• 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.

• Journal of Internet Computing and Services
• /
• v.24 no.4
• /
• pp.127-135
• /
• 2023

This research paper explores the application of FinBERT, a variational BERT-based model pre-trained on financial domain, for sentiment analysis in the financial domain while focusing on the process of identifying suitable training data and hyperparameters. Our goal is to offer a comprehensive guide on effectively utilizing the FinBERT model for accurate sentiment analysis by employing various datasets and fine-tuning hyperparameters. We outline the architecture and workflow of the proposed approach for fine-tuning the FinBERT model in this study, emphasizing the performance of various datasets and hyperparameters for sentiment analysis tasks. Additionally, we verify the reliability of GPT-3 as a suitable annotator by using it for sentiment labeling tasks. Our results show that the fine-tuned FinBERT model excels across a range of datasets and that the optimal combination is a learning rate of 5e-5 and a batch size of 64, which perform consistently well across all datasets. Furthermore, based on the significant performance improvement of the FinBERT model with our Twitter data in general domain compared to our news data in general domain, we also express uncertainty about the model being further pre-trained only on financial news data. We simplify the complex process of determining the optimal approach to the FinBERT model and provide guidelines for selecting additional training datasets and hyperparameters within the fine-tuning process of financial sentiment analysis models.

• Journal of Digital Convergence
• /
• v.20 no.3
• /
• pp.131-143
• /
• 2022

Focusing on FinTech keywords, this study is analyzing newspaper articles and Twitter data by using text mining methodology in order to understand trends in the industry of domestic digital financial service. In the growth of FinTech lifecycle, the frequency analysis has been performed by four important points: Mobile Payment Service, Internet Primary Bank, Data 3 Act, MyData Businesses. Utilizing frequency analysis, which combines the keywords 'China', 'USA', and 'Future' with the 'FinTech', has been predicting the FinTech industry regarding of the current and future position. Next, sentiment analysis was conducted on Twitter to quantify consumers' expectations and concerns about FinTech services. Therefore, this study is able to share meaningful perspective in that it presented strategic directions that the government and companies can use to understanding future FinTech market by combining frequency analysis and sentiment analysis.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.232-237
• /
• 2022

본 논문에서는 미세 조정(fine-tuning) 및 비지도 학습 기법을 사용하여 경제 분야 텍스트인 금융 리포트에 대해 속성 기반 감성 분석(aspect-based sentiment analysis) 데이터셋을 반자동적으로 구축할 수 있는 방법론에 대한 연구를 수행하였다. 구축 시에는 속성기반 감성분석 주석 요소 중 극성, 속성 카테고리 정보를 부착하였으며, 미세조정과 비지도 학습 기법인 BERTopic을 통해 주석 요소를 자동적으로 부착하는 한편 이를 수동으로 검수하여 데이터셋의 완성도를 높이고자 하였다. 데이터셋에 대한 실험 결과, 극성 반자동 주석의 경우 기존에 구축된 데이터셋과 비슷한 수준의 성능을 보였다. 한편 정성적 분석을 통해 자동 구축을 동일하게 수행하였더라도 기술의 원리와 발달 정도에 따라 결과가 상이하게 달라짐을 관찰함으로써 경제 도메인의 ABSA 데이터셋 구축에 여전히 발전 여지가 있음을 확인할 수 있었다.

• Journal of Digital Convergence
• /
• v.17 no.9
• /
• pp.47-54
• /
• 2019

This study confirms the polarity of news articles on apartment prices using Opinion Mining which has widely been used for a big data analysis. The analyses were carried out utilizing internet news articles posted on the Naver for two years: 2012 and 2018. We proposed a sentiment analysis model and modeled a topic-oriented sentiment dictionary construction methods. As a result of analyzing the proposed sentiment analysis model, it was confirmed that there was a difference according to the tendency of the media companies in selecting social issues at the time of rising apartment prices. At the same time, we were able to find more affirmative articles in the media companies which share similar sentiment with the government in charge. In this paper, we proposed a sentiment analysis model that can be used in real estate field and analyzed the polarity of unformatted data related to real estate. In order to integrate them into various fields in the future, it is necessary to build the sentiment dictionaries by themes, as well as to collect various unformatted data over extended periods.

