|
|
--- |
|
|
language: en |
|
|
license: mit |
|
|
tags: |
|
|
- sentiment-analysis |
|
|
- imdb |
|
|
- tensorflow |
|
|
- lstm |
|
|
- binary_crossentropy |
|
|
model-index: |
|
|
- name: IMDb Sentiment Classifier (LSTM) |
|
|
results: |
|
|
- task: |
|
|
type: text-classification |
|
|
name: Sentiment Analysis |
|
|
dataset: |
|
|
name: IMDb Movie Reviews |
|
|
type: imdb |
|
|
metrics: |
|
|
- name: Accuracy |
|
|
type: accuracy |
|
|
value: 0.72 |
|
|
--- |
|
|
|
|
|
# IMDb Movie Review Sentiment Analysis using LSTM |
|
|
|
|
|
This model performs binary sentiment classification (positive/negative) on IMDb movie reviews using an LSTM-based neural network built with TensorFlow. |
|
|
|
|
|
## Model Details |
|
|
|
|
|
- **Architecture**: LSTM |
|
|
- **Framework**: TensorFlow/Keras |
|
|
- **Dataset**: IMDb Movie Reviews (50k) |
|
|
- **Tokenizer**: TensorFlow `Tokenizer` |
|
|
- **Accuracy**: 72% |
|
|
|
|
|
## Sentiment Labels |
|
|
|
|
|
- `0`: Negative |
|
|
- `1`: Positive |
|
|
|
|
|
|
