Efficient discovery of optimal word-association patterns in large text databases

2000 ◽  
Vol 18 (1) ◽  
pp. 49-60 ◽  
Author(s):  
Shinichi Shimozono ◽  
Hiroki Arimura ◽  
Setsuo Arikawa
1998 ◽  
Vol 83 (3_suppl) ◽  
pp. 1419-1424
Author(s):  
Elaine J. Klionsky ◽  
Judy B. Okawa ◽  
Robert W. Holmstrom ◽  
David E. Silber ◽  
Stephen A. Karp

The present study concerned an attempt to corroborate the classic findings of Rapaport, Gill, and Schafer in distinguishing mental hospital patients from control subjects, using a new objectively scored word-association test. 12 computer-scored objective scales were used to compare groups of 101 mental hospital inpatients ( n = 75, including 37 on a prison ward) and outpatients ( n = 26) and 101 control subjects matched with the patients for sex, age, racial and ethnic status, and education. A stepwise multiple discrimination analysis of the scores on the 12 scales of the test significantly distinguished the groups. Scales weighted most highly were Masochism, Antonyms, and Aggression. Subsequent t tests suggested that control subjects scored higher on Aggression, Self-reference, and Masochism scales, whereas patients scored higher on Rejections (nonresponses to stimulus words). Of these, only Self-reference and Rejections were items identified by Rapaport, et al.


2006 ◽  
Vol 11 (4) ◽  
pp. 411-433 ◽  
Author(s):  
Winnie Cheng ◽  
Chris Greaves ◽  
Martin Warren

Uncovering the extent of word associations and how they are manifested has been an important area of study in corpus linguistics since the 1960s (Sinclair et al. 1970). This paper defines and describes a new way of categorising word association, the concgram, which constitutes all of the permutations of constituency and positional variation generated by the association of two or more words. Concgrams are identified without prior input from the user (other than to set the size of the span) employing a fully automated search that reveals all of the word association patterns that exist in a corpus. This study argues that concgrams represent more fully word associations in a corpus. Most concgrams seem to be non-contiguous, and show both constituency (AB, ACB) and positional (AB, BA) variations. Further studies of concgrams will help in the task of uncovering the full extent of the idiom principle (Sinclair 1987).


1981 ◽  
Vol 24 (2) ◽  
pp. 233-246 ◽  
Author(s):  
Robert Goldfarb ◽  
Harvey Halpern
Keyword(s):  

2013 ◽  
Author(s):  
Katharine Donnelly Adams ◽  
Janet G. Van Hell ◽  
Natasha Tokowicz

Sign in / Sign up

Export Citation Format

Share Document