Account-based recommenders in open discovery environments

Abstract

Purpose This paper introduces a machine learning based "My Account" recommender for implementation in open discovery environments such as VuFind, among others.Design/methodology/approach The approach to implementing machine learning based personalized recommenders is undertaken as applied research leveraging data streams of transactional checkout data from discovery systems.Findings The authors discuss the need for large data sets from which to build an algorithm; and introduce a prototype recommender service, describing the prototype's data flow pipeline and machine learning processes. Practical implicationsThe browse paradigm of discovery has neglected to leverage discovery system data to inform the development of personalized recommendations, with this paper, the authors show novel approaches to providing enhanced browse functionality by way of a user account. Originality/valueIn the age of big data and machine learning, advances in deep learning technology and data stream processing make it possible to leverage discovery system data to inform the development of personalized recommendations.

Understanding Artificial Intelligence in Research Libraries – Extensive Literature Review

“…A group of papers (code U-8, 9/126 papers) takes a different approach, arguing for a trade-off between AI and its users. In some cases, users give away their personal data to achieve better service (Alam et al, 2020;Iqbal et al, 2020) and personalised recommendations (Hahn & McDonald, 2018;Hepworth, 2007;Zhu & Wang, 2007) through personalised search engines (Montaner et al, 2003;Porcel et al, 2009). Research has also found methods to use algorithms to develop users' recommendations for relevant books using small amounts of data (Neumann & Geyer-Schulz, 2008).…”

“…Many of the authors address new technology as a useful tool or system (code AI-1, 51/126) or beneficial extension of human skills (code AI-2, 8/126). This viewpoint is most obvious in research and case reports that describe the concept or features of new applications for a variety of library services and operations, including acquisition and circulation (e.g., Iqbal et al, 2020;Ochilbek, 2019;Walker & Jiang, 2019), classification and subject indexing (e.g., Bethard et al, 2009;Golub et al, 2020;, resource retrieval and recommendations (e.g., Färber & Sampath, 2020;Hahn, 2019;Hahn & McDonald, 2018;Smith, 1976), or overall performance analysis (Ennis et al, 2013). These tools are described as automating some laborious or error-prone library operations, enabling faster processes in larger volumes and assisting with librarians' tasks.…”

“…Finally, the analysis revealed the theme of "AI needs" (code D-5, 2/126). From this perspective, AI is an entity with its own need for large quantities of correct data (Brygfjeld et al, 2017;Hahn & McDonald, 2018).…”

Evaluating systematic transactional data enrichment and reuse

“…A previous study on mobile account-based recommender systems detailed the processing and middleware development steps taken to develop such a system [11]. While the case study was descriptive of the machine learning process, it did not undertake the systematic evaluation of the topic outputs of the machine learning processes.…”

“…With these topic metadata clusters a rule set for the recommender system was developed. The prototype recommender study began in October 2016 with seed data of 33,060 consequent subject association rules from initial machine learning processes [11]. These clusters form the basis for the prototype library account-based recommender incorporated into the library mobile app.…”