article · Big Data and Cognitive Computing
With the daily influx of scientific publications, identifying relevant literature remains difficult for researchers who often rely on basic keyword searches or manual browsing. To address this, an evaluation of collaborative filtering, content-based filtering, and hybrid recommendation approaches was conducted using two distinct datasets. The first dataset contains 1,895 users and 3,122 articles from the CI&T Deskdrop collection, while the second includes 7,947 users and 25,975 articles from CiteULike-t. These recommendation techniques automatically suggest articles by analysing user preferences and historical behaviour. Evaluation across accuracy, ranking quality, and novelty demonstrates that the hybrid approach significantly outperforms individual methods. Furthermore, the hybrid technique helps resolve common challenges in recommendation systems, including cold start and data sparsity problems, offering practical value for software tools designed to improve scientific content discovery and researcher productivity.
Researchers face significant difficulties navigating vast volumes of newly published scientific literature using conventional search tools. Identifying effective recommendation methods helps improve content discovery and researcher productivity. Demonstrating that hybrid filtering models successfully handle sparse data and new users provides clearer pathways for designing platforms that deliver relevant scholarly articles more accurately and efficiently.
The findings can inform the design of scientific literature recommendation tools and research platforms aimed at improving discovery and productivity for researchers. Tested on historical collections of up to 25,975 articles and 7,947 users, the work is at an applied research stage, establishing that hybrid algorithms are viable candidates for integration into production-grade scientific discovery and academic reference systems.
AI-generated from the published abstract. Always read the original work before citing.
Given the daily influx of scientific publications, researchers often face challenges in identifying relevant content amid the vast volume of available information, typically resorting to conventional methods like keyword searches or manual browsing. Utilizing a dataset comprising 1895 users and 3122 articles from the CI&T Deskdrop collection, as well as 7947 users and 25,975 articles from CiteULike-t, we examine the effectiveness of collaborative filtering and content-based and hybrid recommendation approaches in scientific literature recommendations. These methods automatically generate article suggestions by analyzing user preferences and historical behavior. Our findings, evaluated based on accuracy (Precision@K), ranking quality (NDCG@K), and novelty, reveal that the hybrid approach significantly outperforms other methods, tackling some challenges such as cold starts and sparsity problems. This research offers theoretical insights into recommendation model effectiveness and practical implications for developing tools that enhance content discovery and researcher productivity.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.3390/bdcc8120190
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
