Multi-Unit Directional Measures of Association: Moving Beyond Pairs of Words

by   Jonathan Dunn, et al.

This paper formulates and evaluates a series of multi-unit measures of directional association, building on the pairwise ΔP measure, that are able to quantify association in sequences of varying length and type of representation. Multi-unit measures face an additional segmentation problem: once the implicit length constraint of pairwise measures is abandoned, association measures must also identify the borders of meaningful sequences. This paper takes a vector-based approach to the segmentation problem by using 18 unique measures to describe different aspects of multi-unit association. An examination of these measures across eight languages shows that they are stable across languages and that each provides a unique rank of associated sequences. Taken together, these measures expand corpus-based approaches to association by generalizing across varying lengths and types of representation.



There are no comments yet.


page 28

page 40

page 42


LAST at SemEval-2021 Task 1: Improving Multi-Word Complexity Prediction Using Bigram Association Measures

This paper describes the system developed by the Laboratoire d'analyse s...

Preference rules for label ranking: Mining patterns in multi-target relations

In this paper we investigate two variants of association rules for prefe...

Frequency vs. Association for Constraint Selection in Usage-Based Construction Grammar

A usage-based Construction Grammar (CxG) posits that slot-constraints ge...

Using Fisher's Exact Test to Evaluate Association Measures for N-grams

To determine whether some often-used lexical association measures assign...

Associative Measures and Multi-word Unit Extraction in Turkish

Associative measures are "mathematical formulas determining the strength...

Stochastic Deep Networks

Machine learning is increasingly targeting areas where input data cannot...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.