TICTOCRON: an automatic solution for propagating quality metadata to scholarly TOC RSS feed metadata

Institutions and researchers stand to benefit from the facilitation of more widespread syndication of, and easier access to, Table of Content (TOC) RSS (Really Simple Syndication [1]) feeds produced for scholarly journals. However, many journal TOC RSS feeds are at present being produced with erroneous, poor or incomplete metadata. This can hamper the usefulness of scholarly current awareness services, and also cause problems for individual subscribers to those feeds. This is exactly what the ticTOCron software toolkit aims to overcome. The ticTOCron toolkit automatically enhances poor, heterogeneous and incomplete metadata found in TOC RSS feeds by making use of a pre-defined "Best Practice" metadata scheme suitable for scholarly journals. In this work we depict the main issues and "bad practices" found in TOC RSS metadata obtained from more than 435 scholarly publishers. Then, we describe software solutions implemented via ticTOCron. Some references are made to the algorithms for generating semantic relations within, between and from the harvested TOCs and to the mechanisms for propagating "metadata associations" from a previously crawled metadata-rich reference set. However, an effort is made to avoid technical jargon and to replace complex technical descriptions with samples and simple comparisons. The original metadata is converted to a canonical format using the "Best Practices metadata set" for scholarly papers proposed by the ticTOCs Project [2]. We also present the results produced by ticTOCron when it was used for enhancing and normalizing TOC RSS feeds collected from more than 12,000 journals. Finally we propose a sustainable and scalable computational model whereby the automatic solution is complemented and fine-tuned by a cost-effective human cross-validation process.