Data and AIAI and data.
You almost always hear the two terms pronounced in the same breath. Why is that?
If you’re a founder trying to understand more about these topics, be it improving your workflows or products or some aspect of your operations, here’s a business owner’s introduction to what people mean when they insist to say the two together.
AI needs data to do anything.
At its core, AI is an algorithm, which in plain English is a process that takes input and produces output. Just like your car, which is just a piece of metal sitting in the garage until it has fuel to run it, an algorithm on its own with no data to process can’t make anything useful. In fact, it can’t do anything at all.
This means that if you want your business to benefit from AI, the first job is to get your data together and get it into shape. This can be a stumbling block, says Phong Nguyen, founder of data science consultancy Partners in Company. “From the customer we’ve worked with and talked to, the barriers to being more data-driven are usually the foundation of having clean, consistent data and being centralized and secure,” she says.
That usually means pulling your data from spreadsheets or bringing your data from multiple platforms — such as a customer relationship management (CRM) platform and a marketing platform — together into a centralized repository, where the data can be combined and compared. for analysis. It usually then needs to be cleaned and normalized in various ways to ensure it is consistent and in the right shape before data teams can draw the right conclusions and then build on the data with AI
Moreover, most AI needs large amounts of data to produce reliable results, for the same reason that you need a large sample of everything to make a reasonable judgment. We’re all familiar with political polls, where professionals typically claim greater than 95 percent accuracy on how the larger population plans to vote in an election by sampling somewhere around 300 people.
That is for a simple choice between two options. If you’re trying to make more complex predictions, such as distinguishing types of customer behavior in your marketing data, you should start with many thousands of examples. Often you will use a lot more to gain strong confidence in your results.
How much data are we talking about? Good statistical analysis can give you a precise number for what you’re trying to do, but as a general rule, hundreds of thousands of rows are usually on the low end for machine learning-based analytics. “I’m not used to working with anything less than a million rows,” said Chantel Perry, an experienced data scientist at large corporations and author of the book Data Newbie to Guru.
And for something like a marketing analysis, where the customer trends you’re trying to understand can vary from day to day and month to month, you also want to collect enough data over a period long enough to make actionable predictions: be active for at least six months and collect data about your customers for at least six months,” says Perry.
So now you understand why AI needs data. That dependence also goes the other way. The truth is, you can’t have one without the other.
A lot of data comes from AI
Just as AI algorithms need data as input, their output is often some form of data.
Suppose your marketing data is cracked in such a way that you find that you have eight major clusters of customers. You might further discover that different clusters of customers should receive different types of pitches or advertisements. Those outputs are data that you can feed into another algorithm, one where you can then use that labeling to predict which cluster a prospective customer will belong to and then have an automated process that assigns them the pitches or ads that are predicted to happen. will be the most. effective.
If you think about it, all data is the result of a process that resembles an algorithm, often AI. Sometimes AI drives that data collection process, sometimes it doesn’t, and sometimes the distinction isn’t so clear. Take, for example, data on average income and spending patterns in a geographic location that you focus on: it can come from a combination of surveys, government data, data collected by credit card companies and merchants and then put back together into a single number for a single counting block, which your marketing algorithms can then use to help you target different customers in different ways.
There’s a common saying I often use when talking about data science: “No one believes in a model except the person who wrote it, and everyone believes in a particular dataset except the person responsible for compiling it. ” Noodle on that for a minute.
We tend to believe that data is necessarily true and does not depend on any human or AI process to be as it is. But that is often not true. To get meaningful results, you need to examine the data that feeds your models — as well as the models that produced the data you feed your models.
“The biggest issue I see issues with is data quality,” says Perry. “Anything that enters the decision-making process should be checked for cleanliness, bias, and other issues, especially with machine learning models.”
Understanding this back and forth between data and AI and their feedback loop can help you avoid relying on analytics that aren’t as good as they seem at first glance.
Enter my scayt-misspell-word” data-scayt-word=”scayt-misspell-word” data-wsc-lang=”en_US” data-wsc-id=”l5k105hb2cgk009tg”>scayt-misspell-word” scayt-misspell -word” data-scayt-word=”data-scayt-word” data-wsc-lang=”en_US” data-wsc-id=”l5k105ha10uz2t7yk”>data-scayt-word="scayt-misspell-word” data -scayt-word=”Wufoo” data-wsc-lang=”en_US” data-wsc-id=”l5k105h9mnssidath”>Wufoo” scayt-misspell-word” data-scayt-word=” data-wsc-lang” data- wsc-lang=”en_US” data-wsc-id=”l5k105h7cwy8or5nw”>data-wsc-lang="en_US" scayt-misspell-word” data-scayt-word=”data-wsc-id” data-wsc-lang=”en_US” data-wsc-id=”l5k105h6tq0gsw22s”>data-wsc-id="l513s1pcuj91h7ddd">scayt -misspell-word” data-scayt-word=”Wufoo” data-wsc-lang=”en_US” data-wsc-id=”l5k105h5t0nzfr303″>Wufoo form!
This post Why do ‘data’ and ‘AI’ always go together?
was original published at “https://www.inc.com/steven-i-weiss/data-ai-together.html”