首页> 外文期刊>International Journal of Engineering and Technology >A Metric to Assess the Performance of MLIR Services
【24h】

A Metric to Assess the Performance of MLIR Services

机译:评估MLIR服务性能的指标

获取原文
           

摘要

Information Retrieval plays a vital role in extraction of relevant information. Many researches have been working on for satisfying user needs, tough the problem arises when accessing multilingual information. This Multilingual environment provides a platform where a query can be formed in one language and the result can be in the same language and/or different languages. Performance evaluation of Information Retrieval for monolingual environments especially for English are developed and standardized from its inception. There is no specialized evaluation model available for evaluating the performance of services related to multilingual environments or systems. The unavailability of MLIR domain specific standards is a challenging task. This paper presented enhanced metric to assess the performance of MLIR systems over its counterpart IR metric. This analysis shows that the performance of the enhanced metric is better than the conventional metric. And also these metric can facilitate the researchers and developers to improve the quality of the MLIR systems in the present and future scenarios.
机译:信息检索在提取相关信息中起着至关重要的作用。为了满足用户需求,已经进行了许多研究,但是在访问多语言信息时会出现问题。这种多语言环境提供了一个平台,在该平台上可以用一种语言来形成查询,并且结果可以是相同语言和/或不同语言。从一开始就开发并标准化了单语言环境(尤其是英语)的信息检索性能评估。没有专门的评估模型可用于评估与多语言环境或系统相关的服务的性能。 MLIR域特定标准的不可用是一项艰巨的任务。本文提出了增强的指标,以评估MLIR系统的性能超过其对应的IR指标。该分析表明,增强指标的性能优于常规指标。而且,这些度量标准还可以方便研究人员和开发人员在当前和将来的场景中提高MLIR系统的质量。

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