SP500215 Incorporating Semantics Within a Connectionist Model and a Vector Processing Model chapter R. Boyd J. Driscoll National Institute of Standards and Technology D. K. Harman word idfof the word Iog10[OCRerr]Nf Document#1 L[OCRerr]comotives pull the trains. and .6 canopy .6 carry .6 Document #2 depart undefined do undefined People meet people under the canopy and within trains. freight .6 from .6 hourly .6 Document #3 leave .6 locomotives .6 Trains carry freight from the station. meet .6 noon .6 people .6 Document #4 pull .6 station .3 Trains leave the station hourly until noon. the 0 trains 0 under .6 Ouery until .6 when undefined When do trains depart the station? within .6 Figure 6. Four Documents and a Query. Figure 8. The i[OCRerr]of Each Word. word frequency category probability depart 1 AMDR 1/4 word number of documents TA[OCRerr] 1/8 the word is in (dJ) do 1 AUSE 1/21 ATh{P 1/21 and 1 TCSE 1/21 canopy 1 TCNV 2/21 carry 1 ThEs 1/21 do 0 TSRC 1/21 depart 0 freight 1 station 1 APOS 3/16 from 1 AORD 1/8 hourly 1 TAMF 1/16 leave 1 TCND 1/8 locomotives 1 TDGR 1/16 meet 1 TSPL 3/16 noon 1 people 1 the 1 pull 1 trains 1 AORD 7/24 station 2 AMDR 1/12 the 4 AMFR 1/12 trains 4 TACM 1/24 under 1 TCNV 1/12 until 1 when 0 when 1 TAMT 1/3 within 1 TnM 2/3 Figure 7. Ust of Words in the Documents and Query. Figure 9. Words in the Query. 296