system is developed with small size and low energy consumption. The system uses a Micro-Electro-Mechanical System (MEMS) sensor to collect acceleration data for machine learning training. A neural network model for recognizing the running state of rail vehicles is built and trained by defining a machine learning running state classification model. The trained recognition model is deployed to the IoT edge device at the vehicle side, and an offset time window method is utilized for real-time state sensing. In addition, the sensing results are uploaded to the IoT server for visualization. The experiments on the subway vehicle showed that the system could identify six complex running states in real-time with over 99% accuracy using only one IoT microcontroller. The model with three axes converges faster than the model with one. The model recognition accuracy remained above 98% and 95%, under different installation positions on the rail vehicle and the zero-drift phenomenon of the MEMS acceleration sensor, respectively. The presented method and system can also be extended to edge-aware applications of equipment such as automobiles and ships.

View this article on IEEE Xplore

Deep Learning Applications in Medical Image Analysis

The tremendous success of machine learning algorithms at image recognition tasks in recent years intersects with a time of dramatically increased use of electronic medical records and diagnostic imaging. This review introduces the machine learning algorithms as applied to medical image analysis, focusing on convolutional neural networks, and emphasizing clinical aspects of the field. The advantage of machine learning in an era of medical big data is that significant hierarchal relationships within the data can be discovered algorithmically without laborious hand-crafting of features. We cover key research areas and applications of medical image classification, localization, detection, segmentation, and registration. We conclude by discussing research obstacles, emerging trends, and possible future directions.

View this article on IEEE Xplore

Most Popular Article of 2017: Disease Prediction by Machine Learning Over Big Data From Healthcare Communities

With big data growth in biomedical and healthcare communities, accurate analysis of medical data benefits early disease detection, patient care, and community services. However, the analysis accuracy is reduced when the quality of medical data is incomplete. Moreover, different regions exhibit unique characteristics of certain regional diseases, which may weaken the prediction of disease outbreaks. In this paper, we streamline machine learning algorithms for effective prediction of chronic disease outbreak in disease-frequent communities. We experiment the modified prediction models over real-life hospital data collected from central China in 2013-2015. To overcome the difficulty of incomplete data, we use a latent factor model to reconstruct the missing data. We experiment on a regional chronic disease of cerebral infarction. We propose a new convolutional neural network (CNN)-based multimodal disease risk prediction algorithm using structured and unstructured data from hospital. To the best of our knowledge, none of the existing work focused on both data types in the area of medical big data analytics. Compared with several typical prediction algorithms, the prediction accuracy of our proposed algorithm reaches 94.8% with a convergence speed, which is faster than that of the CNN-based unimodal disease risk prediction algorithm.

View this article on IEEE Xplore