How AI Changed the History of The Beatles

A recent study published in Harvard Data Science Review is changing the way people think about the legendary band, The Beatles. The research also opens up some exciting possibilities when it comes to artificial intelligence and how we use this technology from a developer standpoint.

At this point, business owners and developers are using AI as a way to design websites, communicate with customers via chatbots, generate leads, and send out relevant ads. It’s an extraordinarily helpful tool when it comes to breaking down data and producing results, which leads to the study. How was AI able to solve a hotly-debated conversation surrounding The Beatles? What future uses can we uncover from this breakthrough?

Let’s take a look at the debate around The Beatles, the purpose of the study, the results, and the possible future benefits of AI.

Image title

The Debate

The study. “(A) Data in the Life: Authorship Attribution in Lennon-McCartney Songs” tackles a debate decades old and involves the origins of some of the most popular songs by The Beatles. John Lennon and Paul McCartney worked on many of their songs together, but after Lennon’s murder in 1980, it became hard to distinguish who was responsible for each track, especially considering McCartney’s apparent agitation over Lennon getting more credit for their work.

Fans of the band picked sides between the potential creators and assumed that they would never know for sure who was the primary creative force behind their favorite Beatles hit. A group of scientists decided to try and get answers for everyone.

Study Parameters

The scientists who started this project developed an AI program that can distinguish between Lennon and McCartney’s creative talent. To make their idea work, they had to introduce hundreds of songs with known creative talent out of the two stars, plus their own performances.

Their machine was able to compile the data from all of the music and develop artist profiles for both Lennon and McCartney. When a new song was played, the machine could determine the creative influence of each man by percentage. As time went on, the AI was able to display clear rules that it learned based on the music submitted. For example, it noted that McCartney’s style “tended to use more non-standard musical motifs.” So, what were the results?

AI Results

The final test consisted of the eight most hotly debated songs. When the songs were put through the program, they determined that most of the music within those eight tracks were primarily Lennon’s style.

For example, “A Hard Day’s Night,” which McCartney sang, and mentioned in multiple interviews that he played a big part in its creation, was mostly Lennon’s work. McCartney was proven to be 97 percent responsible for “Baby’s in Black” and “The Word.”

Out of all of the song tests, the scientists stated that “In My Life” was the song that raised the most questions with fans of The Beatles. The AI program determined that there’s an 81.1 percent chance that Lennon wrote the lyrics to the song. Finally, the system concluded that there’s a 43.5 percent chance that McCartney wrote the music, which contradicts McCartney’s claims.

Potential Impact in Other Industries

It’s clear that AI has managed to blend in almost every aspect of our lives. We use AI or machine learning to keep our websites running, as a component to create new programs, and as a way to monitor and improve our business. Right now, 79 percent of the top businesses use automation and AI to interact with their customers and run their companies. How can this study change the way we use AI in the future?

The possibilities are virtually endless. Can you imagine a music application run on AI that tracks your performance on instruments, and then allows you to master certain aspects of your instrument. The app would understand your style, tell you when you’re improving, and offer solutions for mistakes that happen often.

As time goes on, it’s safe to say that we are going to continue to learn and grow with artificial intelligence. If we are now able to solve decades-old mysteries, the potential for future AI innovations is limitless.

This UrIoTNews article is syndicated fromDzone