WikiRef: Wikilinks as a route to recommending appropriate references for scientific Wikipedia pages

Proceedings of the 27th International Conference on Computational Linguistics, pages 379–389 Santa Fe, New Mexico, USA, August 20-26, 2018. 379 WikiRef: Wikilinks as a route to recommending appropriate references for scientific Wikipedia pages Abhik Jana IIT Kharagpur, India abhik.jana@iitkgp.ac.in Pawan Goyal IIT Kharagpur, India pawang@cse.iitkgp.ac.in Pranjal Kanojiya IIT Kharagpur, India QUALCOMM India Private Limited pranjal989091@gmail.com Animesh Mukherjee IIT Kharagpur, India animeshm@gmail.com Abstract The exponential increase in the usage of Wikipedia as a key source of scientific knowledge among the researchers is making it absolutely necessary to metamorphose this knowledge repository into an integral and self-contained source of information for direct utilization. Unfortunately, the references which support the content of each Wikipedia entity page, are far from complete. Why are the reference section ill-formed for most Wikipedia pages? Is this section edited as frequently as the other sections of a page? Can there be appropriate surrogates that can auto- matically enhance the reference section? In this paper, we propose a novel two step approach – WikiRef – that (i) leverages the wikilinks present in a scientific Wikipedia target page and, thereby, (ii) recommends highly relevant references to be included in that target page appropri- ately and automatically borrowed from the reference section of the wikilinks. In the first step, we build a classifier to ascertain whether a wikilink is a potential source of reference or not. In the following step, we recommend references to the target page from the reference section of the wikilinks that are classified as potential sources of references in the first step. We perform an extensive evaluation of our approach on datasets from two different domains – Computer Science and Physics. For Computer Science we achieve a notably good performance with a precision@1 of 0.44 for reference recommendation as opposed to 0.38 obtained from the most competitive baseline. For the Physics dataset, we obtain a similar performance boost of 10% with respect to the most competitive baseline. 1 Introduction Wikipedia, the largest online encyclopedia, is a collaborative work of 33,778,487 users and 1,210 ad- mins, with 5,662,889 English articles and a total of 45,132,517 articles in more than 250 languages as of June 7, 20181. Introduced in 2001, it has become one of the most visited websites having global Alexa Rank – 52. Each Wikipedia article contains a rich collection of concepts along with a set of hyper- links associating crucial terms to other wiki pages (termed as wikilinks), which makes the article more comprehensive. In addition, the ability to edit the entity pages easily with proper source citations and to view the edits reflected in the pages timely are some of the key reasons behind Wikipedia’s growth and popularity. As a consequence, Wikipedia has become a destination of all kinds of information about This work is licenced under a Creative Commons Attribution 4.0 International Licence. Licence details: http:// creativecommons.org/licenses/by/4.0/ 1https://en.wikipedia.org/wiki/Wikipedia:Size of Wikipedia 2http://www.alexa.com/siteinfo/www.wikipedia.org 380 entities and events. Further, Wikipedia is being widely utilized by many applications focused on entity linking and disambiguation (Moro et al., 2014; …

Content truncated to 500 words