AT deleted beginitemize__item_S.tex  almost 9 years ago

Commit id: e3356e290bb8ea4be3ca3a3dde15785fc45c94ce

deletions | additions      

         

\begin{itemize}  \item Syphon module that detects and requires named entities (enwiki titles, dates, \dots) and verbs, rejects if anything extra is present.  \item Pre-processed question is transferred from Syphon to Argus back-end.  \item News titles (and possibly article texts, structured databases) are scanned for relevant data, entailment (yes, no, unknown) is modelled for each.  \item Relative entailment counts (normalized by general frequency of the named entity in the souces) are used to estimate probability.  \item The output is accompanied with a justification --- summary of evidence for or against the system decision.  \end{itemize}         

sectionStructure_of_.tex  sectionHow_Can_We_Kn.tex  sectionA_Concrete_Pr.tex  beginitemize__item_S.tex  sectionImplementatio.tex