Narrative Collection and Analysis with StoryGarden

Stories can reveal what is really happening in your community or organization. StoryGarden assists in the collection and analysis of pertinent narratives.

SOMA's sister company, StoryGarden Media Inc. assists clients who wish to uncover the trends and emergent issues concerning a community through an analysis of its stories. The company uses proprietary tools and methods for gathering and analyzing stories contributed by the members of a specified community. The content of the stories, along with some associated survey questions, are processed using a methodology pioneered by Dave Snowden and licensed by his company Cognitive Edge.

The Process

Once objectives and goals have been clarified and the constituents of a specific community identified, StoryGarden works with a client to develop a strategy to motivate the contribution of stories from the constituents.

Demographic information is collected:
  • by asking questions of the story contributor.
  • is based on interests of system researchers/administrators.

Semantic search tools:

  • instantly recognize and report the use of specified words and phrases (including synonyms and related words) in the narrative content of the stories.
  • recognize and report specified concepts found in the narratives.
  • group narratives:
  • that contain similar words, phrases, names, concepts
  • based on inferred sentiment (i.e. pro or con, satisfaction – dissatisfaction etc.)

Concept Framework

    • is the central, organizing system for the information contained in the narrative collection that enables the machine recognition of concepts used in the narrative collection.
    • organizes and classifies the concepts, vocabulary, and entities of a particular field. (ie. health care, global warming, poverty etc.)
    • is created by developing a set of definitions of vocabulary used in discourse by lay people and experts in a community of interest. In the IT world, these terms are used to associate classes, relations, functions and objects in the correct context using the common interpretation therein. This is often referred to as an “ontology”.

Sense–Making Process

  • allows contributors' voices, values and concerns to emerge using statistically valid methods.
  • uncovers the presence and weight of values and concerns around specific topics/issues determined to be important by researchers/administrators.
  • This is done by a questionnaire presented at the time of narrative submission.
  • It allows the narrative contributor to indicate the presence and importance of attributes and values in a narrative.
  • this self-signification provides highly accurate representation of the intention of a contributor.

The benefits include:

  • the ability to cluster stories around issues and provide statistical analyses
  • the visualization of the distribution of stories on various parameters
  • the identification of emergent patterns including outliers (narratives that are not part of the centre of distribution) thus indicating possible future trends/issues.
  • scalability to any size of narrative collection without extraordinary technology

Analytical Tools

  • Various statistical analyses are supported by the software systems.
  • narratives contributed by community members
  • associated demographic data
  • values and attitudes held by narrative contributors


  • using demographic information
  • using semantic search tools
  • using concept framework created by system administrators
  • using sense making process to understand narrative intent of story contributor
  • using other analytical tools


    • The narrative collection can be shown in network form and subset clusters can be selected for exploration based on attributes determined by user.
    • Display and navigation by related links allows intuitive “browsing” through the collection. Because attributes can be chosen by the user, only the relevant stories are selected for display.
    • Various visualization tools have been developed for the analysis of the narrative collection.