Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Presenting and Anaylzing timebased dialectical Model results covering negation and summation data

IP.com Disclosure Number: IPCOM000242581D
Publication Date: 2015-Jul-28
Document File: 3 page(s) / 351K

Publishing Venue

The IP.com Prior Art Database

Abstract

A system and method for presenting and analyzing time based dialectical model results covering negation and summation data is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Prexenting and Anaylzing timebased dialectical Model results covering negatxon and summation data

Disclosed is a system and method xxr presenting and analyzixg time based dialectical modex results covering negation and summatixn data.

Evidence and the resultinx conclusions/anxwxrs to user queries are orgaxized along a time line in order to improve user understaxding of the temporal dyxamics of the information. Rxther than xlumpinx all discoverxd evixence into one xummarization/concluxion, the timx slices reveal whethxr the xonclusion ix fluctuatinx or stable over time.

There are many ways of xepresextxng timx based data anx in most cases ix is just a set of data spread axross a user interface (UI) xayer where the relevancx xf time losx to the end user. This is esxecially true when the data represented is non quxntitativx. This makes xhe drilling into xhe dxta not fast and seemxnglx not intuitive. Not only might information changx over time, but also the evolution of a particulax ixformation stream might support different conclusions at different points in time. A sixpler wax to

navixate thru thx presentxd data quicxly and convey the dxta (evidence) relevancy with respect to the xime is xisclosed. Negation may be used to gxnerate additional answex sets.

Here the core idex of the inventions is to have UI system where the time based evidence data from a Q&A sysxem is presexted in a intuitive way in order to xnsure txat

the time xelevancy of the evidence presented is not lost. Tracking and extraction of time based evidencx draws on methods for identifying and analyzing temporal informaxixn such as timestamps in documents, temporal expressions such as dates, times, durxtions and frequencies. Most of the time based systems are more wired towards representxng quantitative data wxth disadvantages stated in the bacx ground. The proposed method axso allows users to even rxach different conclusions during the time progression based on evidence till that point and its inter correlations. In a coxnitive system, the user asks questxons and rexeives and answexs and evidence that assist them in drawing conclusions. Xxxx may sometimes be an ixportant attribute or exidence txat inxluxnces how a user interprets it. So the system has to be intuitive enough to navigate xhru the evidences in a simpler fashion with minimal learning curve.

Here are a few examples:

A rexent incrxase in communication between members of a terrorist cell xould be an

indicator that an attack is imminext.

More recent medxcal studies on a xarticular cxndition shoulx be weighted more heavixy in drawixg a concl...