no code implementations • EMNLP 2020 • Husam Quteineh, Spyridon Samothrakis, Richard Sutcliffe
In this paper we propose a novel data augmentation approach where guided outputs of a language generation model, e. g. GPT-2, when labeled, can improve the performance of text classifiers through an active learning process.
no code implementations • COLING 2022 • Husam Quteineh, Spyridon Samothrakis, Richard Sutcliffe
To overcome this challenge, we present a novel approach where knowledge can be distilled from a teacher model to a student model through the generation of synthetic data.
no code implementations • 31 Dec 2023 • Richard Sutcliffe
Third, we define the methods used to embody personality in a CA, and review recent models using them.
no code implementations • 20 Jul 2023 • Ephrem Afele Retta, Richard Sutcliffe, Jabar Mahmood, Michael Abebe Berwo, Eiad Almekhlafi, Sajjad Ahmed Khan, Shehzad Ashraf Chaudhry, Mustafa Mhamed, Jun Feng
For English, German and Urdu we use the existing RAVDESS, EMO-DB and URDU datasets.
1 code implementation • 29 Jan 2022 • Mustafa Mhamed, Richard Sutcliffe, Xia Sun, Jun Feng, Eiad Almekhlafi, Ephrem A. Retta
This SCM+MMA model is applied to SudSenti2 and SudSenti3 with accuracies of 92. 75% and 84. 39%.
no code implementations • 20 Jan 2022 • Ephrem A. Retta, Richard Sutcliffe, Eiad Almekhlafi, Yosef K. Enku, Eyob Alemu, Tigist D. Gemechu, Michael A. Berwo, Mustafa Mhamed, Jun Feng
In this paper, we create EMIR, the first-ever Music Information Retrieval dataset for Ethiopian music.
no code implementations • 7 Jan 2022 • Ephrem A. Retta, Eiad Almekhlafi, Richard Sutcliffe, Mustafa Mhamed, Haider Ali, Jun Feng
This was done by training two VGGb SER models on ASED, one using Mel-spectrograms and the other using MFCC.
no code implementations • LREC 2016 • Nikolaos Katris, Richard Sutcliffe, Theodore Kalamboukis
The SMT systems were KantanMT, a cloud-based machine translation (MT) platform, and Moses, an open-source MT application.
no code implementations • LREC 2012 • Anselmo Pe{\~n}as, Eduard Hovy, Pamela Forner, {\'A}lvaro Rodrigo, Richard Sutcliffe, Corina Forascu, Caroline Sporleder
The evaluation architecture is completely multilingual: test documents, questions, and their answers are identical in all the supported languages.