Predictive Analysis Research in the UK
April 3, 2013
The field of predictive analysis is proceeding apace, we learn from Science Daily‘s “Predictive Analysis: New Generation of Computational Intelligence Systems.” Predictive analysis, as the name suggests, is the art/science of making predictions (about the past, present, or future) from heaps of data. This article looks at progress being made at the Smart Technology Research Centre, a part of Bournemouth University in Dorset, U.K. The article explains that the researchers:
“. . . are developing computer programs capable of learning. With this intelligent software, computers can make judgments about the quality and reliability of the data they gather. They look for patterns and adapt according to what the information will be used for.”
That would be the work of an “adaptive algorithm” that effectively learns as it encounters data. The Centre has already put some of its findings to work helping companies in tourism and communications, but are looking to build a more general-use system. The write-up relates:
“Building a learning computer system capable of adapting according to the information that is fed into it is no easy task. Most prediction software until recently has been tailor made to solve specific problems. This can make them expensive to maintain and hard to adapt.
“For this, Professor Gabrys and his team have turned to one of the most successful problem solvers on the planet for inspiration — Mother Nature herself. They are building systems which process information in a similar way to the human brain, with its networks of neurons that constantly rewire themselves as we learn.”
Other natural inspirations include genetics and certain animal behaviors, like those of social insects (think ants and bees) or of flocking and swarming creatures (birds and fish.) However, Professor Bogdan Gabrys, the research center’s chair of computational intelligence, emphasizes the limits of the technology—once the system you are studying becomes as complex as, say, financial markets, accurate predictions become unrealistic. The key, he says, is knowing where to draw that line.
Cynthia Murrell, April 03, 2013
Sponsored by ArnoldIT.com, developer of Augmentext
Comments
-
- Subscribe to Beyond Search
Feature archive
News archive
-
