A handful of documentation updates

docs/django/index.rst

@@ -51,8 +51,8 @@ If you need a ChatterBot API endpoint you will want to add the following to your
    )
 
 
-Sync your database
-------------------
+Migrations
+----------
 
 .. sourcecode:: sh
 
@@ -63,4 +63,15 @@ Sync your database
    Looking for a working example? Check our the example Django app using
    ChatterBot on GitHub: https://github.com/gunthercox/ChatterBot/tree/master/examples/django_app
 
+MongoDB and Django
+------------------
+
+ChatterBot has a storage adapter for MongoDB but it does not work with Django.
+If you want to use MongoDB as your database for Django and your chat bot then
+you will need to install a **Django storage backend** such as `Django MongoDB Engine`_.
+
+The reason this is required is because Django's storage backends are different
+and completely separate from ChatterBot's storage adapters.
+
 .. _Django documentation: https://docs.djangoproject.com/en/dev/intro/install/
+.. _Django MongoDB Engine: https://django-mongodb-engine.readthedocs.io/

docs/faq.rst

@@ -29,3 +29,39 @@ There are a number of existing examples that show how to do this.
 
 Additional details and recommendations for configuring Django can be found
 in the :ref:`Webservices` section of ChatterBot's Django documentation.
+
+What kinds of machine learning does ChatterBot use?
+---------------------------------------------------
+
+In brief, ChatterBot uses a number of different machine learning techniques to
+generate its responses. The specific algorithms depend on how the chat bot is
+used and the settings that it is configured with.
+
+Here is a general overview of some of the various machine learning techniques
+that are employed throughout ChatterBot's codebase.
+
+1. Search algorithms
+++++++++++++++++++++
+
+Searching is the most rudimentary form of artificial intelligence. To be fair,
+there are differences between machine learning and artificial intelligence but
+lets avoid those for now and instead focus on the topic of algorithms that make
+the chat bot talk intelligently.
+
+Search is a crucial part of how a chat bot quickly and efficiently retrieves
+the possible candidate statements that it can respond with.
+
+Some examples of attributes that help the chat bot select a response include
+
+- the similarity of an input statement to known statements
+- the frequency in which similar known responses occur
+- the likeliness of an input statement to fit into a category that known statements are a part of
+
+2. Classification algorithms
+++++++++++++++++++++++++++++
+
+Several logic adapters in ChatterBot use `naive Bayesian classification`_
+algorithms to determine if an input statement meets a particular set of
+criteria that warrant a response to be generated from that logic adapter.
+
+.. _naive Bayesian classification: https://en.wikipedia.org/wiki/Naive_Bayes_classifier