Data science is changing the way organizations do just about everything. That change is apparent in fields like social media or e-commerce, but lesser known fields have massive potential for a career in data science.
If you aren’t interested in social media or e-commerce, there could be a way to unite your passion in another field. Here are a few unconventional fields hiring for data science and machine learning that could be your next big break.
Know anything about art appraisal? Appraisals can be a laborious process because they rely on historical data and comparable appraisals. Truly rare or unusual pieces may require research just to find those comparables.
Data science can shorten the research time considerably by locating and logging patterns in the historical record and lateral data through comparable art pieces. The result? Faster, more informed appraisals.
This is a wide open field, so data is waiting to be curated. As styles and trends change, the data sets will also need to be retrained. Data science can give appraisers a faster presentation of relevant data for valuation purposes and rely on the human element to make the final call. Even Sotheby’s is on board.
Data science could also help us identify forgeries more quickly as machines begin to learn the data sets belonging to more than a few details common to a single artist. Any of those datasets goes awry, and you could have a potential forgery on your hands. Again, machines aren’t going to detect a forgery with 100% accuracy just yet, but just flagging potential issues for further, more refined review could be a game changer.
We have more data from the past two years than we have in all of recorded human history, but that doesn’t mean we can’t comb through that information to find subtle shifts in human behavior and trends in cultures over time.
The historical record is a tricky subject, but we can glean a lot of how our forbearers lived using analytical data. Big Data’s potential to reveal patterns in past human behavior could give historians a better grasp of how people lived through the analysis of data previously too unwieldy to yield insights.
For example, historians have long thought that shifts in cultural patterns could be detected by reading newspapers, but no human can consume enough newspapers to collect that kind of data. In 2017, however, a research team used big data to analyze nearly 35 million articles from 100 regional newspapers across the past 150 years, proving that newspapers can show subtle shifts in cultural norms over time.
In 2012, Steven Pinker, a psychologist and linguist, released one of his best-known works, The Better Angels of Our Nature. In it, he used big data to argue that contrary to our perception, human violence is on the decline. A book like this would only be possible with machine learning and that was seven years ago. The field is only getting bigger.
One of the biggest projects, something the Swiss Federal Institute of Technology in Lausanne (EPFL), calls the Venice Time Machine, led by Frederic Kaplan, is analyzing over 1000 years of history through things like maps, monograms, financial records, and even sheet music, to provide insight into how the city of Venice developed.
Machine learning isn’t just cataloging data; it’s learning. It separates significant data points from inconsequential based on sets of targeted queries, returning more valuable results. Data scientists will be involved in the narrative that comes out of those results, interpreting and presenting a coherent narrative valuable to the historical record.
Few things have changed more than the field of music. Streaming, crowdsourcing, and marketing, in addition to studies in music related to history and sociological trends, will transform the field even further.
The potential for big data in music is one of the most varied of all. Industry insiders are using better, faster analytics to turn on a dime to market to newer audiences. For example, streaming services such as Spotify uses analytics to tailor suggestions and also to handle increasingly complicated compensation formulas.
Big music industries are sensing the shift. They’re betting more on proven artists, using analytics such as YouTube or SoundCloud stats to make safer investments. There’s a significant gap, however, in the amount of music owned by one of the big three music companies and the amount of music begin created.
Smaller companies are also using analytics to buy up song rights to potentially big, but currently unknown, songs, and use massive audience analytics to understand how and when to market the song.
The Michigan Institute For Data Science funded four different research teams specifically in music projects with $75,000 to study the role of human behavior, social media, and performance in music as well as the connection between words and music. They intend to broaden our understanding of the human relationship to music.
Data Science Is Changing The Humanities
If your area of expertise leans hard towards the humanities, you aren’t stuck choosing between your data science career and your love of art. The capacity for all fields to utilize big data effectively depends on the expertise of data scientists who can build a narrative, present information, and see projects through to the end. As our understanding of human behavior changes, so will our fields dedicated to humanities.