ELAINE - Narrative Cognition

Autonomous Narrative Discovery

For those who have to read a lot to get narrative from written documents, SiteFocus has the perfect solution. It reduces hours of reading into seconds.

ELAINE is the only AI that works with narrative cognition

ELAINE implements a novel AI that finds casual linkages among connected stories in a narrative. It is the only AI that features narrative cognition. ELAINE discovers while others uses models from pre-training. As most narratives are new and previously unknown, discovery is the only technique that works.

ELAINE is made for tasks that require lots of reading and understanding. Here is how it works:

  • Provide unbiased analysis of natural language documents
  • Deconstruct and identify casual reasoning in narrative
  • Self detect connecting stories
  • Identify symbols that link stories together that formed the narrative
  • Provide traceable reasoning for stories and casual linkages
  • Draw reference from original text to support stories
  • Detect logic in data and provide key context
  • ELAINE taps the logic in narrative and expresses it with a topological diagram


    Example:

    A complex litigation, with voluminous amount of documents, ended in a decision in favor of the API user. What does this mean going forward for software developers? The devil is in the details.

    Narrative Cognition uses symbolic logic to find casual linkages. In this case, Supreme Court applied Fair Use Law in deciding Google v Oracle. Automated Narrative Cognition completed its analysis in seconds on a corpus of 61 pages of legal writing. This is an area of AI that is novel and goes beyond most prevalent AI works such as those offered by BERT or GPT3. The difference is between "discovery" and "pre-trained". When a decision prescribes an opinion that is new and previously unknown, software that relies on ML, DL or RL will not be able to take advantage of "pre-trained" knowledge regardless how large is its knowledge base. Symbolic Logic discovers.

    "Unlike many other computer programs, the value of the copied lines is in significant part derived from the investment of users (here computer programmers) who have learned the API's system. Given these differences, application of fair use here is unlikely to undermine the general copyright protection that Congress provided for computer programs"

    Narrative Cognition Process with ELAINE

  • ELAINE discovers key words as symbols using predicate calculus
  • Symbols are connected to form high-level abstractions (HLA)
  • Behind each HLA is the context and semantics that made up the stories within a narrative
  • An interactive topological diagram is rendered to show symbols connecting HLAs that it discovers. Hovering over each end node will trigger a pop-up text-box showing the corresponding semantic excerpt. Clicking on the node will open up a context sensitive display of the original document pin-pointing the relative context and location of the excerpt.
  • Each HLA is ranked by relevancy according to its semantics posture. Semantics posture is determined by three components - momentum, challenge, and work-in-progress. HLA ranking is derived from the relative significance of semantics according to the author’s composition style.
  • Submit the original document to ELAINE in verbatim, it gives back an interactive WEB report.

    A. Key focus - most relevant excerpts
    B. High Level Abstractions (HLA) identified as momentum, challenge and work-in-progress
    C. Donut graph depicting the proportion of (B)
    D. A table of HLAs ranked by relevancy
    E. Excerpts associated with corresponding HLA
    F. Narrative Diagram linking actors, facts, and stories; a visual navigational diagram with pop-up text box to show corresponding excerpts

    Understanding a complex narrative is always difficult. Using a topological narrative map to deconstruct a narrative makes it possible to gain insights. Leveraging on this technology to mitigate risk can certainly help management to make better decisions. This is especially true for conducting due diligence, such as merger/acquisition, legal and compliance research ...


    Automated Narrative Cognition is central to the success in life and business

    With better narrative cognition:

  • Product team produces better products
  • Development team delivers work product on time
  • CFO manages budgets with ease
  • CEO runs company with more efficiency
  • Meeting covers more subjects with precision
  • Analyst performs due diligent in less time
  • M&A produces better outcome
  • Financial advisor offers more comprehensive research
  • Politician better connects with constituents
  • Economist makes better forecast
  • and more...

  • Benefits

  • Boost productivity without increasing head-count
  • Get up to speed on unfamiliar subject quickly
  • Make decisions with a full deck
  • See the big picture with clarity
  • Detect fallacy and contradiction
  • Inspire alternatives and options in strategic planning

  • To view more ELAINE Narrative Cognition examples, visit www.bizpage.com