Far from Skynet, a simple natural language processing method–fuzzy matching algorithms–empowers smart guesses on the meaning of your work orders.

Did you know the first CD-ROM was pressed in clay over 3,500 years ago? The Phaistos Disk is comprised of 45 individual symbols grouped together in 241 different combinations, organized as a spiral on both sides of the disk. Discovered on the island of Crete in 1908, there has yet to be any conclusive deciphering of its meaning. Fortunately for your building, this is not the case for any emails you choose to send Tikkit.

What is natural language processing, you ask?

Natural language processing is a field of computer science, artificial intelligence, and computational linguistics concerned with the interactions between computers and human (natural) languages. As such, NLP is related to the area of human–computer interaction.

If this piques your interest, consider an online Stanford University class. Or, for now, take Microsoft’s word for it that NLP is “knowing what concepts a word or phrase stands for and knowing how to link those concepts together in a meaningful way.” However, Microsoft also helpfully points out that “It’s ironic that natural language, the symbol system that is easiest for humans to learn and use, is hardest for a computer to master.”

We’re just getting started with layering deep analytics into your building’s work order data. That said, we’re excited that fuzzy matching algorithms–our first expedition into the universe of NLP–is already providing practical benefits for building operators.

tikkit NLP

Shaving time off of filing your work orders, with informed guesses.

When sending a new or forwarding an existing email to Tikkit, it measures the similarity of the text included to a predefined set of expectations. The similarity measures common characters and allows for typos and abbreviations. If it finds a match above a certain probability of certainty, it makes the assignment. The feedback loop is pictured above, where guesses are returned via email to the user such that they can either go on with their busy to-do lists or opt to make refinements to the work order directly.

Tikkit is not your ancestor’s CMMS, so let us know if you’d like to check it out.

 

About Millen Paschich

Millen began his career at Cambridge Associates, trained in finance at SMU, and has an MBA from UCLA. Talk to him about bicycling, business, and green chile burritos.

0 replies on “Using Natural Language Processing to understand your emails”

You may also be interested in...