Automated MeSH term suggestion for effective query formulation in systematic reviews literature search

• The introduction of the new task suggesting MeSH terms for systematic review literature search, modelled within context an information specialist looking to add a Boolean query. formulation term suggestion methods, based on pre-trained language models, help specialists and researchers construct effective queries creation. An empirical evaluation effectiveness different methods. understanding how suggested by proposed automatic methods differ from those originally selected when formulating queries. High-quality medical reviews require comprehensive searches ensure recommendations outcomes are sufficiently reliable. Indeed, searching relevant is key phase in constructing often involves domain (medical researchers) search (information specialists) experts developing Queries this highly complex, logic, include free-text index standardised terminologies (e.g., Medical Subject Headings (MeSH) thesaurus), difficult time-consuming build. use terms, particular, has been shown improve quality results. However, identifying correct query difficult: unfamiliar with database unsure about appropriateness Naturally, full value terminology not fully exploited. This article investigates suggest initial that includes only terms. In context, we devise lexical models These promise automatically identify inclusion Our study contributes several We further contribute extensive analysis suggestions each method these impact