The present study examines the availability and recovery of web references cited in scholarly journals selected based on their high impact factor published between 2008 and 2017. A PHP script was used to crawl the Uniform Resource Locators (URL) collected from the references. A total of 5720 articles were downloaded and 237418 references were extracted. A total of 33512 URLs were checked for their availability. Further the lexical features of URLs like file extension, path depth, character length and top-level domain was determined. The research findings indicated that out of 33512 web references, 20218 contained URLs, DOIs were found in 12799 references and 495 references contained arXiv or WOS identifier. It was found that 29760 URLs were accessible and the remaining 3752 URLs were missing. Most errors were due to HTTP 404 error code (Not found error). The study also tried to recover the inaccessible URLs through Time Travel. Almost 60.55% of inaccessible URLs were archived in various web archives. The findings of the study will be helpful to authors, publishers, and editorial staff to ensure that web references will be accessible in future.
The values of web archives
