https://github.com/vubacktracking/deep-neural-network-vietnamese-student-feedback-sentiment-analysis

Vietnamese Student Feedback Sentiment Analysis
https://github.com/vubacktracking/deep-neural-network-vietnamese-student-feedback-sentiment-analysis

deep-learning keras-tensorflow lstm-model lstm-neural-networks lstm-sentiment-analysis nlp sentiment-analysis streamlit streamlit-webapp tensforflow tfidf-vectorizer vietnamese-nlp word2vec word2vec-model

Last synced: 25 days ago
JSON representation

Vietnamese Student Feedback Sentiment Analysis

• Host: GitHub
• URL: https://github.com/vubacktracking/deep-neural-network-vietnamese-student-feedback-sentiment-analysis
• Owner: VuBacktracking
• License: mit
• Created: 2023-12-12T14:17:43.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2024-03-18T14:34:00.000Z (11 months ago)
• Last Synced: 2024-11-10T01:13:11.531Z (3 months ago)
• Topics: deep-learning, keras-tensorflow, lstm-model, lstm-neural-networks, lstm-sentiment-analysis, nlp, sentiment-analysis, streamlit, streamlit-webapp, tensforflow, tfidf-vectorizer, vietnamese-nlp, word2vec, word2vec-model
• Language: Jupyter Notebook
• Homepage:
• Size: 34.4 MB
• Stars: 6
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Vietnamese Student Feedback Sentiment Analysis Project

---

In this Project, I have used the Corpus named Vietnamese Student Feedback Corpus, which belongs to The UIT NLP Group and has more than 16,000 pieces of feedback from Vietnamese students, to solve the Sentiment Analysis task. I have trained some models and chose the best model with the result in the following table.\

**Note**: You can download the corpora in https://nlp.uit.edu.vn/

## Model used:

|Model|Feature|Precision|Recall|F1-Score|Accuracy|

|---|---|---|---|---|---|

|LSTM| Word2Vec|90.0|89.7|89.8|89.7|

### Word2Vec



    



---

## Web Application

We have built a web application for Analyzing the Sentiment of Vietnamese Student's Feedback

### Corpus Analysis

You can browse the corpus which contains feedback like 4 corpora in the [**Data/testForApp**](Data/testForApp) directory. The application will predict the sentiment for each feedback, and then display the data frame of the result and the chart for analysis.



    





    





    



### Feedback's Sentiment Analysis

You can input Vietnamese feedback about schools and universities including lecturers, facilities, curriculum, etc, the application will predict the sentiment and display the polarity beside.



    





    



---

## How to use?

#### Clone the project

```bash

git clone https://github.com/VuBacktracking/Deep-Neural-Network-Vietnamese-Student-Feedback-Sentiment-Analysis.git

```

#### Install required packages

```bash

pip install -r requirements.txt

```

#### Generate model (optional)

We already ran and saved the model `lstm_model.h5` in `models` directory

```bash

#run this file to generate the models

lstm_create_model.ipynb

```

```bash

#run this file to check the result of model

lstm_result.ipynb

```

#### Now run

``` bash

streamlit run app.py

```

___

## Future Work