• KIISE Transactions on Computing Practices
• /
• v.22 no.5
• /
• pp.240-245
• /
• 2016

In this study, comments on restaurants, movies, and mobile devices, as well as tweet messages regardless of specific domains were analyzed for sentimental information content. We proposed a system for extraction of objects (or aspects) and opinion words from each sentence and the subsequent evaluation. For the sentiment analysis, we conducted a comparative evaluation between the Structural SVM algorithm and the Latent Structural SVM. As a result, the latter showed better performance and was able to extract objects/aspects and opinion words using VP/NP analyzed by the dependency parser tree. Lastly, we also developed and evaluated the sentiment detector model for use in practical services.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.385-388
• /
• 2005

감성이란 외부의 자극에 대해 직관적이고 반사적으로 발생하는 저절로 반응하는 현상이다. 감성은 살아온 사회$\cdot$문화적 배경에 따라 흑은 현재 상태에 따라서 다르게 나타난다. 하지만 다소 개인적인 차이가 있을 수 있을지라도 개인이 속한 사회에 따라서 비슷한 상황 아래서는 비슷한 유형의 반응이 나타난다. 현재 감성 인식을 위해서 개인의 행동이나 신체적인 표현을 이용한 감성 인식 연구가 진행 중이다. 이러한 방법은 감성을 표현하는 방식에서 개인차가 커지면 효용성이 떨어질 수밖에 없다. 우리가 거짓말 탐지기를 사용하는 것처럼 본 논문에서는 감정에 따라 달라지는 개인의 생체 신호를 이용해서 감성 인식을 하고자 한다. 이를 위해서 감성에 따른 여러 가지 생체 신호를 추출하고 감성 인식을 위한 생체 신호의 특징점을 파악하고 패턴분류를 하고자 한다.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.3
• /
• pp.148-155
• /
• 2024

Despite the increasing importance of irony and sarcasm detection in the field of natural language processing, research on the Korean language is relatively scarce compared to other languages. This study aims to experiment with various models for irony detection in Korean text. The study conducted irony detection experiments using KoBERT, a BERT-based model, and ChatGPT. For KoBERT, two methods of additional training on sentiment data were applied (Transfer Learning and MultiTask Learning). Additionally, for ChatGPT, the Few-Shot Learning technique was applied by increasing the number of example sentences entered as prompts. The results of the experiments showed that the Transfer Learning and MultiTask Learning models, which were trained with additional sentiment data, outperformed the baseline model without additional sentiment data. On the other hand, ChatGPT exhibited significantly lower performance compared to KoBERT, and increasing the number of example sentences did not lead to a noticeable improvement in performance. In conclusion, this study suggests that a model based on KoBERT is more suitable for irony detection than ChatGPT, and it highlights the potential contribution of additional training on sentiment data to improve irony detection performance.

• Science of Emotion and Sensibility
• /
• v.12 no.4
• /
• pp.557-564
• /
• 2009

For scoring polygraph charts, the Prosecutors' Office of the Republic of Korea uses a consistency criterion in which an elevated signal on one physiological channel is scored as a deceptive response only if the signal is also elevated on other channels. In the current study, the effects of this scoring criterion on reliability and accuracy (validity) of polygraph scores were assessed. Polygraph tests on 26 suspects were evaluated twice by the same examiners. The examiners used the consistency criterion in the first evaluation. In the second evaluation, the examiners were prevented from using the criterion; the signals from each physiological channel were separated and randomly arranged before they were rescored by the same examiner. Reliability was assessed by the variation among the scores for each suspect. Accuracy was assessed by establishing a standard, based on a Latent Class Analysis model, using the results of polygraph tests on each of 182 additional suspects. Reliability and accuracy were both improved by the use of the consistency criterion which therefore was recommended.