atulshgl / ContextualEventSummary

Applied Natural Language Processing class project

Contextual Event Summary Generation

Versions

The parser uses Python 2.7, NLTK 3.2.2 and Stanford version 3.7.0

Installation and Usage

1. Install nltk:

    $ pip install nltk 
   

2. Download nltk_data:

    >>> import nltk
    >>> nltk.download()
   

3. The esp_parser.py script uses jar files present in stanford-postagger-full-2016-10-31 and stanford-parser-full-2016-10-31.
4. Make sure to place unzipped stanford-postagger-full-2016-10-31 and stanford-parser-full-2016-10-31 folders in the same folder as esp_parser.py script.
5. Extract stanford-parser-3.7.0-models.jar in the stanford-parser-full-2016-10-31 folder.
6. To check the completeness of parser intsallation:

    $ python parser_driver.py
   

7. Install Google NLP dependency

    $ pip install --upgrade google-cloud-language
   

8. Set GOOGLE_APPLICATION_CREDENTIALS environment variable

    $ export GOOGLE_APPLICATION_CREDENTIALS=<path_to_google_service_json_file>
   

9. Install Google Cloud SDK (https://cloud.google.com/sdk/downloads)
10. Authenticate the Google client

     $ gcloud auth application-default login
    

11. Install Watson NLU dependency

     $ pip install --upgrade watson-developer-cloud