Ephrem Retta

Dr. Ephrem Retta

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2021 in Computational Intelligence and Neuroscience
Verified
Journal Article
Open access
PDF file
MMMustafa mhamed Mhamed
Mustafa mhamed Mhamed
RSRichard Sutcliffe
Richard Sutcliffe
XSXia Sun
Xia Sun
JFJun Feng
Jun Feng
EAeiad almekhlafi
eiad almekhlafi
EREphrem Retta
Ephrem Retta
Sentiment analysis is an essential process which is important to many natural language applications. In this paper, we apply two models for Arabic sentiment analysis to the ASTD and ATDFS datasets, in both 2‐class and multiclass forms. Model MC1 is a 2‐layer CNN with global average pooling, followed by a dense layer. MC2 is a 2‐layer CNN with max pooling, followed by a BiGRU and a dense layer. On the difficult ASTD 4‐class task, we achieve 73.17%, compared to 65.58% reported by Attia et al., 2018. For the easier 2‐class task, we achieve 90.06% with MC1 compared to 85.58% reported by Kwaik et al., 2019. We carry out experiments on various data splits, to match those used by other researchers. We also pay close attention to Arabic preprocessing and include novel steps not reported in other works. In an ablation study, we investigate the effect of two steps in particular, the processing of emoticons and the use of a custom stoplist. On the 4‐class task, these can make a difference of up to 4.27% and 5.48%, respectively. On the 2‐class task, the maximum improvements are 2.95% and 3.87%.

31 January 2023 in ACM Transactions on Asian and Low-Resource Language Information Processing
Verified
Journal Article
Open access
EREphrem Retta
Ephrem Retta
EAeiad almekhlafi
eiad almekhlafi
RSRichard Sutcliffe
Richard Sutcliffe
MMMustafa mhamed Mhamed
Mustafa mhamed Mhamed
HAHaider Ali
Haider Ali
JFJun Feng
Jun Feng
In this article we present the Amharic Speech Emotion Dataset (ASED), which covers four dialects (Gojjam, Wollo, Shewa, and Gonder) and five different emotions (neutral, fearful, happy, sad, and angry). We believe it is the first Speech Emotion Recognition (SER) dataset for the Amharic language. Sixty-five volunteer participants, all native speakers of Amharic, recorded 2,474 sound samples, 2 to 4 seconds in length. Eight judges (two for each dialect) assigned emotions to the samples with high agreement level (Fleiss kappa = 0.8). The resulting dataset is freely available for download. Next, we developed a four-layer variant of the well-known VGG model, which we call VGGb. Three experiments were then carried out using VGGb for SER, using ASED. First, we investigated which features work best for Amharic, FilterBank, Mel Spectrogram, or Mel-frequency Cepstral Coefficient (MFCC). This was done by training three VGGb SER models on ASED, using FilterBank, Mel Spectrogram, and MFCC features, respectively. Four forms of training were tried, standard cross-validation and three variants based on sentences, dialects, and speaker groups. Thus, a sentence used for training would not be used for testing, and the same for a dialect and speaker group. MFCC features were superior under all four training schemes. MFCC was therefore adopted for Experiment 2, where VGGb and three well-known existing models were compared on ASED: RESNet50, AlexNet, and LSTM. VGGb was found to have very good accuracy (90.73%) as well as the fastest training time. In Experiment 3, the performance of VGGb was compared when trained on two existing SER datasets—RAVDESS (English) and EMO-DB (German)—as well as on ASED (Amharic). Results are comparable across these languages, with ASED being the highest. This suggests that VGGb can be successfully applied to other languages. We hope that ASED will encourage researchers to explore the Amharic language and to experiment with other models for Amharic SER.

22 November 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
EREphrem Retta
Ephrem Retta
RSRichard Sutcliffe
Richard Sutcliffe
Jabar MahmoodJabar Mahmood
Jabar Mahmood
MBMICHAEL ABEBE BERWO
MICHAEL ABEBE BERWO
EAeiad almekhlafi
eiad almekhlafi
JFJun Feng
Jun Feng
In a conventional speech emotion recognition (SER) task, a classifier for a given language is trained on a pre-existing dataset for that same language. However, where training data for a language do not exist, data from other languages can be used instead. We experiment with cross-lingual and multilingual SER, working with Amharic, English, German, and Urdu. For Amharic, we use our own publicly available Amharic Speech Emotion Dataset (ASED). For English, German and Urdu, we use the existing RAVDESS, EMO-DB, and URDU datasets. We followed previous research in mapping labels for all of the datasets to just two classes: positive and negative. Thus, we can compare performance on different languages directly and combine languages for training and testing. In Experiment 1, monolingual SER trials were carried out using three classifiers, AlexNet, VGGE (a proposed variant of VGG), and ResNet50. The results, averaged for the three models, were very similar for ASED and RAVDESS, suggesting that Amharic and English SER are equally difficult. Similarly, German SER is more difficult, and Urdu SER is easier. In Experiment 2, we trained on one language and tested on another, in both directions for each of the following pairs: Amharic↔German, Amharic↔English, and Amharic↔Urdu. The results with Amharic as the target suggested that using English or German as the source gives the best result. In Experiment 3, we trained on several non-Amharic languages and then tested on Amharic. The best accuracy obtained was several percentage points greater than the best accuracy in Experiment 2, suggesting that a better result can be obtained when using two or three non-Amharic languages for training than when using just one non-Amharic language. Overall, the results suggest that cross-lingual and multilingual training can be an effective strategy for training an SER classifier when resources for a language are scarce.

