Word2Vec to Graph

The models trained using data from various corpora released under TS Corpus
Existing models are listed below and more will be added in time.
  • TS Corpus - 1.25 Billion Words: A skipgram model trained by using 1.25 billion words.
  • TS TimeLine Corpus - 700 Million Words: A skipgram model trained over TS TimeLine Corpus
  • TS Corpus - 500 Million Words: A Cbow model trained by using various corpora under TS Corpus
  • Open Subtitles - 242 Million Words: A model trained by using movie subtitles from Open Subtitles
  • Wikipedia TR - 35 Million Words: A skipgram model trained over TS Wikipedia Corpus data set
  • Node + Edges: The first level of nearest neighbors are calculated
  • Node + Edges + Edges: The first level of nearest neighbors and their nearest neighbors are calculated
  • Graphs may be generated by using different gravity models. Please see pyvis Documentation for more information.
  • Forced Atlas: The forceAtlas2Based solver makes use of some of the equations provided by them and makes use of the barnesHut implementation in vis. The main differences are the central gravity model, which is here distance independent, and the repulsion being linear instead of quadratic. Finally, all node masses have a multiplier based on the amount of connected edges plus one.
  • Barnes Hut: BarnesHut is a quadtree based gravity model. It is the fastest. default and recommended solver for non-heirarchical layouts.
  • Repulsion Set the physics attribute of the entire network to repulsion. When called, it sets the solver attribute of physics to repulsion.
  • HRepulsion: This model is based on the repulsion solver but the levels are taken into account and the forces are normalized.
  • Nodes and edges are connected by various attributes. The options are "dynamic", "continuous", "discrete", "diagonalCross", "straightCross", "horizontal", "vertical", "curvedCW", "curvedCCW" and "cubicBezier".
    Please see pyvis Documentation for more information.
    The color for the keyword is set to orange by default and is allowed to change. However, the default edge color, red, could be set one of the following.
  • blue
  • black
  • green
  • orange
  • grey
  • yellow
  • Node - edge connections are highlighted by arrows.
  • To The Node: Arrows are drawn from keyword to the nodes and edges.
  • From the Node: Arrows are drawn from nodes and edges to keyword.
  • Nodes and edges could be drawn by different shapes. The text for the nodes and edges could be in-line or off-line.
      In-Line Shapes
    • Box
    • Text
    • Circle
    • Ellipse
      Off-Line Shapes
    • Diamond
    • Dot
    • Star
    • Triangle
    • Triangle Down
    • Square