https://github.com/x4nth055/food-reviews-sentiment-analysis

Sentiment analyzer that predicts the review star ( from 0 to 5, continuously) of given food text review.
https://github.com/x4nth055/food-reviews-sentiment-analysis

amazon-food-reviews deep-learning deep-neural-networks embeddings keras lstm-neural-networks natural-language-processing regression review-sentiments sentiment sentiment-analysis

Last synced: about 2 months ago
JSON representation

Sentiment analyzer that predicts the review star ( from 0 to 5, continuously) of given food text review.

• Host: GitHub
• URL: https://github.com/x4nth055/food-reviews-sentiment-analysis
• Owner: x4nth055
• License: mit
• Created: 2019-04-20T10:47:12.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2023-03-24T22:54:29.000Z (almost 2 years ago)
• Last Synced: 2024-11-02T09:32:00.452Z (about 2 months ago)
• Topics: amazon-food-reviews, deep-learning, deep-neural-networks, embeddings, keras, lstm-neural-networks, natural-language-processing, regression, review-sentiments, sentiment, sentiment-analysis
• Language: Python
• Homepage:
• Size: 18.8 MB
• Stars: 11
• Watchers: 2
• Forks: 6
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# 5-Stars Food Reviews Regressor

Building a simple regressor that predicts the review star of given food text review, **80%** accuracy was achieved and about **0.38** **M**ean-**S**quared **E**rror (MSE).

# How it Works

A neural network with Embedding layer as first layer, **L**ong & **S**hort **T**erm **M**emory (LSTMs) since text is sequential data, then one fully connected neuron (dense) with linear activation function for regression ( continuous ratings ). The basic architecture is in the image below:

![alt text](images/network_diagram.png)

# Requirements

- ## Dataset

[Amazon Fine Food Reviews](https://www.kaggle.com/snap/amazon-fine-food-reviews/): large dataset (more than 500K reviews ) that consists of reviews of fine foods from amazon.

Download and extract `Reviews.csv` to `data` folder (training will not work without)

- **keras** for creating, training and testing model.

- **sklearn** for utils like splitting data.

- **tqdm** for printing progress bars.

- **pandas**

- **numpy**

```

pip3 install requirements.txt

```

# Test

In case you want to test directly.

```

python test.py "Best Product Ever"

```

Output:

```

4.82/5

```

# Train

There is already a trained model in `results` folder. However you can tune some parameters in `config.py` to improve **MSE** such as number of LSTM units, embedding size, etc. then run:

```

python train.py

```