A Survey on Topic-Based Visual Text Analytics

Visual text analytics is a multi-discipline research area involving text mining,computer graphics and imaging,and human computer interaction.It helps users analyze a large amount of text documents interactively and visually to provide timely feedback,find anomalies,detect patterns,and gain insight.It has been used in many fields such as homeland security,business intelligence,and financial analysis and has received much attention from academia,industry,and government.This paper introduces a general pipeline of visual text analytics,followed by a survey of existing text mining and visualization techniques.We finally suggest a few potential research challenges,namely,the challenges of scalability,uncertainty,heterogeneity,and user interactions.