Joint Upper & Lower Bound Normalization for IR Evaluation

TL;DR

This paper introduces a new evaluation framework for IR metrics that incorporates query-specific lower-bound normalization, improving the discriminatory power and consistency of metrics like nDCG and MAP across datasets.

Contribution

The paper proposes a novel lower-bound normalization for IR metrics, providing a new perspective and demonstrating its benefits through empirical case-studies.

Findings

UL normalized metrics show higher discriminatory power.

UL normalization reduces significance of differences in uninformative queries.

UL normalized metrics demonstrate better consistency across datasets.

Abstract

In this paper, we present a novel perspective towards IR evaluation by proposing a new family of evaluation metrics where the existing popular metrics (e.g., nDCG, MAP) are customized by introducing a query-specific lower-bound (LB) normalization term. While original nDCG, MAP etc. metrics are normalized in terms of their upper bounds based on an ideal ranked list, a corresponding LB normalization for them has not yet been studied. Specifically, we introduce two different variants of the proposed LB normalization, where the lower bound is estimated from a randomized ranking of the corresponding documents present in the evaluation set. We next conducted two case-studies by instantiating the new framework for two popular IR evaluation metric (with two variants, e.g., DCG_UL_V1,2 and MSP_UL_V1,2 ) and then comparing against the traditional metric without the proposed LB normalization. Experiments on two different data-sets with eight Learning-to-Rank (LETOR) methods demonstrate the following properties of the new LB normalized metric: 1) Statistically significant differences (between two methods) in terms of original metric no longer remain statistically significant in terms of Upper Lower (UL) Bound normalized version and vice-versa, especially for uninformative query-sets. 2) When compared against the original metric, our proposed UL normalized metrics demonstrate higher Discriminatory Power and better Consistency across different data-sets. These findings suggest that the IR community should consider UL normalization seriously when computing nDCG and MAP and more in-depth study of UL normalization for general IR evaluation is warranted.