发明名称 Computer implemented method, program, and system for identifying non-text element suitable for communication in multi-language environment
摘要 A computer implemented method, a program, and a system for effectively providing versatile non-text information suitable for use in a multi-language environment. The method includes the steps of: receiving search results of a database using a search criterion in a certain language and a search criterion in another language corresponding to the search criterion in which specific language attributes are associated with non-text elements that are included in the search results; scoring the non-text elements included in the search results depending on a similarity to another element with which a different language attribute is associated; and identifying at least one of the non-text elements included in the search results on the basis of the scores.
申请公布号 US9514127(B2) 申请公布日期 2016.12.06
申请号 US201314024771 申请日期 2013.09.12
申请人 INTERNATIONAL BUSINESS MACHINES CORPORATION 发明人 Katsuno Yasuharu;Miyamoto Kohtaroh;Mizuno Ken;Yoshihama Sachiko
分类号 G06F17/28;G06F17/27;G06F17/30 主分类号 G06F17/28
代理机构 代理人 Quinn David M
主权项 1. A computer implemented method for identifying a non-text element suitable for communication in a multilanguage environment by using a database in which a non-text element can be searched for, the method comprising the steps of: receiving search results of the database using a first search criterion in a certain language and a second search criterion in another language corresponding to the first search criterion, wherein specific language attributes are associated with non-text elements that are included in the search results; scoring the non-text elements included in the search results depending on a similarity to another element with which a different language attribute is associated, wherein at least one non-text element of the non-text elements is an image, and wherein scoring the at least one non-text element further comprises: attempting character recognition of a character included in the image by using an optical character recognition technique;calculating character recognition scores for each character included in the image based on the result of the attempted character recognition, wherein the character recognition score is low if the image includes multiple characters; and identifying at least one of the non-text elements included in the search results on the basis of the scores.
地址 Armonk NY US