Redefining Absent Keyphrases and their Effect on Retrieval Effectiveness

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 4185–4193 June 6–11, 2021. ©2021 Association for Computational Linguistics 4185 Redefining Absent Keyphrases and their Effect on Retrieval Effectiveness Florian Boudin LS2N, Université de Nantes, France florian.boudin@univ-nantes.fr Ygor Gallina LS2N, Université de Nantes, France ygor.gallina@univ-nantes.fr Abstract Neural keyphrase generation models have re- cently attracted much interest due to their abil- ity to output absent keyphrases, that is, key- phrases that do not appear in the source text. In this paper, we discuss the usefulness of ab- sent keyphrases from an Information Retrieval (IR) perspective, and show that the commonly drawn distinction between present and absent keyphrases is not made explicit enough. We in- troduce a finer-grained categorization scheme that sheds more light on the impact of absent keyphrases on scientific document retrieval. Under this scheme, we find that only a frac- tion (around 20%) of the words that make up keyphrases actually serves as document expan- sion, but that this small fraction of words is be- hind much of the gains observed in retrieval ef- fectiveness. We also discuss how the proposed scheme can offer a new angle to evaluate the output of neural keyphrase generation models. 1 Introduction Searching the scholarly literature for documents of interest is becoming frustratingly difficult and time- consuming as the volume of published research grows exponentially. One promising approach to address this problem and improve the retrievability of documents is to supplement paper indexing with automatically generated keyphrases (Zhai, 1997; Gutwin et al., 1999; Boudin et al., 2020). Tradi- tionally, keyphrases are defined as a short list of terms that represent the main concepts in a doc- ument (Turney, 2000). In recent years, this defi- nition was further refined to differentiate between keyphrases that are present in the source document or not, and in turn, proposed models for producing keyphrases were divided into extractive (Florescu and Caragea, 2017; Boudin, 2018; Sun et al., 2019; Wang et al., 2020; Santosh et al., 2020, inter alia) and generative models (Meng et al., 2017; Zhao and Zhang, 2019; Chen et al., 2020; Bahuleyan and El Asri, 2020, inter alia) based on their ability to output absent keyphrases. Obviously, keyphrases have different effects on retrieval models depending on whether or not they occur in the document: present keyphrases high- light important parts of the input and make weight- ing terms easier, while absent keyphrases add new terms to the input and provide some form of docu- ment expansion. Intuitively, assigning absent key- phrases is more appealing since it may alleviate the vocabulary mismatch problem between query terms and relevant documents (Furnas et al., 1987), hence enabling the retrieval of relevant documents that otherwise would have been missed. This is especially true for scholarly collections, in which documents are mostly short texts (i.e. scientific abstracts) due to licensing issues and/or resource limitations (Huang et al., 2019). Yet, the extent to which present and absent keyphrases contribute to improved retrieval …

Content truncated to 500 words