15.3 Term Distribution Model

• P_i(k) $\rightarrow$ w_i $B$,(B 1$B$D$N(B document $B$K(B k $B2s=P8=$9$k3NN($r?dDj$7$?(B $B$$(B
• Zipf $B$NK!B'(B
  • 1$B$D$N(B corpus $B$KBP$7$F(B $B$"$kC18l$,2?2s=P8=$9$k$+$N3NN((B
  • $f \cdot r = const$
• no-content word($BHsFbMF8l(B,$BHs=EMW8l(B,keyword$B$K$J$j$K$/$$8l(B) $B$H(B content word ($BFbMF8l(B,$B=EMW8l(B,keyword $B$K$J$j$d$9$$8l(B) $B$N6hJL$r$7$?(B $B$$(B($B3FC18l$K=EMWEY$r3d$jEv$F$?$$(B) $B$=$l$>$l$NC18l$NJ,I[$H(B,$B=EMWEY$N(B $B4X78$O(B?