Eureqa was a proprietary modeling engine created in Cornell's Artificial Intelligence Lab and later commercialized by Nutonian, Inc. The software used genetic algorithms to determine mathematical equations that describe sets of data in their simplest form, a technique referred to as symbolic regression.
| Owner | Datarobot, Inc. |
|---|---|
| Created by | Michael Schmidt and Hod Lipson |
| URL | www |
| Commercial | Yes |
| Launched | November 2009 |
| Current status | Active |
Origin and development
editSince the 1970s, the primary way companies had performed data science was to hire data scientists and equip them with tools like R, Python, SAS, and SQL to execute predictive and statistical modeling.[1] In 2007 Michael Schmidt, then a PhD student in Computational Biology at Cornell, along with his advisor Hod Lipson, developed Eureqa to help automate the curve fitting work of data scientists by creating a tool that would automatically search for the "best" mathematical model to fit a given dataset (where best is defined as the simplest model that can be found to achieve a given level of fit to the data).[2][3]
In November 2009 the program was made available to download as freeware.[4] Lipson described the tool's benefit as dealing with fields that are overwhelmed with data but lack theory to explain it.[5] In the October 2011 edition of "Physical Biology", Lipson described a yeast experiment that predicted seven known equations.[6] This took place after Lipson had asked scientists from different disciplines to share their work to test Eureqa's versatility.[6]
Technology
editEureqa worked by creating random equations with the data through evolutionary search.[5] Initial guesses might not fit the data well but some of the equations will fit better than others and those will be used as the basis for the next round of guesses until the fit cannot be further improved.[7]
Reception and use
editOver 80,000 users downloaded the program.[8] People used the application for many uses including analyzing the herding of cattle and modeling the behavior of the stock market.[4]
In 2017 Nutonian was acquired by DataRobot and Eureqa merged into their payware portfolio.[9]
References
edit- ^ Piatetsky, Gregory. "Four main languages for Analytics, Data Mining, Data Science". www.kdnuggets.com. KDnuggets. Retrieved 19 May 2016.
- ^ Keohane, Dennis. "Nutonian - At the Cutting Edge of Technology, Science, and Data Analysis". www.venturefizz.com. VentureFizz. Archived from the original on 2017-01-04. Retrieved 2016-05-19.
- ^ Regalado, Antonio (August 19, 2014). "35 Innovators Under 35". MIT Technology Review. Retrieved 19 May 2016.
- ^ a b Keim, Brandon (December 3, 2009). "Download Your Own Robot Scientist". Wired Magazine. Retrieved April 22, 2013.
- ^ a b Chang, Kenneth (April 2, 2009). "Hal, Call Your Office: Computers That Act Like Physicists". The New York Times. Retrieved April 22, 2013.
- ^ a b Ehrenberg, Rachel (January 14, 2012). "Software Scientist". Science News Digital. Retrieved April 22, 2013.
- ^ Manjoo, Farhad (September 30, 2009). "Will Robots Steal Your Job?". Slate. Retrieved April 20, 2013.
- ^ Shtull-Trauring, Asaf (February 3, 2012). "An Israeli professor's 'Eureqa' moment". Haaretz. Retrieved April 20, 2013.
- ^ "DataRobot Acquires Nutonian". DataRobot. May 25, 2017. Retrieved December 9, 2023.