• Journal of KIISE
• /
• v.43 no.8
• /
• pp.927-932
• /
• 2016

With the explosive increase in the number of online medical documents, the demand for question-answering systems is increasing. Recently, question-answering models based on machine learning have shown high performances in various domains. However, many question-answering models within the medical domain are still based on information retrieval techniques because of sparseness of training data. Based on various information retrieval techniques, we propose an answer snippet retrieval model for question-answering systems of medical documents. The proposed model first searches candidate answer sentences from medical documents using a cluster-based retrieval technique. Then, it generates reliable answer snippets using a re-ranking model of the candidate answer sentences based on various sentence retrieval techniques. In the experiments with BioASQ 4b, the proposed model showed better performances (MAP of 0.0604) than the previous models.

• Asia pacific journal of information systems
• /
• v.25 no.4
• /
• pp.686-711
• /
• 2015

The performance of a novel medical question-answering engine called CliniCluster and existing search engines, such as CQA-1.0, Google, and Google Scholar, was evaluated using known-item searching. Known-item searching is a document that has been critically appraised to be highly relevant to a therapy question. Results show that, using CliniCluster, known-items were retrieved on average at rank 2 ($MRR@10{\approx}0.50$), and most of the known-items could be identified from the top-10 document lists. In response to ill-defined questions, the known-items were ranked lower by CliniCluster and CQA-1.0, whereas for Google and Google Scholar, significant difference in ranking was not found between well- and ill-defined questions. Less than 40% of the known-items could be identified from the top-10 documents retrieved by CQA-1.0, Google, and Google Scholar. An analysis of the top-ranked documents by strength of evidence revealed that CliniCluster outperformed other search engines by providing a higher number of recent publications with the highest study design. In conclusion, the overall results support the use of CliniCluster in answering therapy questions by ranking highly relevant documents in the top positions of the search results.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.191-206
• /
• 2022

Recently, it is a de-facto approach to utilize a pre-trained language model(PLM) to achieve the state-of-the-art performance for various natural language tasks(called downstream tasks) such as sentiment analysis and question answering. However, similar to any other machine learning method, PLM tends to depend on the data distribution seen during the training phase and shows worse performance on the unseen (Out-of-Distribution) domain. Due to the aforementioned reason, there have been many efforts to develop domain-specified PLM for various fields such as medical and legal industries. In this paper, we discuss the training of a finance domain-specified PLM for the Korean language and its applications. Our finance domain-specified PLM, KB-BERT, is trained on a carefully curated financial corpus that includes domain-specific documents such as financial reports. We provide extensive performance evaluation results on three natural language tasks, topic classification, sentiment analysis, and question answering. Compared to the state-of-the-art Korean PLM models such as KoELECTRA and KLUE-RoBERTa, KB-BERT shows comparable performance on general datasets based on common corpora like Wikipedia and news articles. Moreover, KB-BERT outperforms compared models on finance domain datasets that require finance-specific knowledge to solve given problems.