Data segmenting apparatus and method

摘要

Data segmenting apparatus and method are provided. An apparatus for segmenting an input data sequence includes: a boundary searching device configured to search for one or more candidate boundaries in the data sequence, to obtain one or more candidate segmenting schemes formed by different combinations of these candidate boundaries; and an evaluating device configured to generate, for each candidate segmenting scheme, an evaluation value of each candidate boundary in the scheme by evaluating a segmenting loss caused by using the candidate boundary to segment the data sequence, and determine whether the candidate boundary is valid in the candidate segmenting scheme according to the evaluation value reflecting association relationship between a pair of adjacent data segments adjoining to the candidate boundary and association relationship between each of one or more pairs of non-adjacent data segments, each of which includes two non-adjacent data segments located at two sides of the candidate boundary.

主权项

1. An apparatus for segmenting an input data sequence, comprising:
a boundary searching device, configured to search for one or more candidate boundaries in the input data sequence, in order to obtain one or more candidate segmenting schemes formed by different combinations of these candidate boundaries; an evaluating device, configured to, with respect to each candidate segmenting scheme, generate an evaluation value of each candidate boundary in the candidate segmenting scheme, by evaluating a segmenting loss caused by using the candidate boundary in the candidate segmenting scheme to segment the input data sequence, and determine whether the candidate boundary is valid in the candidate segmenting scheme according to the evaluation value, the evaluation value reflecting association relationship between a pair of adjacent data segments adjoining to the candidate boundary and association relationship between each of one or more pairs of non-adjacent data segments, each pair of non-adjacent data segments comprising two non-adjacent data segments respectively located at two sides of the candidate boundary; an association estimating device, configured to estimate the association relationship between the pair of adjacent data segments adjoining to the candidate boundary and the association relationship between each of the one or more pairs of non-adjacent data segments; a loss calculating device, configured to weight the association relationship between each pair of data segments according to a distance between two data segments in the each pair, and calculate the segment loss according to the weighted association relationships, as the evaluation value of the candidate boundary, wherein the larger the distance is, the smaller a corresponding weight is; and a validity judging device, configured to determine, according to the evaluation value, validity of the candidate boundary in corresponding candidate segmenting scheme.