• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.479-484
• /
• 2023

다중 레이블 오픈 의도 분류란 다중 의도 분류와 오픈 의도 분류가 합쳐져 오픈 도메인을 가정하고 진행하는 다중 의도 분류 문제이다. 발화 속에는 여러 의도들이 존재한다. 이때 사전에 정의된 의도 여부만을 판별하는 것이 아니라 사전에 정의되어 있는 의도에 대해서만이라도 어떤 의도인지 분류할 수 있어야 한다. 본 논문에서는 발화 속 의도 정보를 활용하여 다중 레이블 오픈 의도를 분류하는 모델을 제안한다. 먼저, 문장의 의도 개수를 예측한다. 그리고 다중 레이블 의도 분류기를 통해 다중 레이블 의도 분류를 진행하여 의도 정보를 획득한다. 획득한 의도 정보 속 다중 의도 개수와 전체 의도 개수를 비교하여 전체 의도 개수가 더 많다면 오픈 의도가 존재한다고 판단한다. 실험 결과 제안한 방법은 MixATIS의 75% 의도에서 정확도 94.49, F1 97.44, MixSNIPS에서는 정확도 86.92, F1 92.96의 성능을 보여준다.

• Journal of Internet Computing and Services
• /
• v.22 no.5
• /
• pp.69-77
• /
• 2021

The main channels of the financial industry are rapidly changing to mobile. In this environment, banks are focusing on information and communication technology to secure their competitiveness, and rapid innovation is being pursued especially in the payment settlement sector. In October 2019, Korea also introduced open banking services to accelerate the financial innovation, such as the open conversion of financial settlement networks and the expansion of the use of simple payments. In this paper, an empirical study was conducted on the effect of domestic open banking quality factors on usage intention. The service quality factors for open banking were classified into interface design, innovation, security, and data sharing, and the technology acceptance model (TAM) was used to verify whether it has a significant effect on perceived convenience, usefulness and intention to use. According to the analysis results, while innovation and security did not have a significant effect on convenience and usefulness, interface design and data sharing were found to have an effect on perceived convenience. The results would provide implications for some quality issues for companies seeking to introduce open banking services as well as for the related academic arena.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.6
• /
• pp.261-266
• /
• 2016

Chart suggestion method provided by various existing data visualization tools makes chart recommendations without considering the user intention. Data visualization is not properly carried out and thus, unclear in some tools because they do not follow the segmented quantitative data classification policy. This paper provides a guideline that clearly classifies the quantitative input data and that effectively suggests charts based on user intention. The guideline is two-fold; the analysis guideline examines the quantitative data and the suggestion guideline recommends charts based on the input data type and the user intention. Following this guideline, we excluded charts in disagreement with the user intention and confirmed that the time user spends in the chart selection process has decreased.

• Information Systems Review
• /
• v.11 no.3
• /
• pp.125-146
• /
• 2009

The emergence of open source software(OSS) with its most prominent advantages creates a vast interest among practitioners. A study on Linux, the most well-known OSS, estimated that it would cost as 5.4 billion Euros taking over 73,000 person-years if it had been developed by conventional means. However, Linux has achieved only 0.65 percent of the operating system market for individual users while Microsoft windows family counts for nearly 90 percent of the market. Much of the effort being spent in the development of OSS is going to waste and potential value that OSS can bring to users is not being realized. Adoption of OSS is often accompanied by the discontinuance of existing software that is already in place. If users resist changing, they may not adopt OSS. Using the case of Linux, this study examines user resistance to change from the commercial operating software to the free operating system. This study identifies six sub-types of switching costs (uncertainty, emotional, setup, learning, lost benefit, and sunk costs) and tests their effects on user resistance to change based on a survey of 201 users. The results show that user resistance to change has a negative impact on the adoption of OSS. Further, this study shows that uncertainty and emotional costs have significant effects on user resistance to change. Beyond previous research on technology adoption, this research contributes towards an understanding of the switching costs leading to user resistance to change and offers suggestions to OSS practitioners for developing strategies to improve the adoption of OSS.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.461-466
• /
• 2021

본 연구에서는 DECO(Dictionnaire Electronique du COreen) 한국어 전자사전과 LGG(Local-Grammar Graph)에 기반한 반자동 언어데이터 증강(Semi-automatic Symbolic Propagation: SSP) 방식에 입각하여, 핀테크 분야의 CS(Customer Service) 챗봇 NLU(Natural Language Understanding)을 위한 주석 학습 데이터를 효과적으로 생성하고, 이를 기반으로 RASA 오픈 소스에서 제공하는 DIET(Dual Intent and Entity Transformer) 아키텍처를 활용하여 핀테크 CS 챗봇 NLU 시스템을 구현하였다. 실 데이터을 통해 확인된 핀테크 분야의 32가지의 토픽 유형 및 38가지의 핵심 이벤트와 10가지 담화소 구성에 따라, DECO-LGG 데이터 생성 모듈은 질의 및 불만 화행에 대한 양질의 주석 학습 데이터를 효과적으로 생성하며, 이를 의도 분류 및 Slot-filling을 위한 개체명 인식을 종합적으로 처리하는 End to End 방식의 멀티태스크 트랜스포머 모델 DIET로 학습함으로써 DIET-only F1-score 0.931(Intent)/0.865(Slot/Entity), DIET+KoBERT F1-score 0.951(Intent)/0.901(Slot/Entity)의 성능을 확인하였으며, DECO-LGG 기반의 SSP 생성 데이터의 학습 데이터로서의 효과성과 함께 KoBERT에 기반한 DIET 모델 성능의 우수성을 입증하였다.