Billboard Japan and NTT Data / NTT Data Institute of Management Consulting (NTT Data / IMC) have collaborated on a new project that combines graph data from the former and neuroscientific technology from the latter to develop a new way of analyzing musical trends.
Billboard Japan’s Hot 100 song chart is based on a weighted formula that incorporates eight metrics such as CD sales, downloads, streaming, and Twitter mentions. NTT Data / IMC is a pioneer in advancing the brain information business, in part through its development of the “NeuroAI” which predicts the brain activity and perceptual content of people viewing moving images. The technology behind NeuroAI was adopted at this year̵
The collaborative project used a technology called “NeuroAI-Music” which simulates the brain activity of people listening to music. NeuroAI-Music managed to estimate and characterize the brain activity per second of music from audio signals, making it possible to quantitatively extract the characteristic of music latently expressed in the brain, which is clearly different from conventional music genres such as rock, hip-hop, jazz , etc.
Many streaming services and video platforms offer playlists and recommend music via algorithms based on user activity. NeuroAI-Music technology allows listeners to encounter the songs they like from the types of music they have never heard before. Billboard Japan accounts on Apple Music, Spotify and LINE MUSIC feature playlists with “songs that stimulate the brain in similar ways” as some of the more recent issues.
The project also built a chart prediction model that indicates which song types ranked high during a particular week, taking chart data of the songs and the characteristics of the lyrics and chord progression, as well as the mentioned brain information. above, to quantify the characteristics of songs that rank high on the Japan Hot 100. Thus, considering the time series variations of this chart prediction model as “trend variations”, the project set out to predict the characteristics of songs which would have trended in the future based on past variation patterns.
For example, a long-term memory (LSTM) model, a type of neural network capable of learning long-term time series data, trained with the chart prediction model up to the week of March 16, 2020, was used to predict a future model. The correlation coefficient between the actual chart for the week of March 30, 2020 and the predicted model was 0.737, indicating that the forecast was highly accurate. This LSTM model has been shown to maintain accurate predictions up to four weeks earlier.
Subsequently, a similar prediction was made using the metric “spiking trends” (Top Trends Ranking), which indicates the “songs that are suddenly heard” by standardizing the points of the graph. This time, the correlation coefficient between the actual ranking and the forecast maintained accuracy for up to about four months into the future. [Note: The songs used in this prediction are those that have charted up to May 2020.]
Based on these findings, Billboard Japan and NTT Data / NTT Data Institute of Management Consulting intend to launch a new service that offers tailored forecasts for record labels, music publishers, management companies, advertising agencies, streaming platforms and more.
* NeuroAI is a neuroscience-based technology to predict and simulate brain activities. This technology is applied to marketing, research, development and so on.