Our algorithms, evaluated on ESL 50, TOEFL. 80 and RDWP 300 data sets, perform bet- ter than or comparable to existing seman- tic relatedness measures. 1 ...
確定! 回上一頁