Название: Text Mining Approaches for Biomedical Data
Автор: Aditi Sharan, Nidhi Malik, Hazra Imran, Indira Ghosh
Издательство: Springer
Серия: Transactions on Computer Systems and Networks
Год: 2024
Страниц: 438
Язык: английский
Формат: pdf (true), epub
Размер: 102.8 MB

The book 'Text Mining Approaches for Biomedical Data' delves into the fascinating realm of text mining in healthcare. It provides an in-depth understanding of how Artificial Intelligence (AI) and Machine Learning (ML) are revolutionizing healthcare research and patient care. The book covers a wide range of topics such as mining textual data in biomedical and health databases, analyzing literature and clinical trials, and demonstrating various applications of text mining in healthcare. This book is a guide for effectively representing textual data using vectors, knowledge graphs, and other advanced techniques. It covers various text mining applications, building descriptive and predictive models, and evaluating them. Additionally, it includes building Machine Learning models using textual data, covering statistical and Deep Learning approaches. This book is designed to be a valuable reference for Computer Science professionals, researchers in the biomedical field, and clinicians. It provides practical guidance and promotes collaboration between different disciplines. Therefore, it is a must-read for anyone who is interested in the intersection of text mining and healthcare.

Deep Learning architectures have become increasingly popular in recent years for various Natural Language Processing (NLP) tasks, including NER (Named Entity Recognition) task. Deep Learning architectures, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), can learn complex relationships between input data and output labels, making them well-suited for NER tasks. In this subsection, we will present a comprehensive overview of the various Deep Learning models that have been applied to the task of NER. With the coming of neural network-based Deep Learning models, many advancements came in the field of NLP, and these models are now becoming the state-of-the-art models for most text mining tasks. Though we are discussing it in the context of sequence tagging problems, it is useful for sequence classification and almost all text mining-related tasks because of its sequence-capturing tendency and semantics-retaining property. The main advantage of neural network-based Deep Learning models was to overcome the problem of manual feature extraction and provide an efficient vectorization technique for representing natural language text. These vectors pretrained through Machine Learning can capture the semantics of the text at the text representation level itself.