• Korean Journal of Heritage: History & Science
• /
• v.57 no.3
• /
• pp.24-35
• /
• 2024

Recently, research using deep learning technologies such as artificial intelligence, convolutional neural networks, etc. has been actively conducted in various fields including healthcare, manufacturing, autonomous driving, and security, and is having a significant influence on society. In line with this trend, the present study attempted to apply deep learning to the classification of archaeological artifacts, specifically ancient Korean roof-end tiles. Using 100 images of roof-end tiles from each of the Goguryeo, Baekje, and Silla dynasties, for a total of 300 base images, a dataset was formed and expanded to 1,200 images using data augmentation techniques. After building a model using transfer learning from the pre-trained EfficientNetB0 model and conducting five-fold cross-validation, an average training accuracy of 98.06% and validation accuracy of 97.08% were achieved. Furthermore, when model performance was evaluated with a test dataset of 240 images, it could classify the roof-end tile images from the three dynasties with a minimum accuracy of 91%. In particular, with a learning rate of 0.0001, the model exhibited the highest performance, with accuracy of 92.92%, precision of 92.96%, recall of 92.92%, and F1 score of 92.93%. This optimal result was obtained by preventing overfitting and underfitting issues using various learning rate settings and finding the optimal hyperparameters. The study's findings confirm the potential for applying deep learning technologies to the classification of Korean archaeological materials, which is significant. Additionally, it was confirmed that the existing ImageNet dataset and parameters could be positively applied to the analysis of archaeological data. This approach could lead to the creation of various models for future archaeological database accumulation, the use of artifacts in museums, and classification and organization of artifacts.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.221-238
• /
• 2019

Recently, companies and public institutions have been actively introducing chatbot services in the field of customer counseling and response. The introduction of the chatbot service not only brings labor cost savings to companies and organizations, but also enables rapid communication with customers. Advances in data analytics and artificial intelligence are driving the growth of these chatbot services. The current chatbot can understand users' questions and offer the most appropriate answers to questions through machine learning and deep learning. The advancement of chatbot core technologies such as NLP, NLU, and NLG has made it possible to understand words, understand paragraphs, understand meanings, and understand emotions. For this reason, the value of chatbots continues to rise. However, technology-oriented chatbots can be inconsistent with what users want inherently, so chatbots need to be addressed in the area of the user experience, not just in the area of technology. The Fourth Industrial Revolution represents the importance of the User Experience as well as the advancement of artificial intelligence, big data, cloud, and IoT technologies. The development of IT technology and the importance of user experience have provided people with a variety of environments and changed lifestyles. This means that experiences in interactions with people, services(products) and the environment become very important. Therefore, it is time to develop a user needs-based services(products) that can provide new experiences and values to people. This study proposes a chatbot development process based on user needs by applying the design thinking approach, a representative methodology in the field of user experience, to chatbot development. The process proposed in this study consists of four steps. The first step is 'setting up knowledge domain' to set up the chatbot's expertise. Accumulating the information corresponding to the configured domain and deriving the insight is the second step, 'Knowledge accumulation and Insight identification'. The third step is 'Opportunity Development and Prototyping'. It is going to start full-scale development at this stage. Finally, the 'User Feedback' step is to receive feedback from users on the developed prototype. This creates a "user needs-based service (product)" that meets the process's objectives. Beginning with the fact gathering through user observation, Perform the process of abstraction to derive insights and explore opportunities. Next, it is expected to develop a chatbot that meets the user's needs through the process of materializing to structure the desired information and providing the function that fits the user's mental model. In this study, we present the actual construction examples for the domestic cosmetics market to confirm the effectiveness of the proposed process. The reason why it chose the domestic cosmetics market as its case is because it shows strong characteristics of users' experiences, so it can quickly understand responses from users. This study has a theoretical implication in that it proposed a new chatbot development process by incorporating the design thinking methodology into the chatbot development process. This research is different from the existing chatbot development research in that it focuses on user experience, not technology. It also has practical implications in that companies or institutions propose realistic methods that can be applied immediately. In particular, the process proposed in this study can be accessed and utilized by anyone, since 'user needs-based chatbots' can be developed even if they are not experts. This study suggests that further studies are needed because only one field of study was conducted. In addition to the cosmetics market, additional research should be conducted in various fields in which the user experience appears, such as the smart phone and the automotive market. Through this, it will be able to be reborn as a general process necessary for 'development of chatbots centered on user experience, not technology centered'.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.177-190
• /
• 2021

