Deep Learning versus Traditional Classifiers on Vietnamese Students’ Feedback Corpus

Abstract

Student's feedback is an important source of collecting students' opinions to improve quality of training activities. Implementing sentiment analysis into student feedback data, we can determine sentiments polarities which express all problems in the institution since changes necessary will be applied to improve the quality of teaching and learning. This study focused on the machine learning and natural language processing techniques (Naive Bayes, Maximum Entropy, Long Short-Term Memory, Bi-Directional Long Short-Term Memory) on the Vietnamese Students' Feedback Corpus collected from a university. The final results were compared and evaluated to find the most effective model based on different evaluation criteria. The experimental results show that Bi-Directional Long Short-Term Memory algorithm outperformed than three other algorithms in term of the F1-score measurement with 92.0% on the sentiment classification task and 89.6% on the topic classification task. In addition, we developed a sentiment analysis application analyzing student feedback. The application will help the institution to recognize students' opinions about a problem and identify shortcomings that still exist. With the use of this application, the institution can propose an appropriate method to improve the quality of training activities in the future.• Bi-Directional Long Short-Term Memory (Bi-LSTM)

Lyric-Based Classification of Music Genres Using Hand-Crafted Features

“…Another avenue for future work is the use of deep-learning approaches for multi-class classification. Deep-learning models have been shown to outperform classical machinelearning models in a variety of different prediction tasks (Nguyen et al, 2018;Paterakis et al, 2017). Natural-language processing (NLP) transformers, such as RoBERTa developed by Facebook, may be particularly suited to tasks such as music-genre classification from lyrics due to the textual nature of the problem.…”

Técnicas de machine learning aplicadas a la evaluación del rendimiento y a la predicción de la deserción de estudiantes universitarios, una revisión.

“…Naïve Bayes [54] [55] [56] [57] SVM [54] [58] [55] [57] Árboles de Decisión [58] [55] [57] Redes Neuronales [54] KNN [54] K-Means [55] Entropía Máxima [56] Redes Neuronales Profundas [56] Dentro de los aspectos que se extraen de los comentarios de retroalimentación se puede mencionar la polaridad del comentario lo cual ayuda a crear un análisis amplio de los comentarios recibidos. Trabajos como [54] buscan la polaridad mediante la utilización de un dataset de comentarios que contiene la observación personal de cada alumno relativa a los exámenes, el proceso de enseñanza, el contenido de los módulos y los recursos de laboratorio.…”

“…Con el auge del DL, el campo del análisis de comentarios ha obtenido nuevas maneras de evaluar la información y por lo tanto se realizan comparativas entre métodos tradicionales y métodos y los nuevos enfoques. En este caso, el estudio [56] compara un clasificador entrenado utilizando algoritmos tradicionales como Nave Bayes y Máxima Entropía, para compararlos con algoritmos de Deep Learning, en específico las Redes Neuronales Recurrentes (LSTM) y Redes Neuronales Recurrentes Bidireccionales (BLSTM). Usan en este estudio el UIT-VSFC: Vietnamese Students' Feedback Corpus for Sentiment Analysis [59] que contiene más de 16, 000 comentarios recolectados en durante 3 años.…”

Emotion Recognition for Vietnamese Social Media Text

“…In Vietnamese, there are quite a quantity of research works on other NLP tasks such as parsing [26,27], part-of-speech [28,29], named entity recognition [30,31], sentiment analysis [16,17,18] and question answering [32,33]. However, there are no research publications on emotion recognition for Vietnamese social media texts.…”