Gürültülü Metinlerdeki Sayisal İfadelerin Rakamsal Temsile Dönüştürülmesi

Translated title of the contribution: Conversion of number expressions within noisy text into numerical representation

Gozde Gul Ssahin*, Erdem Emekligil, Secil Arslan, Onur Aggin, Gulsen Eryiggit

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this work, we propose a method to detect spelling of numbers from noisy text and efficiently convert them into their numerical representations. We design a greedy algorithm that detects spanning of numbers and use a fast and memory efficient data structure based on word graphs for digit conversion. Our proposed data structure scored 88.2% on synthetic dataset and overall system resulted with 70.3% on real dataset.

Translated title of the contributionConversion of number expressions within noisy text into numerical representation
Original languageTurkish
Title of host publication2017 25th Signal Processing and Communications Applications Conference, SIU 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509064946
DOIs
Publication statusPublished - 27 Jun 2017
Event25th Signal Processing and Communications Applications Conference, SIU 2017 - Antalya, Turkey
Duration: 15 May 201718 May 2017

Publication series

Name2017 25th Signal Processing and Communications Applications Conference, SIU 2017

Conference

Conference25th Signal Processing and Communications Applications Conference, SIU 2017
Country/TerritoryTurkey
CityAntalya
Period15/05/1718/05/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Fingerprint

Dive into the research topics of 'Conversion of number expressions within noisy text into numerical representation'. Together they form a unique fingerprint.

Cite this