--set_pmi_threshold

Switch

--set_pmi_threshold <pmi_value>

Description

This is to be used with --feat_colloc_filter. See more info there.

Argument and Default Value

A float — the minimum PMI (pointwise mutual information) score a multigram must have to be kept as a collocation. Default: 3.0 (dlac.DEF_PMI).

Other Switches

Required Switches:

• --feat_colloc_filter

Example Commands

Using the packaged dla_tutorial dataset (see Differential Language Analysis (DLA) Tutorial):

dlatkInterface.py -d dla_tutorial -t msgs -c user_id -f 'feat$1to3gram$msgs$user_id' \
--feat_colloc_filter --set_pmi_threshold 5.0