Filtering

As our reach extends across the net and across the globe we increasingly need to filter the massive amount of information that is flowing through our personal learning environment and network via all of the various inputs we have established. Spam filters are a widespread manifestation of filtering, and the Bayesian statistical classification algorithms many of them employ have wider applications in the filtering of datasets. NASA employs Bayesian filtering in its AutoClass system for classifying stars according to characteristics to subtle for the human analyst to detect. “AutoClass takes a database of cases described by a combination of real and discrete valued attributes, and automatically finds the natural classes in that data. It does not need to be told how many classes are present or what they look like -- it extracts this information from the data itself.” 1 The advanced search function in Google allows for filtering search results based on simple Boolean rules: ‘ands’, ‘ors’, and ‘nots’ connecting various words and phrases. Google’s advanced scholar search affords even more fine-grained control over search results. As more tools become available for semantic search**, incredibly sophisticated intelligent filtering of content will become routine. 1http://ti.arc.nasa.gov/tech/rse/synthesis-projects-applications/autoclass/