This item is available under a Creative Commons License for non-commercial use only
The thesis aims to take the ﬁrst step towards automated extraction of the information found in book reviews, by using machine learning tools to assign a label of ﬁction or non ﬁction to the text. The thesis makes use of neural networks and performs experiments around architecture, hyper-parameters and text processing from which an optimized model is produced. The thesis enjoys certain successes; it was possible to match the state of the art achieved by (Kim, 2014) and computation was sped up considerably from the default to the optimized model by 13.8 seconds per 50 steps. Further it is conﬁrmed by the thesis that labelling a sequence as ﬁction or non ﬁc tion can be performed most accurately with LSTM architectures and that contrary to (Reimers & Gurevych, 2017) every considered hyper parameter had a considerable impact on results.
Manger, Clement (2018), that seems made up: deep learning classifiers for fiction & non fiction book reviews . Masters dissertation, DIT, 2018.