Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Latest Machine Learning Algorithm Helps Astronomers With Research

DZone's Guide to

Latest Machine Learning Algorithm Helps Astronomers With Research

Machine learning is on the rise with loads of applications. Recently machine learning has aided astronomers with star gazing.

· Big Data Zone
Free Resource

Free O'Reilly eBook: Learn how to architect always-on apps that scale. Brought to you by Mesosphere DC/OS–the premier platform for containers and big data.

All kids have a habit of gazing at the stars and other celestial bodies like the sun and the moon. Stars, always remain a center of wonder for humans. The field that study stars and other celestial bodies is, astronomy and the people related to the domain are called as astronomers. Stargazers always study about the celestial bodies and try to predict their size, shape, weight and other measurable factors. Sometimes, they get success, but not always. Astronomers got relief from the discovery of the latest machine learning algorithm.

Now astronomers are taking help from the machine in the sky survey. The algorithm assists in studying the basic properties of the celestial bodies. Instead of studying the properties by spectroscopy, system uses an algorithm to quickly identify the several factors. Although, machine learning was used before for the study of the cosmos, but the latest algorithm is unique as it predicts specific factors related to stars. This is a unique way of studying the star. What one has to do is just keeping a watch on the results obtained by the machine.

How the Whole Process Works

Machine learning requires an enormous amount of data to be fed to the system. Even the algorithm requires a huge database for the storage purpose. Before the final working, the machine must be trained.

For the training, the researchers used the data of 9000 stars, which include information about variable brightness and spectra. All this data was recorded by Sloan Digital Sky Survey into a supercomputer. From the star, scientists have obtained spectra which disclosed several basic properties of the star like temperature, size and the quantity of heavy metals like iron. The variation in brightness of the star was recorded by plotting light curves. Feeding these sets of information to the computer, the system made the link between the light curves and the rest of the star properties to obtain the result.

The result obtained is the prediction of the newly discovered star's physical properties. The system was able to predict on the basis of the data previously fed into it. The technique used is similar to the filter used for avoiding spam email. The filters are provided with the keywords that are associated with the junk mail and removing the mails containing those words. As the time passes, the filter learns more keywords taught by the user and performs better than the previous. Next time, if the system finds the junk mail, it filters it better and learns more words again.

In the same way, the algorithm prepared by the researcher learns from the new data entries and performs predictions better than before.

In Conclusion

Artificial Intelligence is helpful to mankind in many forms and example is mentioned in the article. You can see that machine learning can change the way research and help humans predict many things. Not just about stars, but we can predict anything. Star prediction algorithm assists astronomers in their research and predict the accurate dimensions and physical quantities of the star. Humans are prone to errors, but the machine work according to the algorithms and are less prone to mistakes.

In machine learning, the system learns automatically from its previous experiences and improves the performance from better to the best. Researchers have fixed their next goal to 50 million variable stars at a time. Ultimately, this proves that artificial intelligence, in the form of machine learning is helpful in research fields and can help to get more perfect results for the problem.

References:

  • http://www.jpl.nasa.gov/news/news.php?
  • http://www.sciencedaily.com/releases/2015/01/150116145656.htm
  • http://www.skyatnightmagazine.com/news/machines-used-teach-astronomers-about-stars

Easily deploy & scale your data pipelines in clicks. Run Spark, Kafka, Cassandra + more on shared infrastructure and blow away your data silos. Learn how with Mesosphere DC/OS.

Topics:
artificial intelligence ,machine intelligence ,machine learning ,machine to machine

Opinions expressed by DZone contributors are their own.

THE DZONE NEWSLETTER

Dev Resources & Solutions Straight to Your Inbox

Thanks for subscribing!

Awesome! Check your inbox to verify your email so you can start receiving the latest in tech news and resources.

X

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}