To achieve a function for obtaining latest information, while keeping versatility by schema-less retrieval, in a history type information storage retrieval system.
A storage device 107 stores a tuple as information to be expressed by the alignment form of a plurality of "attribute=value" pairs including time attributes. A history retrieval processing part 104 selects a tuple group, in which the respective attributes of retrieval expressions to be input are included and also the corresponding values coincide with a corresponding value range condition in the retrieval expressions, from the tuple groups in the storage device 107. While the tuples in a data table 108 are checked one by one from top down, collating with the retrieval object of the retrieval expression is performed, and for the respective coincident tuples, the "attribute=value" pairs, which correspond to the attributes included in the tuples concerning the respective attributes included in key attributes, are extracted. Then, by adopting the alignment as a key, for a latest value table 106, the corresponding values are updated or added by the value.
COPYRIGHT: (C)2011,JPO&INPIT
| JP2009093213A | ||||
| JP2001084476A | ||||
| JP2006195788A |
落合 秀也 外4名,“広域センサネットワークの運用構造と多属性検索”,電子情報通信学会論文誌,日本,社団法人電子情報通信学会,2008年10月 1日,第J91-B巻,第10号,p.1160-1170
Masatake Shiga
Yasuhiko Murayama