• Journal of the Korea Society of Computer and Information
• /
• v.29 no.9
• /
• pp.69-77
• /
• 2024

This study evaluates the performance of second language acquisition models learning Korean and English using the GPT-2-Small model, analyzing the impact of various training conditions on performance. Four training conditions were used: monolingual learning, sequential learning, sequential-interleaved learning, and sequential-EWC learning. The model was trained using datasets from the National Institute of Korean Language and English from BabyLM Challenge, with performance measured through PPL and BLiMP metrics. Results showed that monolingual learning had the best performance with a PPL of 16.2 and BLiMP accuracy of 73.7%. In contrast, sequential-EWC learning had the highest PPL of 41.9 and the lowest BLiMP accuracy of 66.3%(p < 0.05). Monolingual learning proved most effective for optimizing model performance. The EWC regularization in sequential-EWC learning degraded performance by limiting weight updates, hindering new language learning. This research improves understanding of language modeling and contributes to cognitive similarity in AI language learning.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.450-452
• /
• 2020

기술의 발전과 함께 사용자에게 가까이 자리 잡은 소셜 네트워크 서비스는 이미지, 동영상, 텍스트 등 활용 가능한 데이터의 수를 폭발적으로 증가시켰다. 작성자의 감정을 포함하고 있는 텍스트 데이터는 시장 조사, 주가 예측 등 다양한 분야에서 이용할 수 있으며, 이로 인해 긍부정의 이진 분류가 아닌 다중 감정 분석의 필요성 또한 높아지고 있다. 본 논문에서는 딥러닝 기반 감정 분류에 심리학 이론의 기반 감정 모델을 활용한 결합 모델과 단일 모델을 비교한다. 학습을 위해 AI Hub에서 제공하는 데이터와 노래 가사 데이터를 복합적으로 사용하였으며, 결과에서는 대부분의 경우에 결합 모델이 높은 결과를 보였다.

• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1045-1055
• /
• 2007

Latent topic models are statistical models which automatically captures salient patterns or correlation among features underlying a data collection in a probabilistic way. They are gaining an increased popularity as an effective tool in the application of automatic semantic feature extraction from text corpus, multimedia data analysis including image data, and bioinformatics. Among the important issues for the effectiveness in the application of latent topic models to the massive data set is the efficient learning of the model. The paper proposes an accelerated learning technique for PLSA model, one of the popular latent topic models, by an incremental EM algorithm instead of conventional EM algorithm. The incremental EM algorithm can be characterized by the employment of a series of partial E-steps that are performed on the corresponding subsets of the entire data collection, unlike in the conventional EM algorithm where one batch E-step is done for the whole data set. By the replacement of a single batch E-M step with a series of partial E-steps and M-steps, the inference result for the previous data subset can be directly reflected to the next inference process, which can enhance the learning speed for the entire data set. The algorithm is advantageous also in that it is guaranteed to converge to a local maximum solution and can be easily implemented just with slight modification of the existing algorithm based on the conventional EM. We present the basic application of the incremental EM algorithm to the learning of PLSA and empirically evaluate the acceleration performance with several possible data partitioning methods for the practical application. The experimental results on a real-world news data set show that the proposed approach can accomplish a meaningful enhancement of the convergence rate in the learning of latent topic model. Additionally, we present an interesting result which supports a possible synergistic effect of the combination of incremental EM algorithm with parallel computing.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.11
• /
• pp.207-214
• /
• 2012

This study tried to propose an optimal instruction-learning model for the cyber home learning system2.0 through grounded theory. In-depth interviews were conducted to investigate causes of underachievement and the causes were categorized according to common concepts. A total of 25 causes of underachievement could be grouped into four categories and eight sub-categories, as a result. Underachievers, then, participated in the lessons utilizing the cyber home learning system2.0 and their cognitive change process about learning was analyzed from reflectional journals and in-depth interviews with a teacher. It was found that underachievers were participated in learning by passing through 5 processes; adaptation to the cyber home learning system2.0, basic knowledge learning, task implementing, rounds of group discussions, feedbacks and evaluation. Based on analysis of these five processes, this study proposed a conditional matrix for the cyber home learning system 2.0 as the most personalized model for underachieving students.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.251-253
• /
• 1999

