A chat application that provides real-time recommendations when user discusses about movies, restaurants, products, travel etc. on chat. The application works on user click tracking and explicit user feedback to learn preferences and personalize future recommendations and improve classification accuracy.
Install server 1a. Install tasksel sudo apt-get install tasksel 1b. Install LAMP Server from tasksel 1c. Copy the files from src/smart-chat to where you want to host (eg: var/www/smart-chat or htdocs/smart-chat/)
Install scikit-learn from http://www.lfd.uci.edu/~gohlke/pythonlibs/ 2a. Install dependencies: sudo apt-get install build-essential python-dev python-setuptools
libatlas-dev libatlas3gf-base 2b. Download source 2c. Extract source and build with sudo python setup.py install (Make sure you have c++ installed) note: Don't use 3rd party distro as it gives an error
Install beautiful soup (http://www.crummy.com/software/BeautifulSoup/#Download) sudo apt-get install python-bs4
Download nltk and nltk data (NOTE: Use sudo -E if you need user variables eg: proxy settings) 6a. Install settuptools (http://pypi.python.org/pypi/setuptools) wget https://bootstrap.pypa.io/ez_setup.py sudo pyhon ez_setup.py 6b. Install Pip: run sudo easy_install pip 6c. Install NLTK: run sudo pip install -U nltk 6d. Install wordnet data. Run python and type import nltk nltk.download() Ensure that server is running using root user account and the nltk.data directory is downloaded for that user account
Configuration: 5a. config.py (only if you have proxy internet connection on the server end of the application) 5b. config.php (most options are self-explanatory)
- Issue Tracker: https://github.com/rishirdua/smart-chat/
- Source Code: github.com/rishirdua/smartchat
- Project page: http://rishirdua.github.io/smart-chat
If you are having issues, contact the authors directly.
This project is licensed under the terms of the MIT license. See LICENCE.txt for details
Refer to the detailed Documentation and presentation in the doc folder