• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.164-167
• /
• 2022

In table Machine comprehension, the knowledge required for language models or the structural form of tables changes depending on the domain, showing a greater performance degradation compared to text data. In this paper, we propose a pre-learning data construction method and an adversarial learning method through meaningful tabular data selection for constructing a pre-learning table language model robust to these domain changes in table machine reading. In order to detect tabular data sed for decoration of web documents without structural information from the extracted table data, a rule through heuristic was defined to identify head data and select table data was applied. An adversarial learning method between tabular data and infobax data with knowledge information about entities was applied. When the data was refined compared to when it was trained with the existing unrefined data, F1 3.45 and EM 4.14 increased in the KorQuAD table data, and F1 19.38, EM 4.22 compared to when the data was not refined in the Spec table QA data showed increased performance.

• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.71-79
• /
• 2023

In this paper, we analyze the characteristics of machine learning workloads and, based on them, propose a distributed in-memory caching technique to improve the performance of machine learning workloads. The core of machine learning workload is model training, and model training is a computationally intensive task. Performing machine learning workloads in a Kubernetes-based cloud environment in which the computing framework and storage are separated can effectively allocate resources, but delays can occur because IO must be performed through network communication. In this paper, we propose a distributed in-memory caching technique to improve the performance of machine learning workloads performed in such an environment. In particular, we propose a new method of precaching data required for machine learning workloads into the distributed in-memory cache by considering Kubflow pipelines, a Kubernetes-based machine learning pipeline management tool.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.6 no.3
• /
• pp.291-300
• /
• 2016

The emergence of smart devices and IoT leads to the appearance of O2O service to blur the difference between online and offline. As online services' merits were added to the offline market, it caused a change in the dynamics of the offline industry, which means the offline-space's digitization. Unlike these changing aspects of the offline market, exhibition industry grows steadily in the industry, however it is also possible to create a new value added by combining O2O service. We conducted a survey targeting 20 spectators in '2015 Seoul Design Festival' at COEX. The survey was used to analysis of the spatial structure and generate the dataset for machine learning. We identified problems with the analysis study of the existing spatial structure, and based on this investigation we propose a new method for analyzing a spatial structure. Also by processing a machine learning technique based on the generated dataset, we propose a novel evaluation model of exhibition-space cells for O2O service layout.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.2
• /
• pp.165-170
• /
• 2021

Machine learning has recently been applied to research in most areas. This is because the results of machine learning are not determined, but the learning of input data creates the objective function, which enables the determination of new data. In addition, the increase in accumulated data affects the accuracy of machine learning results. The data collected here is an important factor in machine learning. The proposed system is a convergence system of cloud systems and local fog systems for service delivery. Thus, the cloud system provides machine learning and infrastructure for services, while the fog system is located in the middle of the cloud and the user to collect and refine data. The data for this application shall be based on the Sensitive data generated by smart devices. The machine learning technique applied to this system uses SVM algorithm for classification and RNN algorithm for status recognition.

• Journal of Korea Multimedia Society
• /
• v.17 no.2
• /
• pp.232-239
• /
• 2014

As online texts have been rapidly growing, their automatic classification gains more interest with machine learning methods. Nevertheless, comparatively few research could be found, aiming for Korean texts. Evaluating them with statistical methods are also rare. This study took a sample of tweets and used machine learning methods to classify emotions with features of morphemes and n-grams. As a result, about 76% of emotions contained in tweets was correctly classified. Of the two methods compared in this study, Support Vector Machines were found more accurate than Na$\ddot{i}$ve Bayes. The linear model of SVM was not inferior to the non-linear one. Morphological features did not contribute to accuracy more than did the n-grams.

