Skip to content

The complete social media intelligence platform with dynamic sentiment analysis​.

Notifications You must be signed in to change notification settings

tanmaypardeshi/PULSE-X

Repository files navigation

Pasckathon_git_commit

PULSE - X : The complete social media intelligence platform with dynamic sentiment analysis​

Prequisites on the local machine to project to be run:

  1. Python 3.6 or above
  2. Node JS v12.0.0 or above
  3. Yarn 1.22.4
  4. Expo SDK 38
  5. pip 3 v20.0.x or above

Steps to run the project on the local machine

  1. Clone this repository into your local machine by using
    git clone https://github.com/tanmaypardeshi/Pasckathon_git_commit.git
  • Setting up the django development server:

  1. Install virtualenv using pip3 install virtualenv
  2. Create a new virtualenv using virtualenv venv
  3. Activate virtualenv using source venv/bin/activate
  4. Install all required dependencies using pip install -r requirements.txt
  5. cd into the backend directory and run the following commands to set up database and populate data into it
    1. python manage.py makemigrations user
    2. python manage.py makemigrations employee
    3. python manage.py migrate
    4. python manage.py loaddata user.json
    5. python manage.py loaddata review.json
  6. To run the development server python manage.py runserver.
  7. Deactivate the virtualenv using deactivate.

Note: Step 1,2,4 and 5 are one time process to set up data and dependencies.

  • Setting up the react local server:

  1. cd into the frontend directory.
  2. Use command yarn to install node modules for the front end.
  3. Keep the django server up and running to make the website functional.
  4. Once the node modules have been installed, run yarn start to start the development server.
  • Setting up expo SDK for the mobile app:

  1. cd into the mobile directory.
  2. Use command yarn to install node modules for the mobile app.
  3. Keep the django server up and running on your local network to make the app functional. Note that the command for the django server in this case will be python manage.py runserver :8000
  4. Once the node modules have been installed, run yarn start to start the expo server.
  5. Scan the QR code on your mobile device using the expo app available on play store.

Information about the directories in the repository

(Click on the file name to open the folder)

Directory Name Description
backend Django REST framework app with the APIs
frontend Front end for the web app made in React
mobile Mobile app using React Native and Expo SDK
Machine_Learning Notebooks used for the multilayered filtering approach
scripts Python scripts to extract data from csv

Information about the notebooks in the Machine_Learning directory

( Click on the notebook to access )

Notebook Name Description
Json_Processing.ipynb Extracts the data from json format and converts it to a csv
Text_Preprocessing.ipynb Preprocesses the data by performing lemmatisation, tokenisation and removes stop words to bring the data in the required format by the models
SentimentAnalysis.ipynb The polarity of a tweet is analysed by using a pretrained model from TextBlob
KeywordsExtraction.ipynb Extraction of aspects and keywords essential to the product
SarcasmDetection.ipynb To detect the true intent behind the review
NER_BERT.ipynb To detect the product by fine-tuning BERT
NER_KeywordExtraction.ipynb To detect the product by using the DeepPavlov Library and the keywords using TextRank Algorithm
BertSentimentAna.ipynb Sentiment analysis module
Helpfulness.ipynb Measures helpfulness of review
ProfanityDetection.ipynb Checks if the reviews are abusive or not

Snippets of the platform:

Web app

1.jpeg 2.jpeg

1.jpeg 2.jpeg

1.jpeg 2.jpeg

1.jpeg 2.jpeg

1.jpeg 2.jpeg

1.jpeg 2.jpeg

1.jpeg 2.jpeg


Mobile app

1.jpeg   2.jpeg   1.jpeg   2.jpeg   1.jpeg   2.jpeg   1.jpeg   2.jpeg   1.jpeg   2.jpeg   2.jpeg   2.jpeg

About

The complete social media intelligence platform with dynamic sentiment analysis​.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages