SLIIT >
Miscellaneous/Other Publication >
Miscellaneous/Information Technology >

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/327

Title: Intelligent Document Writer
Authors: De Silva, D.I.
Samaratunge, S. A. U.
Srirankan, K.
Keywords: Speech recognition
reference management
grammar checker
Issue Date: 2016
Publisher: ICCTICT 2016
Abstract: With the advancement in technology now it has come to an era where a vast majority of people opting to use computers for their day to-day life activities. However, in terms of document writing most of the people struggle to create appropriate documents according to defined formats in a shorter period of time. In addition, people with less information technology literacy find it extremely difficult to create documents by themselves using existing software. The proposed system acts as a solution to this problem. It allows users to operate most of the features that are helpful in creating documents through their voice. In addition, it will automatically create documents based on a user given title, identify almost all the punctuation issues in a document and correct them automatically, suggest related words while a user is speaking or typing, search for references based on the user’s input and list and cite them accordingly, allow users to easily and quickly place the images from the internet in appropriate places in the document file, and will read the documents according to the user’s desires. Through this application users will be able to create document easily, quickly and conveniently. To check the accuracy level of recognizing the pronunciation accents of the Sri Lankan people, 50 people who resided in different locations were chosen randomly. For that sample, the system managed to achieve an accuracy level of 75 percent. Correcting grammar mistake feature achieved an 80 percent accuracy. Reference management feature was tested against 150 different documents and achieved 70 percent accuracy. The feature which was used for easy handling of images was also tested was accuracy using 50 different documents and it managed to achieve 85 percent accuracy.
URI: http://hdl.handle.net/123456789/327
Appears in Collections:Miscellaneous/Information Technology

Files in This Item:

File Description SizeFormat
Mr. Dilshan and Mr. Udara.pdf335.1 kBAdobe PDFView/Open
View Statistics

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Valid XHTML 1.0! DSpace Software Copyright © 2002-2010  Duraspace - Feedback