Special Issue "Deep Learning and Semantic Technologies"

A special issue of Algorithms (ISSN 1999-4893).

Deadline for manuscript submissions: 15 December 2018

Special Issue Editors

Guest Editor
Dr. George Fazekas

School of Electronic Engineering and Computer Science, Queen Mary University of London, London E1 4FZ, UK
Website | E-Mail
Interests: semantic audio; music information retrieval; semantic web; music emotion recognition; deep learning
Guest Editor
Prof. Dr. Robert Stevens

School of Computer Science, University of Manchester, Oxford Rd, Manchester M13 9PL, UK
Website | E-Mail
Interests: bioinformatics; bio-ontologies; accessibility; visual disability

Special Issue Information

Dear Colleague,

Sustained increase in computational capacity, advances in training and optimisation techniques and the availability of big data caused a resurgence of interest in neural networks. Deep learning opened new avenues in information extraction and processing in a wide range of application domains, including natural language processing, audio and visual object recognition and synthesis, bioinformatics, genomics, health informatics, recommendation systems and many other areas where learning effective representations from raw data or recognising small patterns amid large variations in data is beneficial. At the same time, semantic technologies including ontologies provide a well-established mechanism for structured knowledge representation and inference. They allow domain experts to construct and maintain knowledge bases, often without training data, which may be used in high-level decision-making procedures. These approaches can be distinctly complementary. They may facilitate solving problems where very complex decisions are needed, where large datasets are not yet available, or when expert knowledge can augment big data analytics. Deep learning provides the state-of-the-art in converting raw data into symbols that may be manipulated using logic. In this Special Issue, we invite original research papers and reviews related to the combination of these techniques, including new paradigms for complex reasoning over semantic structures and applications where deep learning and semantic technologies are used in tandem.

Topics of interest include but are not limited to the following:

* Ontology structure and content learning from text and media

* Ontology matching and evaluation using deep neural networks

* Named Entity Recognition and term disambiguation using e.g. word embeddings or enhanced by using knowledge representations

* Using ontologies as priors for deep neural network training

* Learning neural networks from knowledge graphs

* Ontology learning from non-textual data (e.g. music signals, social networks, graph signals etc.)

* Deep Learning for ontology reasoning

* Recurrent and memory networks for complex inference

* Statistical Relational Learning and Reasoning

* Semantic deep mining and knowledge completion using big data analytics

* Applications in Semantic Web, biomedical research, media, audio, video, music, recommendation systems, intelligent user interfaces, broadcasting, manufacturing, etc. 

Dr. George Fazekas
Prof. Dr. Robert Stevens
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Algorithms is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 850 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Published Papers (1 paper)

View options order results:
result details:
Displaying articles 1-1
Export citation of selected articles as:

Research

Open AccessArticle Bidirectional Grid Long Short-Term Memory (BiGridLSTM): A Method to Address Context-Sensitivity and Vanishing Gradient
Algorithms 2018, 11(11), 172; https://doi.org/10.3390/a11110172
Received: 10 July 2018 / Revised: 24 September 2018 / Accepted: 29 October 2018 / Published: 30 October 2018
PDF Full-text (1669 KB) | HTML Full-text | XML Full-text
Abstract
The Recurrent Neural Network (RNN) utilizes dynamically changing time information through time cycles, so it is very suitable for tasks with time sequence characteristics. However, with the increase of the number of layers, the vanishing gradient occurs in the RNN. The Grid Long
[...] Read more.
The Recurrent Neural Network (RNN) utilizes dynamically changing time information through time cycles, so it is very suitable for tasks with time sequence characteristics. However, with the increase of the number of layers, the vanishing gradient occurs in the RNN. The Grid Long Short-Term Memory (GridLSTM) recurrent neural network can alleviate this problem in two dimensions by taking advantage of the two dimensions calculated in time and depth. In addition, the time sequence task is related to the information of the current moment before and after. In this paper, we propose a method that takes into account context-sensitivity and gradient problems, namely the Bidirectional Grid Long Short-Term Memory (BiGridLSTM) recurrent neural network. This model not only takes advantage of the grid architecture, but it also captures information around the current moment. A large number of experiments on the dataset LibriSpeech show that BiGridLSTM is superior to other deep LSTM models and unidirectional LSTM models, and, when compared with GridLSTM, it gets about 26 percent gain improvement. Full article
(This article belongs to the Special Issue Deep Learning and Semantic Technologies)
Figures

Figure 1

Back to Top