Overcoming the Imbalance Between Tag Recommendation Approaches and Real-World Folksonomy Structures with Cognitive-Inspired Algorithms

TL;DR

This paper addresses the challenge of recommending tags in social tagging systems by proposing cognitive-inspired algorithms based on ACT-R to better handle the sparsity of real-world folksonomies.

Contribution

It introduces a novel approach using ACT-R's activation equation to improve tag recommendation in sparse folksonomy structures.

Findings

Cognitive-inspired algorithms outperform traditional methods in sparse settings

ACT-R based model effectively models user tag vocabulary

Hybrid approaches can adapt to different folksonomy densities

Abstract

In this paper, we study the imbalance between current state-of-the-art tag recommendation algorithms and the folksonomy structures of real-world social tagging systems. While algorithms such as FolkRank are designed for dense folksonomy structures, most social tagging systems exhibit a sparse nature. To overcome this imbalance, we show that cognitive-inspired algorithms, which model the tag vocabulary of a user in a cognitive-plausible way, can be helpful. Our present approach does this via implementing the activation equation of the cognitive architecture ACT-R, which determines the usefulness of units in human memory (e.g., tags). In this sense, our long-term research goal is to design hybrid recommendation approaches, which combine the advantages of both worlds in order to adapt to the current setting (i.e., sparse vs. dense ones).