Due to the development of the fourth industrial revolution technology, efforts are being made to improve areas that humans cannot handle by utilizing artificial intelligence techniques such as machine learning. Although on-demand production companies also want to reduce corporate risks such as delays in delivery by predicting total production time for orders, they are having difficulty predicting this because the total production time is all different for each order. The Theory of Constraints (TOC) theory was developed to find the least efficient areas to increase order throughput and reduce order total cost, but failed to provide a forecast of total production time. Order production varies from order to order due to various customer needs, so the total production time of individual orders can be measured postmortem, but it is difficult to predict in advance. The total measured production time of existing orders is also different, which has limitations that cannot be used as standard time. As a result, experienced managers rely on persimmons rather than on the use of the system, while inexperienced managers use simple management indicators (e.g., 60 days total production time for raw materials, 90 days total production time for steel plates, etc.). Too fast work instructions based on imperfections or indicators cause congestion, which leads to productivity degradation, and too late leads to increased production costs or failure to meet delivery dates due to emergency processing. Failure to meet the deadline will result in compensation for delayed compensation or adversely affect business and collection sectors. In this study, to address these problems, an entity that operates an order production system seeks to find a machine learning model that estimates the total production time of new orders. It uses orders, production, and process performance for materials used for machine learning. We compared and analyzed OLS, GLM Gamma, Extra Trees, and Random Forest algorithms as the best algorithms for estimating total production time and present the results.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.131-154
• /
• 2022

Recently, researches on unstructured data analysis have been actively conducted with the development of information and communication technology. In particular, topic modeling is a representative technique for discovering core topics from massive text data. In the early stages of topic modeling, most studies focused only on topic discovery. As the topic modeling field matured, studies on the change of the topic according to the change of time began to be carried out. Accordingly, interest in dynamic topic modeling that handle changes in keywords constituting the topic is also increasing. Dynamic topic modeling identifies major topics from the data of the initial period and manages the change and flow of topics in a way that utilizes topic information of the previous period to derive further topics in subsequent periods. However, it is very difficult to understand and interpret the results of dynamic topic modeling. The results of traditional dynamic topic modeling simply reveal changes in keywords and their rankings. However, this information is insufficient to represent how the meaning of the topic has changed. Therefore, in this study, we propose a method to visualize topics by period by reflecting the meaning of keywords in each topic. In addition, we propose a method that can intuitively interpret changes in topics and relationships between or among topics. The detailed method of visualizing topics by period is as follows. In the first step, dynamic topic modeling is implemented to derive the top keywords of each period and their weight from text data. In the second step, we derive vectors of top keywords of each topic from the pre-trained word embedding model. Then, we perform dimension reduction for the extracted vectors. Then, we formulate a semantic vector of each topic by calculating weight sum of keywords in each vector using topic weight of each keyword. In the third step, we visualize the semantic vector of each topic using matplotlib, and analyze the relationship between or among the topics based on the visualized result. The change of topic can be interpreted in the following manners. From the result of dynamic topic modeling, we identify rising top 5 keywords and descending top 5 keywords for each period to show the change of the topic. Existing many topic visualization studies usually visualize keywords of each topic, but our approach proposed in this study differs from previous studies in that it attempts to visualize each topic itself. To evaluate the practical applicability of the proposed methodology, we performed an experiment on 1,847 abstracts of artificial intelligence-related papers. The experiment was performed by dividing abstracts of artificial intelligence-related papers into three periods (2016-2017, 2018-2019, 2020-2021). We selected seven topics based on the consistency score, and utilized the pre-trained word embedding model of Word2vec trained with 'Wikipedia', an Internet encyclopedia. Based on the proposed methodology, we generated a semantic vector for each topic. Through this, by reflecting the meaning of keywords, we visualized and interpreted the themes by period. Through these experiments, we confirmed that the rising and descending of the topic weight of a keyword can be usefully used to interpret the semantic change of the corresponding topic and to grasp the relationship among topics. In this study, to overcome the limitations of dynamic topic modeling results, we used word embedding and dimension reduction techniques to visualize topics by era. The results of this study are meaningful in that they broadened the scope of topic understanding through the visualization of dynamic topic modeling results. In addition, the academic contribution can be acknowledged in that it laid the foundation for follow-up studies using various word embeddings and dimensionality reduction techniques to improve the performance of the proposed methodology.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1463-1478
• /
• 2022

