How to apply AI to small data problems – TechCrunch

In the past or so, the digital revolution has given us a surplus Data. This is exciting for a number of reasons, but mostly about how AI will be able to further revolutionize business.

However, in the world of B2B – which I am deeply involved in – we are still experiencing data shortages, largely due to a much lower number of transactions compared to B2C. So to AI to fulfill its promise to revolutionize business, it must also be able to deal with these small data problems. Thankfully, it is possible.

The problem is that many data scientists turn to bad practices, creating self-fulfilling prophecies, which reduce AI’s effectiveness in small data scenarios – and ultimately hinder AI’s influence. in promoting business.

The trick to applying AI correctly to small data problems is to follow the right data science practices and avoid the bad ones.

The term “self-fulfilling prophecy” is used in psychology, investing, and elsewhere, but in the world of data science it can simply be described as “predicting the obvious. ” We see this when companies find a model that predicts what has worked for them, sometimes even “by design,” and applies it to different situations.

For example, a retail company determines that people who have purchased their products online are more likely to purchase than those who have not, so they market aggressively to that group. They are anticipating the obvious!

Instead, they should adopt models that optimize what Not works well – converts first-time buyers who don’t have an item in their cart. By solving the latter – or anticipating the not-so-obvious – the retail company will be more likely to influence sales and attract new customers rather than simply keeping the old ones.

To avoid the trap of creating self-fulfilling prophecies, here’s the process you should follow to apply AI to small data problems:

  1. Enrich your data: When you find you don’t have a lot of existing data to deal with, the first step is to enrich the data you already have. This can be done by tapping into external data to apply the same model. We’re seeing this more than ever thanks to the proliferation of recommendation systems used by Amazon, Netflix, Spotify and more. Even if you only buy once or twice on Amazon, they have a lot of information about the products in the world and the people who buy them, so they can make pretty accurate predictions about your next purchase. your. If you’re a B2B company that uses “one dimension” to categorize your transactions (e.g. “big companies”), follow Pandora’s example and break down each customer by levels. the most granularity (e.g. song title, artist, singer gender, melody composition, tempo, etc.). The more you know about your data, the richer it becomes. You can go from low dimensional data with trivial predictions to high dimensional knowledge with powerful predictive and recommender models.

Source link


News of max: Update the world's latest breaking news online of the day, breaking news, politics, society today, international mainstream news .Updated news 24/7: Entertainment, the World everyday world. Hot news, images, video clips that are updated quickly and reliably

Related Articles

Leave a Reply

Your email address will not be published.

Back to top button