| CPC G06F 16/3323 (2019.01) [G06F 16/313 (2019.01); G06F 16/3346 (2019.01); G06F 40/279 (2020.01); G06V 30/414 (2022.01)] | 12 Claims |
|
1. A document search method, comprising the step of:
selecting a first search criterion for existence of a translation,
preparing a first search text block from a part of a search document;
calculating first relevance of each text block included in a first target to the first search text block by performing full-text search using the first search text block as a second search criterion;
determining a second target from the first target depending on a level of the first relevance;
calculating first similarities of each sentence included in the first search text block to sentences included in the second target;
obtaining a normalization similarity by dividing a sum of the first similarities by a number of sentences; and
retrieving at least one text block similar to the first search text block using the normalization similarity,
wherein the first target is at least some of a plurality of text blocks created from a plurality of search target documents,
wherein the first similarities take into consideration a sequence of words in the sentences,
wherein the first similarities are configured to be determined to be low when the sequence of the words is different, and
wherein the first similarities are used with values higher than or equal to a threshold value.
|