• The Journal of the Korea Contents Association
• /
• v.19 no.7
• /
• pp.526-537
• /
• 2019

This study researched the effect of application systems for media industry by using machine learning method focusing on industrial organization theory. First, for applying the system successfully, formation of sympathy about needs is required. The introduction of machine learning can bring change in each stage of value chain especially, decision making process of investment and production process. In investment side, objective performance prediction data can enhance efficiency, and content diversity can decrease with concentrated investment phenomenon to secured content by the system. In production side, if the system support to make creators decrease simple repeat works, production efficiency will increase.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.555-562
• /
• 2016

Semantic Role Labeling (SRL) is to determine the semantic relation of a predicate and its argu-ments in a sentence. But Korean semantic role labeling has faced on difficulty due to its different language structure compared to English, which makes it very hard to use appropriate approaches developed so far. That means that methods proposed so far could not show a satisfied perfor-mance, compared to English and Chinese. To complement these problems, we focus on suffix information analysis, such as josa (case suffix) and eomi (verbal ending) analysis. Korean lan-guage is one of the agglutinative languages, such as Japanese, which have well defined suffix structure in their words. The agglutinative languages could have free word order due to its de-veloped suffix structure. Also arguments with a single morpheme are then labeled with statistics. In addition, machine learning algorithms such as Support Vector Machine (SVM) and Condi-tional Random Fields (CRF) are used to model SRL problem on arguments that are not labeled at the suffix analysis phase. The proposed method is intended to reduce the range of argument instances to which machine learning approaches should be applied, resulting in uncertain and inaccurate role labeling. In experiments, we use 15,224 arguments and we are able to obtain approximately 83.24% f1-score, increased about 4.85% points compared to the state-of-the-art Korean SRL research.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.4
• /
• pp.147-152
• /
• 2016

One of major issues in machine learning that extracts and acquires knowledge implicit in data is to find an appropriate way of representing it. Knowledge can be represented by a number of structures such as networks, trees, lists, and rules. The differences among these exist not only in their structures but also in effectiveness of the models for their problem solving capability. In this paper, we propose partition utility as a criterion function for clustering that can lead to simplification of the model and thus avoid overfitting problem. In addition, a heuristic is proposed as a way to construct balanced hierarchical models.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.873-876
• /
• 2017

최근 정보검색의 효율성을 위해 데이터를 분석하여 해당 데이터를 가장 잘 나타내는 연관단어를 추출 및 추천하는 연구가 활발히 이루어지고 있다. 현재 관련 연구들은 출현 빈도수를 사용하는 방법이나 LDA와 같은 기계학습 기법을 활용해 데이터를 분석하여 연관단어를 생성하는 방법을 제안하고 있다. 기계학습 기법은 결과 값을 찾는데 사용되는 특징들을 전문가가 직접 설계해야 하며 좋은 결과를 내는 적절한 특징을 찾을 때까지 많은 시간이 필요하다. 또한, 파라미터들을 직접 설정해야 하므로 많은 시간과 노력을 필요로 한다는 단점을 지닌다. 이러한 기계학습 기법의 단점을 극복하기 위해 인공신경망을 다층구조로 배치하여 데이터를 분석하는 딥러닝이 최근 각광받고 있다. 본 논문에서는 기존 기계학습 기법을 사용하는 연관단어 추출연구의 한계점을 극복하기 위해 딥러닝을 활용한다. 먼저, 인공신경망 기반 단어 벡터 생성기인 Word2Vec를 사용하여 다양한 텍스트 데이터들을 학습하고 룩업 테이블을 생성한다. 그 후, 생성된 룩업 테이블을 바탕으로 인공신경망의 한 종류인 합성곱 신경망을 활용하여 사용자가 입력한 주제어와 관련된 최근 뉴스데이터를 분석한 후, 주제별 최신 연관단어를 추출하는 시스템을 제안한다. 또한 제안한 시스템을 통해 생성된 연관단어의 정확률을 측정하여 성능을 평가하였다.

• Journal of Digital Convergence
• /
• v.18 no.2
• /
• pp.261-268
• /
• 2020

Machine learning is achieved by making a cost function from a given digital signal data and optimizing the cost function. The cost function here has local minimums in the cost function depending on the amount of digital signal data and the structure of the neural network. These local minimums make a problem that prevents learning. Among the many ways of solving these methods, our proposed method is to change the learning step-size. Unlike existed methods using the learning rate (step-size) as a fixed constant, the use of multivariate function as the cost function prevent unnecessary machine learning and find the best way to the minimum value. Numerical experiments show that the results of the proposed method improve about 3%(88.8%→91.5%) performance using the proposed method rather than the existed methods.