Man Luu

This current research presents an inventive multilevel named entity recognition scheme for explaining the confrontation with biomedical entity recognition which based on divergent algorithms. The presented scheme contains multilevels, which enables Biomedical entity recognition tasks to extract and identify important biomedical concept: DNA, RNA, CELL-LINE, CELL-TYPE, PROTEIN, and O classes with ease. The BioNLP/NLPBPA 2004 challenge datasets have been used and evaluated, resulted in promising outcomes in terms of biomedical recognition model performance.

In this paper, we propose a novel approach for clinical name entity recognition based on deep machine learning architecture. The proposed scheme based on two different deep learning architectures: the feed forward networks (FFN), and the recurrent neural network (RNN), allow significant improvement in performance, in terms of different performance measures, including precision, recall and F-score, when evaluated with the CLEF 2016 Challenge task 1 A dataset corresponding to Clinical Nursing Handover. It was possible to achieve an F-score of 66% with RNN architecture, which was higher than most of the other participating systems in the Challenge task.

This paper describes a novel machine learning approach based on deeper and wider deep learning model, for better feature learning and latent feature discovery for the clinical name entity recognition task. The performance evaluation of the proposed framework with a benchmark clinical NLP dataset, the clinical CLEF eHealth challenge 2016 dataset, has led to promising performance, when assessed in terms of F-measure, Recall and Precision. The Hybrid CNN model with hyperparameter optimization led to an F-score 89 % for the CLEF eHealth 2016 Challenge task involving synthetic nursing handover dataset.

In this working notes report/paper, we describe the details of two submis- sions for CLEF 2015 eHealth challenge for Task 1a, with details of methods and tools developed for automatic speech recognition of NICTA synthetic nursing handover dataset. The first method involves a novel zero-resource approach based on unsuper- vised acoustic only modeling of speech involving word discovery, and the second method is based on combination of acoustic, language, grammar and dictionary models, using well known open source speech recognition toolkit from CMU, the CMU Sphinx(7). The experimental evaluation of the two methods was done on Challenge dataset (NICTA synthetic nursing handover dataset).

In this paper, we propose a novel multilevel NER framework, for addressing the challenges of clinical name entity recognition, based on different machine learning and text mining algorithms. The proposed framework, with multiple levels, allows models for increasingly complex NER tasks to be built. The experimental evaluation on two different publicly available datasets, corresponding to different application contexts - the CLEF 2016 challenge shared task 1A for nursing handover context, and the BIONLP/NLPBPA 2004 challenge shared task on GENIA corpus for recognizing entities in microbiology, has validated the proposed framework.

This paper describes a novel deep learning-based framework for biomedical name entity recognition. Bio-Entity name entity recognition task based on three different deep learning techniques: Feedforward Networks (FFNs), Recurrent Neural Networks (RNNs), and Hybrid Convolutional Neural Networks (CNNs), has allowed better latent feature learning and discovery for the complex NLP task. The performance evaluation of the proposed framework with the BioNLP dataset corresponding to biomedical entity recognition task, has led to promising performance, when assessed in terms of F-measure, Recall and Precision. The best performing deep learner based on Hybrid CNN approach has resulted in an F-score of 70.32%, and surpassed the performance reported by other participants in the Challenge task.