29 May 2023 in Engineering Science and Technology, an International Journal
Verified
Journal Article
Open access
MMMustafa mhamed Mhamed
Mustafa mhamed Mhamed
RSRichard Sutcliffe
Richard Sutcliffe
XSXia Sun
Xia Sun
JFJun Feng
Jun Feng
EREphrem Retta
Ephrem Retta

17 February 2021 in Journal of Healthcare Engineering
Verified
Journal Article
Open access
PDF file
HAHaider Ali
Haider Ali
HLHansheng Li
Hansheng Li
EREphrem Retta
Ephrem Retta
IHIMRAN UL HAQ
IMRAN UL HAQ
ZGZhenzhen Guo
Zhenzhen Guo
JFJun Feng
Jun Feng
The breast cancer microscopy images acquire information about the patient’s ailment, and the automated mitotic cell detection outcomes have generally been utilized to ease the massive amount of pathologist’s work and help the pathologists make clinical decisions quickly. Several previous methods were introduced to solve automated mitotic cell count problems. However, they failed to differentiate between mitotic and nonmitotic cells and come up with an imbalance problem, which affects the performance. This paper proposes a Representation Differential Learning Method (RDLM) for mitosis detection through deep learning to detect the accurate mitotic cell area on pathological images. Our proposed method has been divided into two parts: Global bank Feature Pyramid Network (GLB-FPN) and focal loss (FL). The GLB feature fusion method with FPN essentially makes the encoder-decoder pay attention, to further extract the region of interest (ROIs) for mitotic cells. On this basis, we extend the GLB-FPN with a focal loss to mitigate the data imbalance problem during the training stage. Extensive experiments have shown that RDLM significantly outperforms on visualization view and achieves the best performance in quantitative matrices than other proposed approaches on the MITOS-ATYPIA-14 contest dataset. Our framework reaches a 0.692 F1-score. Additionally, RDLM achieves 5% improvements than GLB with FPN in F1-score on the mitosis detection task.

20 July 2023 in arXiv
Verified
Preprint
Open access
PDF file
EREphrem Retta
Ephrem Retta
RSRichard Sutcliffe
Richard Sutcliffe
Jabar MahmoodJabar Mahmood
Jabar Mahmood
MBMICHAEL ABEBE BERWO
MICHAEL ABEBE BERWO
EAEiad Almekhlafi
Eiad Almekhlafi
JFJun Feng
Jun Feng
In a conventional Speech emotion recognition (SER) task, a classifier for a given language is trained on a pre-existing dataset for that same language. However, where training data for a language does not exist, data from other languages can be used instead. We experiment with cross-lingual and multilingual SER, working with Amharic, English, German and URDU. For Amharic, we use our own publicly-available Amharic Speech Emotion Dataset (ASED). For English, German and Urdu we use the existing RAVDESS, EMO-DB and URDU datasets. We followed previous research in mapping labels for all datasets to just two classes, positive and negative. Thus we can compare performance on different languages directly, and combine languages for training and testing. In Experiment 1, monolingual SER trials were carried out using three classifiers, AlexNet, VGGE (a proposed variant of VGG), and ResNet50. Results averaged for the three models were very similar for ASED and RAVDESS, suggesting that Amharic and English SER are equally difficult. Similarly, German SER is more difficult, and Urdu SER is easier. In Experiment 2, we trained on one language and tested on another, in both directions for each pair: Amharic<->German, Amharic<->English, and Amharic<->Urdu. Results with Amharic as target suggested that using English or German as source will give the best result. In Experiment 3, we trained on several non-Amharic languages and then tested on Amharic. The best accuracy obtained was several percent greater than the best accuracy in Experiment 2, suggesting that a better result can be obtained when using two or three non-Amharic languages for training than when using just one non-Amharic language. Overall, the results suggest that cross-lingual and multilingual training can be an effective strategy for training a SER classifier when resources for a language are scarce.