corpus—based and corpus—driven studies of collocation：ideas and methods.ppt

Pattern Instances V + (art) + (A) + N command devotion, give a loud laugh V + Prep + (art) + (A) + N leap to a sudden conclusion, leap to a decision A+N cogent argument V + Adv take off, take away, sit down I + A dead tired, dead drunk, stark naked [back] * Colligations Collocations DET + N + V This paper describes ART + N + PREP a pair of ADV + ADJ + CONJ + (ART) N as white as snow ART + N + PRE + N a sea of troubles Motive (verb + particle) ambled up, tore across INT + ADJ dead tired [back] 表6.3 Colligations and collocations * 1 deen University but summary and analytical data are held at London. A running series o 2 the implications of these attitudinal data were debatable, some analysts arg 3 olved and the current status of available data INTRODUCTION 1. One of the advantag 4 There is at present no single centralized data file on all individuals living in t 5 s proposals to create centralized data banks containing data on individual 6 disks and tapes) are used to store digital data as well as application programs and ot 7 opment of a completely general distributed data base management system must evolv 8 s well understood, but in a distributed data management system the locking problem 9 ull inventory of the relevant empirical data on the one hand, and the conceptual 10 th improvements largely based on empirical data and practical experience. Today,due t 11 rthwhile discussing all the experimental data obtained, bu one should refer to some 12 has been made with published experimental data which has led to a rationalisation 13 accuracy and the reproducibility of field data A few comparative studies between th 14 are available to perform this task). Input data for this program consist of spacecraft 15 rigger a signal for recording inspection data as soon as the stylus is deflected * 16 This is what Greenbaum calls intuitional data , which he contrasts