24小时热门版块排行榜    

查看: 580  |  回复: 1
本帖产生 1 个 翻译EPI ,点击这里进行查看

unitjava

禁虫 (小有名气)

已阅   关注TA 给TA发消息 送TA红花 TA的回帖

terry_well

荣誉版主 (知名作家)

笑熬浆糊——满脑浆糊

优秀版主优秀版主

【答案】应助回帖

★ ★ ★ ★ ★ ★
sltmac(金币+1): 2011-11-29 08:53:15
unitjava(金币+5, 翻译EPI+1): 谢了! 2011-12-01 09:44:18
sltmac(金币+5): 2011-12-04 13:35:53
可能金币给的有点少,没人回答你,呵呵
应该跟我上次翻译的那段属于一篇文章,就帮你到底吧。
In this paper, the contextual information could be classified according their function in the retrieval model. Such contextual information is composed of four parts: information queried by the users, information that is used to filter the searching results, information that affects the sorting results and the recommended proper information. For instance, extending the inquiry by using users’ positions means that when a inquiry is submitted at position A, but it was perceived that the user is at position B by GPS facility, then the perceived position would be also used as a keyword to submit to the system. In order to improve the recall rate, the model could use some of the ontology in the data base to extend the searching scope for the users. In fact, the relationship has been defined for the synonyms, antonyms, word upper and word lower. It is possible to combine the correlative words to submit to the query system.
签名:英文名“Terry”,EC版专用;中文名“浆糊哥”,灌水专用,谢谢合作!
2楼2011-11-29 03:37:20
已阅   关注TA 给TA发消息 送TA红花 TA的回帖
相关版块跳转 我要订阅楼主 unitjava 的主题更新
信息提示
请填处理意见