Authorship Authentication of Short Messages from Social Networks Using Recurrent Artificial Neural Networks: Massage Batches
Abstract
500 tweets from Twitterare collectedby using the software Nvivo, fromeach of 34 authorsthat meet certain criteria. Dataset consists of 17000 tweets is preprocessed to extract frequencies of 72 features. Since artificial neural networks are more successful distinguishing two classes, for N authors, N×N neural networks are trained for pair wise classification. These experts then organized as N special competing teams (CANNT) to aggregate decisions of these NXN experts. Then to improve the accuracy of author authentication, a novel technique, batch identification is used and up to100% accuracy is achieved.
Keywords
Authorship Authentication;short massages;committee machines;recurrent neural network
Full Text:
PDFDOI: http://dx.doi.org/10.21533/scjournal.v7i1.155
Refbacks
- There are currently no refbacks.
Copyright (c) 2018 Nesibe Merve Demir
ISSN 2233 -1859
Digital Object Identifier DOI: 10.21533/scjournal
This work is licensed under a Creative Commons Attribution 4.0 International License