Machine based expansion of contractions in text in digital media
AssigneeVerint Americas Inc.
InventorIan Beaver
Life of this patent
Estimating the term…
Abstract
As described herein, a system for expanding contractions in electronically stored text includes expanding contractions having only on expanded form. For remaining contractions, a grammar check is performed for all possible expanded forms to determine if an expanded form can be selected based on context and grammar rules. If an expanded form is not evident from the first two steps, all possible expanded forms of the remaining contractions are converted to a vector representation along with the original text. A Word Movers Distance (WMD) for each possible expansion is calculated using the vectors for each possible expansion and the original text. An expanded form is chosen without human intervention based on the grammar score alone or the WMD and the grammar score.
Claims7
A method of expanding contractions in electronically stored text without human intervention, wherein a library of contractions is available electronically comprising:
- identifying a contraction in the electronic text;
- substituting an expanded form of the contraction in the text for the contraction if the library of contractions defines only a single expanded form of the contraction;
- if the library of contractions defines more than one expanded form of the contraction, substituting each expanded form of the contraction in the text and performing a grammar check to provide a grammar score for each expanded form in context of the text and evaluating if only one expanded form is grammatically correct;
- if only one expanded form is grammatically correct, substituting the grammatically correct expanded form in the text for the contraction;
- if the more than one expanded form is grammatically correct or no expanded form is grammatically correct, converting all expanded forms and the contraction to vector representations using a word embedding model and calculating a word movers distance for each expanded form with respect to the electronic text;
- selecting an expanded form for substitution of the contracting in the text based on the grammar score and the word mover's distance calculated;
- and
- substituting the selected expanded form for the contraction in the text.
A method of generating a final text without contractions from an electronically available text without human intervention, comprising:
- (a) accessing the electronically available text;
- (b) analyzing the electronically available text to determine if contractions are present in the electronically available text;
- and
- (c) returning the electronically available text as the final text when no contractions are present after (b).
Dependencies are inferred from the claim text.