Core.matrix is the Clojure library most commonly used for data wrangling. ( Incanter uses core.matrix as the library backing the dataset implementation). If you like to visualize the data using Gorilla-repl and you use core.matrix, you might find this library useful.

Here's a short introduction.

Given my programming background, I seldom had a need to use Microsoft Excel (or other spreadsheet equivalents). However, in the Data Science field, I find that Excel is a preferred weapon of choice to get a quick idea about the data and its characteristics, especially used by folks from a non-programming background. It has the ability to filter, splice and pull up graphs (and a whole host of features that I know nothing about).

While I've seen that Excel is a great tool for this task, I feel slightly hamstrung by the lack of programming support . I know there are macros and other geegaws available, but for someone coming from the Java/Clojure side of the world, investing the time to learn a new set of tools seems like an awfully large side project.

Bots are right now in the Technology trigger phase of the Gartner hype cycle. Are they likely to succeed, or are they likely to fizzle out?

Lets take a look at these statistics:

Here's Part II of the guide to building a Facebook messenger bot using Clojure and Heroku.

In case you missed Part I, here it is

Here's a howto guide on creating and running a bot on the Facebook messenger platform using Clojure on the Heroku PaaS platform. Part inspired from this post by Abhay.

The messenger platform requires the use of a HTTPS URL, and the URL needs to be available and accessible to test against. I found that using Heroku solves multiple goals, no need to expose the URL via ngrok+localhost, nor is getting a SSL certicate required since Heroku dynos are available on an HTTPS URI.