Ontology-Based Recommendation of Editorial Products

Abstract

Major academic publishers need to be able to analyse their vast catalogue of products and select the best items to be marketed in scientific venues. This is a complex exercise that requires characterising with a high precision the topics of thousands of books and matching them with the interests of the relevant communities. In Springer Nature, this task has been traditionally handled manually by publishing editors. However, the rapid growth in the number of scientific publications and the dynamic nature of the Computer Science landscape has made this solution increasingly inefficient. We have addressed this issue by creating Smart Book Recommender (SBR), an ontologybased recommender system developed by The Open University (OU) in collaboration with Springer Nature, which supports their Computer Science editorial team in selecting the products to market at specific venues. SBR recommends books, journals, and conference proceedings relevant to a conference by taking advantage of a semantically enhanced representation of about 27K editorial products. This is based on the Computer Science Ontology, a very large-scale, automatically generated taxonomy of research areas. SBR also allows users to investigate why a certain publication was suggested by the system. It does so by means of an interactive graph view that displays the topic taxonomy of the recommended editorial product and compares it with the topiccentric characterization of the input conference. An evaluation carried out with seven Springer Nature editors and seven OU researchers has confirmed the effectiveness of the solution.

Pragmatic Ontology Evolution: Reconciling User Requirements and Application Performance

“…For example, we could map the concept temporal logic to each book that contains one of the alternative labels (e.g., "temporal logics") or sub-concepts (e.g., "temporal operators") in the CSO ontology. This technique has been applied with good results in a variety of fields, such as automatic classification of proceedings [2], technology forecasting [6], recommender systems [3,13], community detection [5], and others.…”

“…This is particularly challenging in the field of Computer Science, where new areas evolve constantly and taxonomies tend to become obsolete very quickly [22]. In the context of the collaboration between The Open University and Springer Nature [2,13], we focused on the issue of supporting the evolution of the Computer Science portion of PMC, concentrating in particular on some branches that had become obsolete.…”

“…In particular, by applying the Klink-2 algorithm [22] on the Rexplore dataset [23], we generated the Computer Science Ontology (CSO) [24], a large-scale ontology of research topics in Computer Science, which includes about 26K topics linked by about 226K semantic relationships. CSO powers two tools used by SN for tagging and recommending books: Smart Topic Miner [2] and Smart Book Recommender [13].…”

“…The instances are represented as a vector in which the elements correspond to the concepts in the ontology and the values weigh the importance of a concept. In the case of PMC, we used the Smart Topic API [13] for representing books as a vectors of research topics in which each topic is assigned a value equal to the number of chapters in which it appears. This is a convenient representation that can support several tasks.…”

“…Since this taxonomy is used by a lot of different users and software systems, the evolution process needs to take in consideration both user needs and the impact on applications. For instance, a recommender system for suggesting editorial products described by an ontology [13] would perform differently according to the ontology that it is using. In addition, the process need to be transparent, so that every change can be justified in light of these factors.…”