Maritime transport accounts for 99.7% of the exports and imports of the Republic of Korea; therefore, developing a vessel monitoring system for efficient operation is of significant interest. Several studies have focused on tracking and monitoring vessel movements based on automatic identification system (AIS) data; however, ships without AIS have limited monitoring and tracking ability. High-resolution optical satellite images can provide the missing layer of information in AIS-based monitoring systems because they can identify non-AIS vessels and small ships over a wide range. Therefore, it is necessary to investigate vessel monitoring and small vessel classification systems using high-resolution optical satellite images. This study examined the possibility of developing ship monitoring systems using Compact Advanced Satellite 500-1 (CAS500-1) satellite images by first training a deep learning model using satellite image data and then performing detection in other images. To determine the effectiveness of the proposed method, the learning data was acquired from ships in the Yellow Sea and its major ports, and the detection model was established using the You Only Look Once (YOLO) algorithm. The ship detection performance was evaluated for a domestic and an international port. The results obtained using the detection model in ships in the anchorage and berth areas were compared with the ship classification information obtained using AIS, and an accuracy of 85.5% and 70% was achieved using domestic and international classification models, respectively. The results indicate that high-resolution satellite images can be used in mooring ships for vessel monitoring. The developed approach can potentially be used in vessel tracking and monitoring systems at major ports around the world if the accuracy of the detection model is improved through continuous learning data construction.

• Journal of the Korean Institute of Gas
• /
• v.21 no.4
• /
• pp.47-61
• /
• 2017

The preconditions of the usefulness of an operator safety training model in large plants are the versatility and accuracy of operational procedures, obtained by detailed analysis of the various types of risks associated with the operation, and the systematic representation of knowledge. In this study, we consider the artificial intelligence planning method for the generation of operation procedures; classify them into general actions, actions and technical terms of the operator; and take into account the sharing and reuse of knowledge, defining a knowledge expression ontology. In order to expand and extend the general operations of the operation, we apply a Hierarchical Task Network (HTN). Actual boiler plant case studies are classified according to operating conditions, states and operating objectives between the units, and general emergency shutdown procedures are created to confirm the applicability of the proposed method. These results based on systematic knowledge representation can be easily applied to general plant operation procedures and operator safety training scenarios and will be used for automatic generation of safety training scenarios.

• Korean Journal of Remote Sensing
• /
• v.36 no.3
• /
• pp.487-501
• /
• 2020

As a viable option for retrieval of LST (Land Surface Temperature), this paper presents a DNN (Deep Neural Network) based approach using 148 Landsat 8 images for South Korea. Because the brightness temperature and emissivity for the band 10 (approx. 11-㎛ wavelength) of Landsat 8 are derived by combining physics-based equations and empirical coefficients, they include uncertainties according to regional conditions such as meteorology, climate, topography, and vegetation. To overcome this, we used several land surface variables such as NDVI (Normalized Difference Vegetation Index), land cover types, topographic factors (elevation, slope, aspect, and ruggedness) as well as the T₀ calculated from the brightness temperature and emissivity. We optimized four seasonal DNN models using the input variables and in-situ observations from ASOS (Automated Synoptic Observing System) to retrieve the LST, which is an advanced approach when compared with the existing method of the bias correction using a linear equation. The validation statistics from the 1,728 matchups during 2013-2019 showed a good performance of the CC=0.910~0.917 and RMSE=3.245~3.365℃, especially for spring and fall. Also, our DNN models produced a stable LST for all types of land cover. A future work using big data from Landsat 5/7/8 with additional land surface variables will be necessary for a more reliable retrieval of LST for high-resolution satellite images.

• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.351-360
• /
• 2018

Human emotion recognition is a research topic that is receiving continuous attention in computer vision and artificial intelligence domains. This paper proposes a method for classifying human emotions through multiple neural networks based on multi-modal signals which consist of image, landmark, and audio in a wild environment. The proposed method has the following features. First, the learning performance of the image-based network is greatly improved by employing both multi-task learning and semi-supervised learning using the spatio-temporal characteristic of videos. Second, a model for converting 1-dimensional (1D) landmark information of face into two-dimensional (2D) images, is newly proposed, and a CNN-LSTM network based on the model is proposed for better emotion recognition. Third, based on an observation that audio signals are often very effective for specific emotions, we propose an audio deep learning mechanism robust to the specific emotions. Finally, so-called emotion adaptive fusion is applied to enable synergy of multiple networks. The proposed network improves emotion classification performance by appropriately integrating existing supervised learning and semi-supervised learning networks. In the fifth attempt on the given test set in the EmotiW2017 challenge, the proposed method achieved a classification accuracy of 57.12%.

• Journal of The Korean Association For Science Education
• /
• v.38 no.2
• /
• pp.219-234
• /
• 2018

In this study, we explored the possibility of automating the process of analyzing elements of scientific argument in the context of a Korean classroom. To gather training data, we collected 990 sentences from science education journals that illustrate the results of coding elements of argumentation according to Toulmin's argumentation structure framework. We extracted 483 sentences as a test data set from the transcription of students' discourse in scientific argumentation activities. The words and morphemes of each argument were analyzed using the Python 'KoNLPy' package and the 'Kkma' module for Korean Natural Language Processing. After constructing the 'argument-morpheme:class' matrix for 1,473 sentences, five machine learning techniques were applied to generate predictive models relating each sentences to the element of argument with which it corresponded. The accuracy of the predictive models was investigated by comparing them with the results of pre-coding by researchers and confirming the degree of agreement. The predictive model generated by the k-nearest neighbor algorithm (KNN) demonstrated the highest degree of agreement [54.04% (${\kappa}=0.22$)] when machine learning was performed with the consideration of morpheme of each sentence. The predictive model generated by the KNN exhibited higher agreement [55.07% (${\kappa}=0.24$)] when the coding results of the previous sentence were added to the prediction process. In addition, the results indicated importance of considering context of discourse by reflecting the codes of previous sentences to the analysis. The results have significance in that, it showed the possibility of automating the analysis of students' argumentation activities in Korean language by applying machine learning.

• Journal of the Korean Geotechnical Society
• /
• v.20 no.7
• /
• pp.69-78
• /
• 2004

In this study, the back analysis program was developed by applying the genetic algorithm, one of artificial intelligence fields, to the direct method. The optimization process which has influence on the efficiency of the direct method was modulated with genetic algorithm. On conditions that the displacement computed by forward analysis for a certain rock mass model was the same as the displacement measured at the tunnel section, back analysis was executed to verify the validity of the program. Usefulness of the program was confirmed by comparing relative errors calculated by back analysis, which is carried out under the same rock mass conditions as analysis model of Gens et at (1987), one of back analysis case in the past. We estimated the total displacement occurring by tunnelling with the crown settlement and convergence measured at the working faces in three tunnel sites of Kyungbu Express railway. Those data measured at the working face are used for back analysis as the input data after confidence test. As the results of the back analysis, we comprehended the tendency of tunnel behaviors with comparing the respective deformation characteristics obtained by the measurement at the working face and by back analysis. Also the usefulness and applicability of the back analysis program developed in this study were verified.