A Chatbot kernel for Jupyter based on pyAIML
It has been tested with Jupyter 4.x. The code works with either Python 2.7 or Python 3 (tested with Python 3.4)
The installation process requires two steps:
Install the Python package:
pip install aimlbotkernel
Install the kernel into Jupyter:
jupyter aimlbotkernel install [--user] [--logdir <dir>]
The --user option will install the kernel in the current user’s personal config, while the generic command will install it as a global kernel (but needs write permissions in the system directories).
The --logdir specifies the default place into which the logfile will be written (unless overriden at runtime by the LOGDIR environment variable). If no directory is specified, the (platform-specific) default temporal directory will be used.
Note that the Jupyter kernel installation also installs some custom CSS; its purpose is to improve the layout of the kernel results as they are presented in the notebook (but it also means that the rendered notebook will look slightly different in a Jupyter deployment in which the kernel has not been installed, or within an online viewer).
To uninstall, perform the inverse operations (in reverse order), to uninstall the kernel from Jupyter and to remove the Python package:
jupyter aimlbotkernel remove pip uninstall aimlbotkernel
Once installed, an AIML Chatbot kernel will be available in the Notebook New menu. Starting one such kernel will create a chatbot. The chatbot is initially empty but can be loaded with a couple of predefined DBs (use the %help magic for initial instructions).
Notebook input is of two kinds:
The examples directory contains a few notebooks showing some of the provided functionality. They can also be seen with online Notebook viewer (note that, as said above, they will look slightly different than in a running kernel).
AIML is an XML-based specification to design conversational agents. Its most famous application is ALICE, a chatbot (the DB for the free version of ALICE is included in this kernel, as it is included in python-aiml)
The chatbot can load an AIML database (which is basically a bunch of XML files). It can also define AIML rules on the fly, by using the %aiml magic in a cell.