Skip to main content



Machine learning and graph technology

http://www.engineerlive.com/content/machine-learning-and-graph-technology

 

The article I found talks about machine learning and networks.  The article talks not about the theory behind networks but about the technology and applications of networks and graphs.  With the explosion of Big Data, networks can be used to help analyze the huge amount of information.  The article begins by talking about new tools and technology that have been developed to work with networks.  Traditional databases are based on tables, which have a rigid structure.  Better suited to keeping track of networks are graph databases which can adapt much more easily to the dynamic nature of networks.

Such databases keep track of nodes an edges, in addition to information about each node.  This has many advantages over a traditional database—information can be accessed quickly, and large scale trends can be identified.  This is particularly useful since identifying trends is incredibly important in business.  Additionally, new nodes can be added very easily, and analysis can be done in real time.

The article then goes on to describe how machine learning can be used in conjunction with networks to gain insight into large scale trends.  This is useful in areas like automated systems where a machine has to independently make decisions.  For example, sensors on a factory floor can be used to create a representation of the machines, how they are connected, and whether there are any problems.  Machine learning is also used by businesses for predictive analysis to better make business decisions.

In class, we discussed overall structure of graphs.  How certain features of a graph can be used to provide valuable information.  For example, I found the social network of a person’s Facebook friends interesting–the clear groupings of friends was apparent, and it was even easy to pick out who this person’s romantic partner was.  It’s cool that computers can be trained to look for the same types of patterns, and that with machine learning, the algorithms can even improve over time.

I found this article particularly interesting because it talks about the real-world impact of network analysis.  Such analysis can be transformative across a wide range of industries.  Additionally, I thought it was cool to read about how technology has evolved to work more efficiently with fluid data structures like graphs.  I am excited to see how this technology will continue to evolve and improve.

Comments

Leave a Reply

Blogging Calendar

September 2016
M T W T F S S
 1234
567891011
12131415161718
19202122232425
2627282930  

Archives