Deep Learning Based Emotion Detection Model For Amharic Text

Emotions are so important that whenever we need to make a decision, we want to feel other‟srnemotions. This is not only true for individuals but also for organizations. Due to the rapid growthrnof internet peoples expirees their emotions using different social media networks, reviews, blogs,rnonline and so on. The need for finding relevant sources, extracts related sentences with emotion,rnsummarizes them and organize them to useful form is becoming very high. Emotion detectionrncan play an important role in satisfying these needs. The process of emotion detection involvesrncategorizing emotional sentences into predefined categories such as sadness, anger, disgust,rnhappiness, so on based on the emotional terms that appear within the comment. So that it‟srndifficult to manually identifying emotion of a million of users and aggregating them towards arnrapid and efficient decision is quite a challenging task due to the rapid growth of Amharicrnlanguage usage in social media. In this research work, an emotion detection model is proposedrnfor determining the emotion expressed in the Amharic texts or comment.rnIn this study, we proposed deep learning based emotion detection model for Amharic text usingrnCNN with word embedding. The proposed model includes different tasks. The first task is textrnpre-processing which consists of commonly used text pre-processing tasks in many naturalrnlanguage processing applications. We perform text pre-processing in Amharic text and train therndocument using a word embedding in order to generate word embedding model. The embeddingrnresult provides a contextually similar word for every word in the training set then we implementrnour CNN model for emotion classification.rnThe common evaluation metrics such as accuracy, recall, F1 score and precision were used tornmeasure our proposed model performance. Deep learning based emotion detection model forrnAmharic text prototype is developed and used to tests the system performance using the collectedrnAmharic text comments. Finally, this study with four categories (sadness, anger, disgust, andrnhappiness) of classification shows a result of 71.11% accuracy. Also did better when the numberrnof classification is two (positive and negative) shows result of 87.46% accuracy. We alsornevaluate our model using RNN to compare with our CNN model.

Get Full Work

Report copyright infringement or plagiarism