Heatmaps have been instrumental in helping understand deep network decisions,
and are a common approach for Explainable AI (XAI). While significant progress
has been made in enhancing the informativeness and accessibility of heatmaps,
heatmap analysis is typically very subjective and limited to domain experts. As
such, developing automatic, scalable, and numerical analysis methods to make
heatmap-based XAI more objective, end-user friendly, and cost-effective is
vital. In addition, there is a need for comprehensive evaluation metrics to
assess heatmap quality at a granular level.

热力图在解释人工智能（XAI）中起到了重要的作用，这篇论文探讨了提高热力图信息性和可访问性的方法，并提出了开发自动化、可扩展和数值分析方法以使热力图 XAI 更客观、用户友好和成本效益的需求，同时也需要全面的评估指标来评估热力图质量。

热图的基于部件的定量分析

Part-based Quantitative Analysis for Heatmaps

The sequence of visits and procedures performed by the patient in the health
system, also known as the patient's pathway or trajectory, can reveal important
information about the clinical treatment adopted and the health service
provided. The rise of electronic health data availability made it possible to
assess the pathways of a large number of patients. Nevertheless, some
challenges also arose concerning how to synthesize these pathways and how to
mine them from the data, fostering a new field of research. The objective of
this review is to survey this new field of research, highlighting
representation models, mining techniques, methods of analysis, and examples of
case studies.

本综述旨在调查基于患者路径或轨迹的新型研究领域，重点关注路径的表示模型、挖掘技术、分析方法和案例研究，并讨论了采集大量患者的电子医疗数据的挑战。

患者路径建模与挖掘：范围审视

Modelling and Mining of Patient Pathways: A Scoping Review

The field of natural language processing has seen impressive progress in
recent years, with neural network models replacing many of the traditional
systems. A plethora of new models have been proposed, many of which are thought
to be opaque compared to their feature-rich counterparts. This has led
researchers to analyze, interpret, and evaluate neural networks in novel and
more fine-grained ways. In this survey paper, we review analysis methods in
neural language processing, categorize them according to prominent research
trends, highlight existing limitations, and point to potential directions for
future work.

本综述论文介绍了自然语言处理领域中使用神经网络模型替代传统系统的进展情况及其分析方法，并根据主要研究趋势进行分类，指出现有限制，并提出未来工作的潜在方向。

神经语言处理中的分析方法：一项调查

Analysis Methods in Neural Language Processing: A Survey

In recent years, deep neural networks have revolutionized many application
domains of machine learning and are key components of many critical decision or
predictive processes. Therefore, it is crucial that domain specialists can
understand and analyze actions and pre- dictions, even of the most complex
neural network architectures. Despite these arguments neural networks are often
treated as black boxes. In the attempt to alleviate this short- coming many
analysis methods were proposed, yet the lack of reference implementations often
makes a systematic comparison between the methods a major effort. The presented
library iNNvestigate addresses this by providing a common interface and
out-of-the- box implementation for many analysis methods, including the
reference implementation for PatternNet and PatternAttribution as well as for
LRP-methods. To demonstrate the versatility of iNNvestigate, we provide an
analysis of image classifications for variety of state-of-the-art neural
network architectures.

iNNvestigate 是一个处理深度神经网络的 Python 库，提供了多种分析方法和实现，以便领域专家能够理解和分析复杂神经网络结构，本文利用 iNNvestigate 进行了多种最新神经网络图像分类的分析。