Finding Local Experts for Dynamic Recommendations Using Lazy Random Walk

TL;DR

This paper introduces a dynamic recommender system using lazy random walks that effectively suggests local shopping places by leveraging social contact data, outperforming existing methods in accuracy and scalability.

Contribution

It proposes a novel dynamic recommendation algorithm based on lazy random walks that incorporates local and topical authority, demonstrating improved precision and scalability.

Findings

Achieves high precision scores (up to 0.5 at top-1) on Indonesian shopping datasets.

Demonstrates scalability in execution time for larger datasets.

Shows that the system remains effective without frequent database updates.

Abstract

Statistics based privacy-aware recommender systems make suggestions more powerful by extracting knowledge from the log of social contacts interactions, but unfortunately, they are static. Moreover, advice from local experts effective in finding specific business categories in a particular area. We propose a dynamic recommender algorithm based on a lazy random walk that recommends top-rank shopping places to potentially interested visitors. We consider local authority and topical authority. The algorithm tested on FourSquare shopping data sets of 5 cities in Indonesia with k-steps of 5,7,9 of (lazy) random walks and compared the results with other state-of-the-art ranking techniques. The results show that it can reach high score precisions (0.5, 0.37, and 0.26 respectively on precision at 1, precision at 3, and precision at 5 for k=5). The algorithm also shows scalability concerning execution time. The advantage of dynamicity is the database used to power the recommender system; no need to be very frequently updated to produce a good recommendation.