Data Visualization Analysis Based on Explainable Artificial Intelligence: A Survey

Abstract

With the rapid development of computer hardware and big data processing technology, the bottleneck of intelligent analysis of massive data has changed from "how to deal with massive data quickly" to "how to mine valuable information quickly and effectively from massive data". Visualization and visualization analysis based on human visual perception characteristics, combined with data analysis and human-computer interaction and other technologies, use visual charts to deconstruct the knowledge and rules contained in complex data. This technology runs through the whole life cycle of data science, known as the last kilometer in the field of big data intelligence, and has achieved remarkable results in many big data application analysis scenarios. Traditional visualization analysis is extremely dependent on the user's frequent active participation in the whole life cycle of visualization analysis, including data preparation, data conversion, visualization mapping, visual rendering, user interaction, visual analysis and other stages, which require high professional skills of users and low intelligence of the system. Therefore, the traditional visualization analysis mode and systems have the challenges of high threshold of visualization analysis, high cost of data preparation, high latency of interaction response, and low efficiency of interaction mode. Therefore, this paper introduces the application, challenges in visualization based on explainable AI.