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)

The use of modular neural network (MNN) for the recognition of partial discharge (PD) sources has been investigated. Three phase related quantities, the PD pulse counts, the average and maximuim discharge magnitudes form the feature vector of a discharge signal. The MNN consists of 5 sub-networks with identical structure and a maximum selector. Each subnetwork is assigned the task to recognize a particular PD source. Compared with a single neural network which is trained to recognize all PD sources, the MNN has a higher training ability, faster rate of convergence and better recognition rate.

No abstract