On Injecting Entropy-Like Features into Deep Neural Networks for Content Relevance Assessment
TPNC (2021)
Research topics:
Neural NetworksAbstract
This paper describes in details an innovative technique of injection of a global (or generally large-scale) quality measure into a deep neural network (DNN) in order to compensate for the tendency of DNNs to found the resulting classification virtually from a superposition of local neighbourhood transformations and projections. We used a state probability-like feature as the global quality measure and injected it into a DNN-based classifier deployed in a specific task of determining which parts of a web page are of certain interest for further processing by NLP techniques. Our goal was to decompose web sites of various internet discussion forums to useful content, i.e. the posts of users, and useless content, i.e. forum graphics, menus, banners, advertisements, etc.
Authors
BibTex
@inproceedings{sido2021injecting, title={On Injecting Entropy-Like Features into Deep Neural Networks for Content Relevance Assessment}, author={Sido, Jakub and Ek{\v{s}}tein, Kamil and Pra{\v{z}}{\'a}k, Ond{\v{r}}ej and Konop{\'\i}k, Miloslav}, booktitle={International Conference on the Theory and Practice of Natural Computing}, pages={59--68}, year={2021}, organization={Springer} }
