PharmKE: Knowledge Extraction Platform for Pharmaceutical Texts Using Transfer Learning

Abstract

Even though named entity recognition (NER) has seen tremendous development in recent years, some domain-specific use-cases still require tagging of unique entities, which is not well handled by pre-trained models. Solutions based on enhancing pre-trained models or creating new ones are efficient, but creating reliable labeled training for them to learn on is still challenging. In this paper, we introduce PharmKE, a text analysis platform tailored to the pharmaceutical industry that uses deep learning at several stages to perform an in-depth semantic analysis of relevant publications. The proposed methodology is used to produce reliably labeled datasets leveraging cutting-edge transfer learning, which are later used to train models for specific entity labeling tasks. By building models for the well-known text-processing libraries spaCy and AllenNLP, this technique is used to find Pharmaceutical Organizations and Drugs in texts from the pharmaceutical domain. The PharmKE platform also incorporates the NER findings to resolve co-references of entities and examine the semantic linkages in each phrase, creating a foundation for further text analysis tasks, such as fact extraction and question answering. Additionally, the knowledge graph created by DBpedia Spotlight for a specific pharmaceutical text is expanded using the identified entities. The obtained results with the proposed methodology result in about a 96% F1-score on the NER tasks, which is up to 2% better than those of the fine-tuned BERT and BioBERT models developed using the same dataset. The ultimate benefits of the platform are that pharmaceutical domain specialists may more easily identify the knowledge extracted from the input texts thanks to the platform’s visualization of the model findings. Likewise, the proposed techniques can be integrated into mobile and pervasive systems to give patients more relevant and comprehensive information from scanned medication guides. Similarly, it can provide preliminary insights to patients and even medical personnel on whether a drug from a different vendor is compatible with the patient’s prescription medication.

Capítulo 6. Inteligencia Artificial y la producción audiovisual

“…El Procesamiento de Lenguaje Natural -NLPpuede utilizar la herramienta Natural Language Toolkit -NLTK-que es una biblioteca de Python que brinda herramientas para trabajar con datos de lenguaje natural, útil en la transcripción automática de diálogos y la generación de subtítulos. También están las herramientas Spacy y AllenNLP -bibliotecas especializadas en tareas de procesamiento de lenguaje natural, como el análisis de sentimientos y la extracción de información (Jofche et al, 2023).…”

VisFormers—Combining Vision and Transformers for Enhanced Complex Document Classification

“…There are various models designed for tasks related to documents, demonstrating how Transformer architectures can be used effectively in various areas. Nasi et al [30] present PharmKE, a knowledge extraction platform for pharmaceutical texts. Leveraging transfer learning, the platform achieves a 96% F1-score in named entity recognition tasks, outperforming fine-tuned BERT and BioBERT models.…”

“…Our work surpasses this limitation, achieving superior performance across diverse document datasets. Universal Document Processing, developed by the author in their paper [30,32], excels in document-related tasks; one limitation is its potential computational intensity during training. However, our model surpasses this challenge, achieving superior performance in terms of computational efficiency and overall accuracy.…”

Review of Natural Language Processing in Pharmacology

“…Frequent information source is the PubMed engine together with the PubTator model (Wei et al, 2019) for automated annotation. The PharmKE tool (Jofche et al, 2023) labels pharmaceutical entities and the relationships between them. In new diseases, such as COVID-19, LBD technique have proved useful to extract relevant information (Martinc et al, 2020;Pinto et al, 2020).…”