가상현실을 이용한 교육시스템에서 학습자를 대신한 에이전트의 설계에 대한 많은 연구가 이루어지고 있다. 이러한 에이전트의 설계에 있어서 가장 중요한 점은 동적으로 변화되는 환경에서 이루어지는 다양한 현상과 사건들을 인식하는 방법과 이에 대해 반응하는 행동의 설계와 구현에 있다. 본 논문에서는 에이전트의 행동을 설계함에 있어서 교육 시스템의 목적에 부합하기 위한 주위 객체들과의 상호작용을 통한 행위의 인식, 장이론을 이용한 공간객체와의 상호작용 등을 이용하여 좀 더 효율적으로 주위 환경에 반응할 수 있는 자율 에이전트의 행위에 대한 방안을 제시한다.

• International Commerce and Information Review
• /
• v.16 no.2
• /
• pp.23-44
• /
• 2014

This study examined the relationship between entrepreneurship, learning competence and the export performance of international new ventures that enterprises entering foreign markets immediately after starting their business in Korea, and the mediating effect of learning competence. A research framework was developed with two entrepreneurship factors(risk taking and innovativeness), learning competence factors, and export performance. Six hypotheses were proposed and tested on 115 samples using LISREL. Research findings suggest that the two entrepreneurship factors positively influenced learning competence. However, export performance was affected by only innovativeness. Also, the learning competence positively affected export performance. Learning competence was found to mediate between entrepreneurship and export performance.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.2
• /
• pp.456-464
• /
• 1997

We describe a random prcess model that prvides sets of patterms whth prcisely contrlolled within-class varia-bility and between-class distinctions.We used these pattems in a simulation study wity the back-propagation netwoek to chracterize its perfotmance as we varied the process-controlling parameters,the statistical differences between the processes,and the random noise on the patterns.Our results indicated that grneralized statistical difference between the processes genrating the patterns provided a good predictor of the difficulty of the clssi-fication problem. Also we analyzed the performance of the Bayes classifier whith the maximum-likeihood cri-terion and we compared the performance of the neural network to that of the Bayes classifier.We found that the performance of neural network was intermediate between that of the simulated and theoretical Bayes classifier.

• Annual Conference of KIPS
• /
• 2022.05a
• /
• pp.705-707
• /
• 2022

인체의 흉부 X-ray 영상으로부터 척추질환과 관련된 의료 진단지표를 자동으로 추출하는 과정을 위하여 흉추조직의 정확한 분할이 필요하다. 본 연구에서는 HRNet 기반의 학습을 통하여 흉추조직을 분할하는 방법을 고찰한다. 분할 과정에서 영상 내의 상대적인 위치 정보가 효과적으로 반영될 수 있도록, 계층별로 영상의 고해상도의 표현이 그대로 유지되는 구조와 저해상도의 특징 지도로 변환되는 구조가 병렬적으로 연결되는 형태의 심층 신경망 모델을 채택하였다. 흉부 X-ray 영상에서 콥각도(Cobb's angle)를 산출하는 문제를 대상으로 흉추 분할을 위한 학습 방법, 진단지표 추출 방법 등을 소개하며, 부수적으로 피사체의 위치 변화 및 크기 변화 등에 강인한 성능을 제공하기 위하여 학습 데이터를 증강하는 방법론을 제시하였다. 총 145개의 영상을 사용한 실험을 통하여 제안된 이론의 타당성을 평가하였다.

• The Journal of Korean Association of Computer Education
• /
• v.14 no.4
• /
• pp.43-51
• /
• 2011

Recently, by the development of information technology, we can get various information from anywhere in real time. However, personal information is exposed to threats which may incur unwanted information leakage. Cryptography serves as a primary study to prevent this leakage. However, some theories of cryptography are based on complex mathematical theories which make many people confused. Therefore, in this paper, we develope a software which is helpful to understand RSA algorithm, which is widely used algorithm in digital signature to protect personal information.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.993-996
• /
• 2009

The object of this study is to make leaners have studying environment to study adaptively, any where, any one, any time, and just in time. So, it helps leaners find solutions to questions and problems which they can face in the process of learning. This study tried to find a solution to possibility of ontologied electronic circuit, after consideration of the Semantic web and ontology theory through studying of Sundry records. As the result, I established the ontology frame about the electronic circuit, and I studied on application for e-learning based on the Semantic web ontology.