Conference material: "Proceedings of the International Conference on Computer Graphics and Vision “Graphicon” (19-21 September 2023, Moscow)"
Authors:Makarova E.A., Lagerev D.G.
Using Interactive Visualization in the Problem of Feature Extraction from Semi-structured Text Data
Abstract:
The article deals with the visualization of semi-structured text data (SSTD) in order to solve the problems of exploratory analysis and build a model for processing text data for their further use in data analysis models. The problems faced by researchers when adding SSTS to the data analysis model are considered. Existing approaches to visualization of text data for solving various problems of natural language processing are considered. A model of intelligent processing of SSTD and approaches to data transformation within the data processing. A visual model used to visualize the process of transformation of SSTD is based on the Sankey charts. The proposed visual model allows to reduce the expert's time for data processing by increasing the visibility of the process of extracting features from SSTD using interactive visual tools. The developed approach was tested on data from the information system of the employment service.
Keywords:
Text data processing, exploratory analysis, visualization, Sankey chart