This will install the kernel spec into your user home instead of the Python install directory (which may not be writable). This assumes that Jupyter can be called using the command jupyter.

I did install Python 2.7.13 separately for this and also create a kernel spec file similar to that of Python 3.5 version. python2 /usr/local/share/jupyter/kernels/python2 ...

Similarly, a virtual Python environment that was created with either virtualenv or venv can be accessed in Jupyter Notebook via a kernel. Create a Conda virtual environment and install the kernel ...