PROJECT TITLE :
ThemeDelta: Dynamic Segmentations over Temporal Topic Models
We present ThemeDelta, a visual analytics system for extracting and visualizing temporal trends, clustering, and reorganization in time-indexed textual datasets. ThemeDelta is supported by a dynamic temporal segmentation algorithm that integrates with topic modeling algorithms to spot amendment points where significant shifts in topics occur. This algorithm detects not solely the clustering and associations of keywords in a time period, but conjointly their convergence into topics (teams of keywords) that will later diverge into new teams. The visual representation of ThemeDelta uses sinuous, variable-width lines to point out this evolution on a timeline, utilizing color for classes, and line width for keyword strength. We demonstrate how interaction with ThemeDelta helps capture the rise and fall of topics by analyzing archives of historical newspapers, of U.S. presidential campaign speeches, and of social messages collected through iNeighbors, a web-primarily based social web site. ThemeDelta is evaluated employing a qualitative expert user study involving three researchers from rhetoric and history using the historical newspapers corpus.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here