• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.251-273
• /
• 2022

Recently, research on applying text analysis to deep learning has steadily continued. In particular, researches have been actively conducted to understand the meaning of words and perform tasks such as summarization and sentiment classification through a pre-trained language model that learns large datasets. However, existing pre-trained language models show limitations in that they do not understand specific domains well. Therefore, in recent years, the flow of research has shifted toward creating a language model specialized for a particular domain. Domain-specific pre-trained language models allow the model to understand the knowledge of a particular domain better and reveal performance improvements on various tasks in the field. However, domain-specific further pre-training is expensive to acquire corpus data of the target domain. Furthermore, many cases have reported that performance improvement after further pre-training is insignificant in some domains. As such, it is difficult to decide to develop a domain-specific pre-trained language model, while it is not clear whether the performance will be improved dramatically. In this paper, we present a way to proactively check the expected performance improvement by further pre-training in a domain before actually performing further pre-training. Specifically, after selecting three domains, we measured the increase in classification accuracy through further pre-training in each domain. We also developed and presented new indicators to estimate the specificity of the domain based on the normalized frequency of the keywords used in each domain. Finally, we conducted classification using a pre-trained language model and a domain-specific pre-trained language model of three domains. As a result, we confirmed that the higher the domain specificity index, the higher the performance improvement through further pre-training.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.143-148
• /
• 2023

본 논문에서는 금융 도메인 특화 사전학습 언어모델인 KF-DeBERTa(Korean Finance DeBERTa)를 제안한다. KF-DeBERTa는 대규모의 금융 말뭉치를 기반으로 학습하였으며, Transformer 아키텍처와 DeBERTa의 특징을 기반으로 구성되었다. 범용 및 금융 도메인에 대한 평가에서 KF-DeBERTa는 기존 언어모델들에 비해 상당히 높은 성능을 보였다. 특히, 금융 도메인에서의 성능은 매우 두드러졌으며, 범용 도메인에서도 다른 모델들을 상회하는 성능을 나타냈다. KF-DeBERTa는 모델 크기 대비 높은 성능 효율성을 보여주었고, 앞으로 금융 도메인에서의 활용도가 기대된다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.3
• /
• pp.79-87
• /
• 2024

Recently, the importance of pre-trained language models (PLM) has been emphasized for natural language processing (NLP) such as text classification, sentiment analysis, and question answering. Korean PLM shows high performance in NLP in general-purpose domains, but is weak in domains such as finance, medicine, and law. The main goal of this study is to propose a language model learning process and method to build a financial-specific language model that shows good performance not only in the financial domain but also in general-purpose domains. The five steps of the financial-specific language model are (1) financial data collection and preprocessing, (2) selection of model architecture such as PLM or foundation model, (3) domain data learning and instruction tuning, (4) model verification and evaluation, and (5) model deployment and utilization. Through this, a method for constructing pre-learning data that takes advantage of the characteristics of the financial domain and an efficient LLM training method, adaptive learning and instruction tuning techniques, were presented.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.619-623
• /
• 2022

대화생성은 대규모 학습 데이터로부터 사전 학습된 언어모델을 활용한 도전적인 다운스트림 태스크 중 하나이다. 대화에서 특정한 지식에 대한 맥락이 보존된 응답 문장을 생성하기 위한 기술의 일환으로써 지식 기반 대화생성이 연구되고 있으며, 현업에서는 사업목표에 따른 대화 서비스를 제공하는 목적으로 이러한 기술을 적용할 수 있다. 본 논문에서는, 각각의 서비스 도메인에 특화된 모델을 적절히 활용 가능하도록 전체 데이터를 도메인별로 구분하여 학습한 다수의 대화생성 모델을 구축한다. 또한, 특정 도메인의 데이터로 학습된 모델이 나머지 도메인에서 어떤 수준의 대화생성이 가능한지 비교 분석함으로써 개별 학습된 모델들이 도메인의 특성에 따라 서로 다른 영향력이나 연관성을 나타낼 가능성을 확인한다. 이러한 실험적인 분석 결과를 바탕으로 현업의 서비스에서 개별 도메인에 특화된 모델이 적절히 활용하는 것이 유용함을 확인하고자 한다.

