• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Proceedings of the 73rd ASISamp;T annual meeting: navigating streams in an information ecosystems >ImageSieve: Exploratory Search of Museum Archives with Named Entity-Based Faceted Browsing
【24h】

ImageSieve: Exploratory Search of Museum Archives with Named Entity-Based Faceted Browsing

机译：ImageSieve：具有基于实体的分面浏览的博物馆档案探索性搜索

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Over the last few years, faceted search emerged as anrnattractive alternative to the traditional “text box” search andrnhas become one of the standard ways of interaction onrnmany e-commerce sites. However, these applications ofrnfaceted search are limited to domains where the objects ofrninterests have already been classified along severalrnindependent dimensions, such as price, year, or brand.rnWhile automatic approaches to generate faceted searchrninterfaces were proposed, it is not yet clear to what extentrnthe automatically-produced interfaces will be useful to realrnusers, and whether their quality can match or surpass theirrnmanually-produced predecessors. The goal of this paper isrnto introduce an exploratory search interface calledrnImageSieve, which shares many features with traditionalrnfaceted browsing, but can function without the use ofrntraditional faceted metadata. ImageSieve uses automaticallyrnextracted and classified named entities, which playrnimportant roles in many domains (such as news collections,rnimage archives, etc.). We describe one specific applicationrnof ImageSieve for image search. Here, named entitiesrnextracted from the descriptions of the retrieved images arernused to organize a faceted browsing interface, which thenrnhelps users to make sense of and further explore thernretrieved images. The results of a user study of ImageSieverndemonstrate that a faceted search system based on namedrnentities can help users explore large collections and findrnrelevant information more effectively.
机译：在过去的几年中，多面搜索已成为传统的“文本框”搜索的一种吸引人的替代方法，并且已成为许多电子商务站点之间进行交互的标准方式之一。但是，分面搜索的这些应用仅限于已经按照几个独立的维度（例如价格，年份或品牌）对感兴趣的对象进行了分类的领域。虽然提出了自动生成分面搜索接口的方法，但尚不清楚自动进行分面搜索的程度-生产的界面对实际用户将很有用，以及它们的质量是否可以匹配或超过其人工生产的前任产品。本文的目的是引入一个名为rnImageSieve的探索性搜索界面，该界面与传统的多面浏览具有许多功能，但无需使用传统的多面元数据即可运行。 ImageSieve使用自动提取和分类的命名实体，它们在许多域（例如新闻收藏，图像档案等）中扮演重要角色。我们描述了一种用于图像搜索的特定应用程序ImageSieve。在这里，将从检索到的图像的描述中提取的命名实体用于组织多面浏览界面，然后这将帮助用户理解并进一步浏览检索到的图像。用户对ImageSievernd的研究结果表明，基于命名实体的多面搜索系统可以帮助用户更有效地探索大型馆藏并找到相关信息。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号