/Entity-Based-Sentiment-Analysis

Primary LanguageJupyter Notebook

GETTING STARTED

Run flask HTTP server

app.py

API ENDPOINT

'Content-Type: application/json'

POST /predict

Request Body :

```json
{"sentence":<"Enter the list of Sentences">, "entity":<"enter the entities list">}
```

Inference from the model:

```json
{"sentence":["the website was great", "the movie was good but I didn't like it"], "entity":["website","movie"]}
```
```json
[{'Entity': 'website',
  'Sentence': 'the website was great',
  'pred_prob': array([-0.6650544, -0.722052 ], dtype=float32),
  'sentiment': 'Negative'},
 {'Entity': 'movie',
  'Sentence': "the movie was good but I didn't like it",
  'pred_prob': array([-0.6282205 , -0.76258385], dtype=float32),
  'sentiment': 'Negative'}]  



***
***