• 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.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.369-373
• /
• 2022

기존의 관계 추출 태스크에서의 많은 연구들은 사전학습 언어모델을 파인튜닝하여 뛰어난 성능을 달성해왔다. 하지만, 파인튜닝은 사전학습 시의 학습 기법과의 간극으로 인해 일반화 능력을 저해한다. 본 연구는 다운스트림 태스크를 사전학습의 Masked Language Modeling (MLM) 기법을 통해 해결하는 프롬프트 기반의 학습 기법을 활용하여, 특정 한국어 도메인에서의 관계 추출을 위한 프롬프트 기반 파인튜닝 방법론을 제안한다. 실험의 경우, 도메인의 특성이 뚜렷한 전통문화유산 말뭉치를 대상으로 실험을 진행하여 본 방법론의 도메인 적응력을 보이며, 메타 정보 즉, 개체 유형 및 관계 유형의 의미론적 정보를 일종의 지식 정보로 활용하여 프롬프트 기반 지식 주입의 효과성을 검증한다. 프롬프트에의 메타 정보의 주입과 함께 프롬프트 기반으로 파인튜닝된 모델은 오직 MLM 기법만을 이용하여 태스크를 수행하여 기존 파인튜닝 방법론 대비 파라미터 수가 적음에도, 성능 면에서 대부분 소폭 상승하는 경향을 보여줌으로써 그 효과성 및 효율성을 보인다.

• Journal of the Korea Convergence Society
• /
• v.13 no.2
• /
• pp.13-20
• /
• 2022

This study is to develop a named entity recognition model specialized in criminal investigation domains using deep learning techniques. Through this study, we propose a system that can contribute to analysis of crime for prevention and investigation using data analysis techniques in the future by automatically extracting and categorizing crime-related information from text-based data such as criminal judgments and investigation documents. For this study, the criminal investigation domain text was collected and the required entity name was newly defined from the perspective of criminal analysis. In addition, the proposed model applying KoELECTRA, a pre-trained language model that has recently shown high performance in natural language processing, shows performance of micro average(referred to as micro avg) F1-score 98% and macro average(referred to as macro avg) F1-score 95% in 9 main categories of crime domain NER experiment data, and micro avg F1-score 98% and macro avg F1-score 62% in 56 sub categories. The proposed model is analyzed from the perspective of future improvement and utilization.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.600-602
• /
• 2023

사전 학습된 언어 모델은 최근 다양한 도메인 및 응용태스크에 활용되고 있다. 하지만 언어 모델을 활용한 문장 난이도 측정 태스크에 대해서는 연구가 수행된 바 없다. 이에 본 논문에서는 교과서 데이터를 활용해 문장 난이도 데이터 셋을 구축하고, 일반 말뭉치로 훈련된 BERT 모델과 교과서 텍스트를 활용해 적응 학습한 BERT 모델을 문장 난이도 측정 태스크에 대해 미세 조정하여 성능을 비교했다.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.15-23
• /
• 2022

In the field of patents, as NLP(Natural Language Processing) is a challenging task due to the linguistic specificity of patent literature, there is an urgent need to research a language model optimized for Korean patent literature. Recently, in the field of NLP, there have been continuous attempts to establish a pre-trained language model for specific domains to improve performance in various tasks of related fields. Among them, ELECTRA is a pre-trained language model by Google using a new method called RTD(Replaced Token Detection), after BERT, for increasing training efficiency. The purpose of this paper is to propose KorPatELECTRA pre-trained on a large amount of Korean patent literature data. In addition, optimal pre-training was conducted by preprocessing the training corpus according to the characteristics of the patent literature and applying patent vocabulary and tokenizer. In order to confirm the performance, KorPatELECTRA was tested for NER(Named Entity Recognition), MRC(Machine Reading Comprehension), and patent classification tasks using actual patent data, and the most excellent performance was verified in all the three tasks compared to comparative general-purpose language models.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.57-67
• /
• 2024

This research investigates training methods for large language models to accurately identify sentiments and comprehend information about increasing and decreasing fluctuations in the financial domain. The main goal is to identify suitable datasets that enable these models to effectively understand expressions related to financial increases and decreases. For this purpose, we selected sentences from Wall Street Journal that included relevant financial terms and sentences generated by GPT-3.5-turbo-1106 for post-training. We assessed the impact of these datasets on language model performance using Financial PhraseBank, a benchmark dataset for financial sentiment analysis. Our findings demonstrate that post-training FinBERT, a model specialized in finance, outperformed the similarly post-trained BERT, a general domain model. Moreover, post-training with actual financial news proved to be more effective than using generated sentences, though in scenarios requiring higher generalization, models trained on generated sentences performed better. This suggests that aligning the model's domain with the domain of the area intended for improvement and choosing the right dataset are crucial for enhancing a language model's understanding and sentiment prediction accuracy. These results offer a methodology for optimizing language model performance in financial sentiment analysis tasks and suggest future research directions for more nuanced language understanding and sentiment analysis in finance. This research provides valuable insights not only for the financial sector but also for language model training across various domains.