Algorithms Applied in AI Solutions

When you talk about artificial intelligence, it’s not far off that you will also talk about algorithms. Sometimes, people might use one to mean the other.

However, an algorithm is merely a set of instructions that gets carried out when a trigger is met.  Think of it as a recipe that a cook follows to the letter, each ingredient prepared in an exact way to cook a batch of brownies.  When the next batch is prepared, the ingredients are prepared the exact same way so that the resulting confection is similar to the first one.

In algorithms, everything is rigid and unchanging. Not so with AI.  Artificial intelligence uses different algorithms to do its work. As AI gets inputs, it modifies the algorithms it has or creates new one.

It can also incorporate more triggers other than the one that it was trained to recognize. Where algorithms are rigid and unchanging, AI changes, adapt, and grow as it learns new data or encounters new triggers.

Algorithms Are Great, or Are They?

There are many examples of how algorithms work to make our lives better.  You may not even realize it, but you interact with algorithms and artificial intelligence on a daily basis.

They’re responsible for what posts you see on Facebook.  Or the TV shows and movies recommended by Netflix. It also helps give you the best and most relevant results on Google.

In the past, algorithms have made such an impact that they changed the way we work.  For instance:

• Algorithms that can compress data, video, and audio, has given rise to voice over IP or IP television.
• TCP/IP is a set of protocols that manages how the Internet sends or receives data.
• RSA and other security algorithms help keep Internet users secure.

In the context of AI, algorithms are what converts data into something more than just numbers and text.  It makes it more intelligent.  At its most basic, algorithms can help you recognize trends in the data, as well as report to you in such a way that you can easily spot these trends or understand the numbers.

Algorithms help make it easier for us to do analytics on the data we collect. And this in turn helps us manage our environment better and with higher efficiency.

But wait…

However, there are times when algorithms come up with insights that may not make sense.  For instance, some insights that were uncovered by using AI and its algorithms on data sourced by a variety of companies include:

• More deals were closed when there is a new moon than during a full moon.
• People answer phones more when it’s cold or humid, while they reply to emails more when it’s sunny.
• Loan applicants who fill out applications using all capital letters are more likely to default than those who use small caps or use capitalization correctly.

Algorithms Need the Human Touch

But while this may sound ridiculous to some, these are real-world insights that were discovered using algorithms with AI.  How does artificial intelligence judge which insights have value to an individual or business?

The short answer is they don’t. This is where data scientists come in. Humans will need to pore over the various insights and data produced by AI and algorithms to see if they can use it for their purposes

* * *

Like in any process, the soft skills matter as much as the hard skills. In an AI-driven future, the technology will be taking over the repetitive tasks using algorithms to do their work.  However, that doesn’t mean that we humans will have fewer jobs available to us. Instead, data scientists, end users, and other humans will be there to make sense of the data and insights that are presented by AI and algorithms.