A Generic Top-N Recommendation Framework For Trading-off Accuracy, Novelty, and Coverage

TL;DR

This paper introduces a flexible recommendation framework that balances accuracy, novelty, and coverage by learning user preferences for long-tail items, improving recommendation diversity and system coverage.

Contribution

It presents a novel re-ranking framework that personalizes the trade-off between accuracy and coverage, enhancing long-tail item recommendation and system scalability.

Findings

Increases recommendation novelty and coverage.

Maintains accuracy while promoting long-tail items.

Enables personalization of non-personalized algorithms.

Abstract

Standard collaborative filtering approaches for top-N recommendation are biased toward popular items. As a result, they recommend items that users are likely aware of and under-represent long-tail items. This is inadequate, both for consumers who prefer novel items and because concentrating on popular items poorly covers the item space, whereas high item space coverage increases providers' revenue. We present an approach that relies on historical rating data to learn user long-tail novelty preferences. We integrate these preferences into a generic re-ranking framework that customizes balance between accuracy and coverage. We empirically validate that our proposedframework increases the novelty of recommendations. Furthermore, by promoting long-tail items to the right group of users, we significantly increase the system's coverage while scalably maintaining accuracy. Our framework also enables personalization of existing non-personalized algorithms, making them competitive with existing personalized algorithms in key performance metrics, including accuracy and coverage.