Automated Amharic News Categorization Using Deep Learning Models

Endalie, Demeke; Haile, Getamesay

doi:10.1155/2021/3774607

Cited by 16 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…N denotes the number of documents. If you want to calculate the TF-IDF score of an individual phrase inside an entire document, you can do so by multiplying the total number of documents in the document by the frequency of that term (DF) [ 21 ]. Multilayer.…”

Section: Proposed Methodologymentioning

confidence: 99%

Analysing Hate Speech against Migrants and Women through Tweets Using Ensembled Deep Learning Model

Hasan

Sharma²,

Khan

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Twitter’s popularity has exploded in the previous few years, making it one of the most widely used social media sites. As a result of this development, the strategies described in this study are now more beneficial. Additionally, there has been an increase in the number of people who express their views in demeaning ways to others. As a result, hate speech has piqued interest in the subject of sentiment analysis, which has developed various algorithms for detecting emotions in social networks using intuitive means. This paper proposes the deep learning model to classify the sentiments in two separate analyses. In the first analysis, the tweets are classified based on the hate speech against the migrants and the women. In the second analysis, the detection is performed using a deep learning model to organise whether the hate speech is performed by a single or a group of users. During the text analysis, word embedding is implemented using the combination of deep learning models such as BiLSTM, CNN, and MLP. These models are integrated with word embedding methods such as inverse glove (global vector), document frequency (TF-IDF), and transformer-based embedding.

show abstract

Section: Proposed Methodologymentioning

confidence: 99%

Analysing Hate Speech against Migrants and Women through Tweets Using Ensembled Deep Learning Model

Hasan

Sharma²,

Khan

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…Using fastText to produce text vectors and a Convolutional Neural Network (CNN) to automatically extract features, the research [16] provides a deep learning model for Amharic news document categorization. The suggested method outperformed popular machine learning algorithms including SVM, MLP, DT, XGB, and RF on a dataset consisting of six types of news articles, with a classification accuracy of 93.79 percent.…”

Section: Related Workmentioning

confidence: 99%

Unifying Sentence Transformer Embedding and Softmax Voting Ensemble for Accurate News Category Prediction

2023

View full text Add to dashboard Cite

The study focuses on news category prediction and investigates the performance of sentence embedding of four transformer models (BERT, RoBERTa, MPNet, and T5) and their variants as feature vectors when combined with Softmax and Random Forest using two accessible news datasets from Kaggle. The data are stratified into train and test sets to ensure equal representation of each category. Word embeddings are generated using transformer models, with the last hidden layer selected as the embedding. Mean pooling calculates a single vector representation called sentence embedding, capturing the overall meaning of the news article. The performance of Softmax and Random Forest, as well as the soft voting of both, is evaluated using evaluation measures such as accuracy, F1 score, precision, and recall. The study also contributes by evaluating the performance of Softmax and Random Forest individually. The macro-average F1 score is calculated to compare the performance of different transformer embeddings in the same experimental settings. The experiments reveal that MPNet versions v1 and v3 achieve the highest F1 score of 97.7% when combined with Random Forest, while T5 Large embedding achieves the highest F1 score of 98.2% when used with Softmax regression. MPNet v1 performs exceptionally well when used in the voting classifier, obtaining an impressive F1 score of 98.6%. In conclusion, the experiments validate the superiority of certain transformer models, such as MPNet v1, MPNet v3, and DistilRoBERTa, when used to calculate sentence embeddings within the Random Forest framework. The results also highlight the promising performance of T5 Large and RoBERTa Large in voting of Softmax regression and Random Forest. The voting classifier, employing transformer embeddings and ensemble learning techniques, consistently outperforms other baselines and individual algorithms. These findings emphasize the effectiveness of the voting classifier with transformer embeddings in achieving accurate and reliable predictions for news category classification tasks.

show abstract

“…Machine learning suggests that, if properly trained, systems can identify patterns, learn from data, and make decisions with little or no human intervention [13]. The support vector machine, K-nearest neighbors (KNN), DT, and random forest (RF) are the most widely used machine-learning algorithms [14].…”

Section: Learning Modelmentioning

confidence: 99%

Analysis and Detection of Road Traffic Accident Severity via Data Mining Techniques: Case Study Addis Ababa, Ethiopia

Endalie,

Abebe

2023

Mathematical Problems in Engineering

Self Cite

View full text Add to dashboard Cite

Around the world, road traffic accidents are the leading cause of serious injuries and deaths. Ethiopia is one of the countries that suffer the most from traffic accidents. Every government in every country wants to keep its citizens safe from accidents. To keep people safe from accidents, it is necessary to conduct a detailed analysis of the factors that contribute to high-severity accidents and deaths. As a result, we developed a data mining algorithm-based road traffic accident severity analysis for the Addis Ababa subcity in this study. The longest frequent factors in the dataset were generated using the Apriori algorithm. The Apriori algorithm generates the most frequent factors as sex: male, driver–vehicle relationship: employee, weather condition: normal, pedestrian movement: not a pedestrian, road surface type: asphalt, and accident severity: high severity, with 42.21% and 84.35% support and confidence, respectively. In addition, we created an accident severity level predictive model using a support vector machine. The predictive model has an accuracy of 85%. The proposed predictive model outperforms other well-known predictive models, such as K-nearest neighbors, decision trees, and random forests. As a result, when making decisions or policies in Ethiopia, the government or private organizations should consider the association of factors that lead to serious severe accidents.

show abstract

Automated Amharic News Categorization Using Deep Learning Models

Cited by 16 publications

References 15 publications

Analysing Hate Speech against Migrants and Women through Tweets Using Ensembled Deep Learning Model

Analysing Hate Speech against Migrants and Women through Tweets Using Ensembled Deep Learning Model

Unifying Sentence Transformer Embedding and Softmax Voting Ensemble for Accurate News Category Prediction

Analysis and Detection of Road Traffic Accident Severity via Data Mining Techniques: Case Study Addis Ababa, Ethiopia

Contact Info

Product

Resources

About