• Science of Emotion and Sensibility
• /
• v.11 no.4
• /
• pp.599-612
• /
• 2008

Today, communication trend of financial brand has changed fast with more foreign financial brand's introduction, emerging financial brands through the openness. With the trend of changing, companies are introducing various marketing methods to differentiate its brand image. And color marketing becomes an important tool for the differentiation. However, except a few brands, brand color which expresses management character of a company is different from the customer's preferred color which is based on investors' investment tendency. This may be related to the brand Image which is final goal of communication. Therefore, this study suggests effective communication method between company and customers by analyzing preferred color of customers by their investment tendency and comparison analysis security firms' color marketing strategy. As a result, it was found that Roland Barthes symbolic meaning of colors is different from the symbolic meaning of the groups of investor tendency. For example, I assumed that aggressive investors preferred strong color like red or orange, but the survey result was far from my assumption. I hope this study can be a good foundation for logical and scientific marketing in communication between security companies and customers in more open market with introduction to the Capital Market Consolidation Act.

• 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.

• Journal of the HCI Society of Korea
• /
• v.14 no.2
• /
• pp.61-69
• /
• 2019

Recently, as a new ICT trend, emerging chatbots are actively introduced in the field of finance. Chatbot conducts services through the interaction of communication with users. The purpose of this study is to investigate the effect of interaction dialogue type on the efficiency, usability, sensibility and perceived security of financial service chatbot. Based on theoretical considerations, I have divided into closed conversation, open conversation, and mixed conversation type based on the conversation style based on the implementation method of chatbot. Three types of Financial Chatbot prototypes were made and the experiments were conducted after account inquiry, account transfer, Q & A financial task execution. As a result of experimental research analysis, chatbot's interaction dialogue type was found to affect efficiency and usability. Users have shown that the interaction of closed conversations and mixed conversations is an intuitive interface that allows financial services to be easily manipulated without error. This study will be used as a resource to improve the user experience that requires deep understanding of financial chatbot users who should consider both the emotional element of artificial intelligence that provides services through natural conversation and the functional elements that perform financial business can be.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.16 no.6
• /
• pp.141-155
• /
• 2017

As the use of SNS becomes more active, many people are posting their thoughts about specific events in their SNS in the form of text. As a result, SNS is used in various fields such as finance and distribution to conduct service satisfaction surveys and consumer monitoring. However, in the transportation area, there are not enough cases to utilize unstructured data analysis such as emotional analysis. In this study, we developed an emotional analysis methodology that can be used in transportation by using highway VOC data, which is atypical data collected by Korea Expressway Corporation. The developed methodology consists of morpheme analysis, emotional dictionary construction, and emotional discrimination of the collected unstructured data. The developed methodology was verified using highway related tweet data. As a result of the analysis, it can be guessed that many information and information about the construction and the accident were related to the highway during the analysis period. Also, it seems that users complain about the delay caused by construction and accident.

• The Korean Journal of Applied Statistics
• /
• v.35 no.2
• /
• pp.189-201
• /
• 2022

In this study, we propose factor augmentation to improve forecasting power of cryptocurrency return. We consider financial and economic variables as well as psychological aspect for possible factors. To be more specific, financial and economic factors are obtained by applying principal factor analysis. Psychological factor is summarized by news sentiment analysis. We also visualize such factors through impulse response analysis. In the modeling perspective, we consider ARIMAX as the classical model, and random forest and deep learning to accommodate nonlinear features. As a result, we show that factor augmentation reduces prediction error and the GRU performed the best amongst all models considered.