Definitions for "Data Reduction"
Taking unprocessed astronomical data and getting it in a more understandable form (e.g. calibrating it).
1. The transformation of raw data into a more applicable form. 2. The conversion of all information in a data set into fewer dimensions for a particular purpose, as, for example, a single measure such as a reliability measure.
The process of transforming large quantities of raw data, usually gathered by automatic recording equipment, into useful, condensed, or simplified information.
Compression of sound and image signals.
Digital video that has not been compressed carries more picture information than is necessary to produce a quality image. Digitized video identifies the precise brightness and color of each pixel. It is not necessary to assign this large amount of storage space to each pixel since common picture elements can be grouped together and represented by smaller segments of code. This is the basis of MPEG-2. For example, the picture elements of a static image do not need to be stored as new information over and over to create successive frames. Through the use of I-frames, a reference frame that appears once every fifteen frames, bi-directional B- frames, and P-frames, predictive frames that fill in the information between the I-frames, the amount of data storage necessary to reproduce high quality moving images is greatly reduced.
In digital video and audio transmission or storage systems, a process that eliminates nonvisible or nonaudible aspects of pictures or sound that are not ordinarily perceived because of "masking," allowing a much higher storage density. Data reduction-sometimes called lossy compression-is not the same as data compression. The latter allows the compressed information to be restored to its original status; the former permanently eliminates material that cannot be detected by eye or ear. See also Compression; Masking; PCM.
summarizing a stream of data, often in real time, to extract useful information - such as blocks of data surrounding peaks or triggers
Keywords:  fewer, matrix, sample, analysis, larger
reducing the number of cases or variables in a data matrix. For example, a factor analysis can be used to replace a large collection of variables with a smaller number of factors. A Q-factor analysis is used to replace a larger sample of cases with a fewer number of similar groups.