Publication
Leveraging context-aware recommender systems for improving personal knowledge assistants by introducing contextual states
Mahta Bakhshizadeh; Christian Jilek; Heiko Maus; Andreas Dengel
In: Thomas Seidl (Hrsg.). LWDA. GI-Workshop-Tage "Lernen, Wissen, Daten, Analysen" (LWDA-2021), September 1-3, Munich, Germany, Pages 1-12, LWDA2021, 2021.
Abstract
During the last decades, recommender systems have played a remarkable role in putting one step further toward making content platforms more intelligent in a wide variety of domains ranging from music and movies to books and documents. Notwithstanding the various applications of recommender systems, not many contributions have been made regarding their potential capabilities in the domain of personal knowledge management. Hence, it has been tried in this study to shed new light on an innovative application of recommender systems to improve personal knowledge assistants by making them capable of providing knowledge workers with useful information through every single situation during their daily work. This paper provides a comprehensive research tree involving the key information about state of the art approaches with a focus on the three most relevant categories to this research including knowledge-based, sequential, and session-based recommender systems. Furthermore, the idea of the contextual states is introduced as the first step of a promising direction toward collecting the required multi-dimensional information for making such helpful recommendations.
Projects
- CoMem - Corporate Memory
- SensAI - Self-organizing Personal Knowledge Assistants in Evolving Corporate Memories