Statistical measure for assisting the multilingual information retrieval system evaluation using Normalized Distance Measure

摘要

Multilingual information retrieval system (MLIR) retrieves relevant information from multiple languages in response to a user query in a single source language. Effectiveness of multilingual information retrieval is measured using Mean Average Precision (MAP). The main feature of multilingual information retrieval is the score list of one language cannot be compared with other language score list. MAP does not consider this feature. We propose a new metric Normalized Distance Measure (NDM) for measuring the effectiveness of MLIR systems. NDM considers the MLIR features. Our analysis states that NDM metric gives credits to MLIR systems that retrieve highly relevant multilingual documents.