• MALSORI
• /
• no.56
• /
• pp.207-223
• /
• 2005

The goal of language model adaptation is to improve the background language model with a relatively small adaptation corpus. This study presents a language model adaptation technique where additional text data for the adaptation do not exist. We propose the information retrieval (IR) technique with N-gram language modeling to collect the adaptation corpus from baseline text data. We also propose to use a dynamic language model interpolation coefficient to combine the background language model and the adapted language model. The interpolation coefficient is estimated from the word hypotheses obtained by segmenting the input speech data reserved for held-out validation data. This allows the final adapted model to improve the performance of the background model consistently The proposed approach reduces the word error rate by $13.6\%$ relative to baseline 4-gram for two-hour broadcast news speech recognition.

• ETRI Journal
• /
• v.28 no.3
• /
• pp.337-346
• /
• 2006

In this paper, we propose a new language model, namely, a dependency structure language model, for information retrieval to compensate for the weaknesses of unigram and bigram language models. The dependency structure language model is based on the first-order dependency model and the dependency parse tree generated by a linguistic parser. So, long-distance dependencies can be naturally captured by the dependency structure language model. We carried out extensive experiments to verify the proposed model, where the dependency structure model gives a better performance than recently proposed language models and the Okapi BM25 method, and the dependency structure is more effective than unigram and bigram in language modeling for information retrieval.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.81-88
• /
• 2022

The major educational goal of reading part, which occupies important portion in Korean language in Korean SAT, is to evaluated whether a given text can be fully understood. Therefore given questions in the exam must be able to solely solvable by given text. In this paper we developed a datatset based on Korean SAT's reading part in order to evaluate whether a deep learning language model can classify if the given question is true or false, which is a binary classification task in NLP. In result, by applying language model solely according to the passages in the dataset, we were able to acquire better performance than 59.2% in F1 score for human performance in most of language models, that KoELECTRA scored 62.49% in our experiment. Also we proved that structural limit of language models can be eased by adjusting data preprocess.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.43-50
• /
• 2024

This paper evaluates the Korean sentiment analysis performance of large language models like GPT-3.5 and GPT-4 using a zero-shot approach facilitated by the ChatGPT API, comparing them to pre-trained Korean models such as KoBERT. Through experiments utilizing various Korean sentiment analysis datasets in fields like movies, gaming, and shopping, the efficiency of these models is validated. The results reveal that the LMKor-ELECTRA model displayed the highest performance based on F1-score, while GPT-4 particularly achieved high accuracy and F1-scores in movie and shopping datasets. This indicates that large language models can perform effectively in Korean sentiment analysis without prior training on specific datasets, suggesting their potential in zero-shot learning. However, relatively lower performance in some datasets highlights the limitations of the zero-shot based methodology. This study explores the feasibility of using large language models for Korean sentiment analysis, providing significant implications for future research in this area.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.33 no.1
• /
• pp.80-90
• /
• 2010

Performance management is very popular in business area and seems to be an exciting topic. Despite significant research efforts and myriads of performance metrics, performance management today as a rigorous approach is still in an immature state and metrics are often selected based on intuitive and heuristic approach. In a R&D sector, the difficulty to select the proper performance metrics is even more increasing due to the natural characteristics of R&D such as unique or domain-specific problems. In this paper, we present a way of presenting R&D performance framework using ontology language. Based on this, the specific metrics can be derived by reusing or inheriting the context in the framework. The proposed ontological framework is formalized using OWL(Ontology Web Language) and metrics selection rules satisfying the characteristics of R&D are represented in SWRL(Semantic Web Rule Language). Actual metrics selection procedure is carried out using JESS rule engine, a plug-in to Prot$\acute{e}$g$\acute{e}$, and illustrated with an example, incorporating a prevalent R&D performance model : TVP(Technology Value Pyramid).

• 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.

• Language and Information
• /
• v.15 no.2
• /
• pp.23-45
• /
• 2011

This paper proposes a statistical-based linguistic methodology for automatic mapping between large-scale heterogeneous languages resources for NLP applications in general. As a particular case, it treats automatic mapping between two large-scale heterogeneous Korean language resources: Sejong Semantic Classes (SJSC) in the Sejong Electronic Dictionary (SJD) and nouns in KorLex. KorLex is a large-scale Korean WordNet, but it lacks syntactic information. SJD contains refined semantic-syntactic information, with semantic labels depending on SJSC, but the list of its entry words is much smaller than that of KorLex. The goal of our study is to build a rich language resource by integrating useful information within SJD into KorLex. In this paper, we use both linguistic and statistical methods for constructing an automatic mapping methodology. The linguistic aspect of the methodology focuses on the following three linguistic clues: monosemy/polysemy of word forms, instances (example words), and semantically related words. The statistical aspect of the methodology uses the three statistical formulae ${\chi}^2$, Mutual Information and Information Gain to obtain candidate synsets. Compared with the performance of manual mapping, the automatic mapping based on our proposed statistical linguistic methods shows good performance rates in terms of correctness, specifically giving recall 0.838, precision 0.718, and F1 0.774.

• ETRI Journal
• /
• v.43 no.6
• /
• pp.1038-1048
• /
• 2021

We propose an end-to-end neural coreference resolution for the Korean language that uses an attention mechanism to point to the same entity. Because Korean is a head-final language, we focused on a method that uses a pointer network based on the head. The key idea is to consider all nouns in the document as candidates based on the head-final characteristics of the Korean language and learn distributions over the referenced entity positions for each noun. Given the recent success of applications using bidirectional encoder representation from transformer (BERT) in natural language-processing tasks, we employed BERT in the proposed model to create word representations based on contextual information. The experimental results indicated that the proposed model achieved state-of-the-art performance in Korean language coreference resolution.

• Electronics and Telecommunications Trends
• /
• v.39 no.1
• /
• pp.95-105
• /
• 2024

In natural language processing, large language models such as GPT-4 have recently been in the spotlight. The performance of natural language processing has advanced dramatically driven by an increase in the number of model parameters related to the number of acceptable input tokens and model size. Research on multimodal models that can simultaneously process natural language and image data is being actively conducted. Moreover, natural-language and image-based reasoning capabilities of large language models is being explored in robot artificial intelligence technology. We discuss research and related patent trends in robot task planning and code generation for robot control using large language models.

• Journal of computational fluids engineering
• /
• v.15 no.1
• /
• pp.88-94
• /
• 2010

In general, flow analysis programs have been built in not JAVA language but such as Fortran, C or C++ because of a merit in performance of computation. However, most developer of those languages are not free for use. User should purchase the development tools of programs, and install it on their own computer. Fortunately, JAVA is a free software, and can be used to develop WEB-based programs. It is well known that the JAVA language is time consuming for calculating. However, the performance of computer power has been increased continuously, so it is sure that the former states can be overcome. In the present study, a flow analysis program, which is interactive with Graphics, using the Web-based programing language, JAVA, has been developed for solving the two-dimensional flow motion.