S URVEY OF NLP IN P HARMACOLOGY:METHODOLOGY,TASKS , R ESOURCES ,K NOWLEDGE , AND TOOLS APREPRINT Dimitar Trajanov 1 , Vangel Trajkovski 1 , Makedonka Dimitrieva 1 , Jovana Dobreva 1 , Milos Jovanovik 1 , Matej Klemen 2 , Aleˇ s ˇ Zagar 2 , Marko Robnik- ˇ Sikonja 2 1 Ss. Cyril and Methodius University in Skopje, Faculty of Computer Science and Engineering, North Macedonia {dimitar.trajanov,jovana.dobreva,milos.jovanovik}@finki.ukim.mk {vangel.trajkovski,makedonka.dimitrieva}@students.finki.ukim.mk 2 University of Ljubljana, Faculty of Computer and Information Science, Slovenia {matej.klemen,ales.zagar,marko.robnik}@fri.uni-lj.si August 23, 2022 ABSTRACT Natural language processing (NLP) is an area of artificial intelligence that applies information tech- nologies to process the human language, understand it to a certain degree, and use it in various appli- cations. This area has rapidly developed in the last few years and now employs modern variants of deep neural networks to extract relevant patterns from large text corpora. The main objective of this work is to survey the recent use of NLP in the field of pharmacology. As our work shows, NLP is a highly relevant information extraction and processing approach for pharmacology. It has been used extensively, from intelligent searches through thousands of medical documents to finding traces of adversarial drug interactions in social media. We split our coverage into five categories to survey modern NLP methodology, commonly addressed tasks, relevant textual data, knowledge bases, and useful programming libraries. We split each of the five categories into appropriate subcategories, de- scribe their main properties and ideas, and summarize them in a tabular form. The resulting survey presents a comprehensive overview of the area, useful to practitioners and interested observers. 1 Introduction Information processing is indispensable to modern drug design, production, and application. A significant amount of information is stored in the textual form and located in scientific papers, clinical notes, ontologies, knowledge bases, social media posts, and newspaper articles. Extraction and retrieval of this information rely on natural language processing (NLP). NLP is a broad scientific area based on computer science, linguistics, and artificial intelligence [99; 100]. As the whole area of artificial intelligence, it has been completely transformed in recent years by deep learning [63]. It has witnessed numerous new techniques and successful applications, such as intelligent search, machine translation, and speech recognition. Many general NLP techniques and approaches can be applied to the pharmacological area. However, often NLP techniques have to be adapted to the specifics of the field in terms of available knowledge sources, text representation, specific methods, terminology, etc. In this work, we survey modern NLP methodology, tasks, resources, knowledge bases, and tools used and adapted to the area of pharmacology. The review aims to inform practitioners working in the area of the pharmacology of exciting recent development and to give a solid starting reference material to new entrants. Several surveys summarise NLP in pharmacology but only cover specific areas of NLP methods. One of the first reviews of NLP for clinical decision support (CDS) [46] was published in 2009. The authors observed that many CDS data is textual and reviewed existing NLP developments for CDS. Luo et al. [132] present a structured review of NLP for narratives in electronic health records (EHR) for pharmacovigilance. Dreisbach et al. [52] review NLP arXiv:2208.10228v1 [cs.CL] 22 Aug 2022