matcher-pcx-synomia 1.1.2

Extract themes from verbatims

Matcher-pcx-synomia

Extract themes from verbatims.

Version 1.1.2

Release date: 2019-09-18

Getting Started

pip install ...

Prerequisites

• flashtext: $ pip install flashtext

Usage

• You would need a lexicon.txt with the following structure:

manque de place=><manque de place;place;-1< personnel désagréable=><personnel désagréable;personnel;-1< train direct=><train direct;train;1<

• Init matcher:

themes_matcher = matcher.ThemesMatcher('lexicon.txt')

• Extract themes from verbatims:

verbatims = ["Le confort et la propreté","Rapidité (Train Direct), mais personnel désagréable","c'est catastrophique… retards chroniques, personnel désagréable","Le manque de place ds le tgv"] vbs2matches = themes_matcher.match(verbatims) print(vbs2matches) {'Rapidité (Train Direct), mais personnel désagréable': [['train direct', 'train', '1'], ['personnel désagréable', 'personnel', '-1']], 'Le confort et la propreté': [], 'Le manque de place ds le tgv': [['manque de place', 'place', '-1']], "c'est catastrophique… retards chroniques, personnel désagréable": [['personnel désagréable', 'personnel', '-1']]}

Authors

• Synomia