• Journal of the Korean Society of Clothing and Textiles
• /
• v.47 no.5
• /
• pp.873-890
• /
• 2023

This study analyzed users' emotional responses to VI character design through YouTube comments. The researchers applied text-mining to analyze 116,375 comments, focusing on terms related to character design and characteristics of VI. Using the BERT model in sentiment analysis, we classified comments into extremely negative, negative, neutral, positive, or extremely positive sentiments. Next, we conducted a co-occurrence frequency analysis on comments with extremely negative and extremely positive responses to examine the semantic relationships between character design and emotional characteristic terms. We also performed a content analysis of comments about Miquela and Shudu to analyze the perception differences regarding the two character designs. The results indicate that form elements (e.g., voice, face, and skin) and behavioral elements (e.g., speaking, interviewing, and reacting) are vital in eliciting users' emotional responses. Notably, in the negative responses, users focused on the humanization aspect of voice and the authenticity aspect of behavior in speaking, interviewing, and reacting. Furthermore, we found differences in the character design elements and characteristics that users expect based on the VI's field of activity. As a result, this study suggests applications to character design to accommodate these variations.

• Asia pacific journal of information systems
• /
• v.32 no.2
• /
• pp.226-248
• /
• 2022

The existing model for sentiment analysis of product reviews learned from past data and new data was labeled based on training. But new data was never used by the existing system for making a decision. The proposed Aspect-based multi-agent Deep Reinforcement learning Sentiment Analysis (ADRSA) model learned from its very first data without the help of any training dataset and labeled a sentence with aspect category and sentiment polarity. It keeps on learning from the new data and updates its knowledge for improving its intelligence. The decision of the proposed system changed over time based on the new data. So, the accuracy of the sentiment analysis using deep reinforcement learning was improved over supervised learning and unsupervised learning methods. Hence, the sentiments of premium customers on a particular site can be explored to other customers effectively. A dynamic environment with a strong knowledge base can help the system to remember the sentences and usage State Action Reward State Action (SARSA) algorithm with Bidirectional Encoder Representations from Transformers (BERT) model improved the performance of the proposed system in terms of accuracy when compared to the state of art methods.

• Journal of Construction Engineering and Project Management
• /
• v.9 no.4
• /
• pp.1-15
• /
• 2019

Since its adoption by the transportation sector in the early 1990s, partnering has been broadly used with the traditional delivery method by many agencies with significant reported benefits. During the same era, a number of transportation agencies (DOTs) started experimenting with a wide variety of alternative project delivery methods (APDMs) aimed at improving the delivery of highway construction projects. The effect of collaborative working strategies such as partnering, together with the APDMs have become somehow interrelated posing a potential challenge on how to effectively integrate partnering as a concept in the APDMs. The salient question has been if the collaborative nature of these APDMs has affected how partnering is being used by state DOTs. Through an extensive literature review, analysis of 32 CMGC RFPs/RFQs and review of three CMGC case studies, the study found that there is limited information in state DOT documents that show procedures on the usage of partnering with CMGC projects. Majority of DOTs are relying on the inherent nature of the CMGC contract to promote healthy collaborative practices and there is the need to consider partnering during preconstruction and construction separately to cater for any personnel change over. The study also revealed that partnering may become less important at the construction phase due to overlap between partnering and CMGC practices. In support of this finding, a CMGC partnering model was developed that can be adopted by DOTs. This paper contributes to both research and practice by expanding the existing knowledge on partnering on APDMs.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.2 no.3
• /
• pp.80-86
• /
• 2000

Eddy covariance method is widely used in measuring vertical fluxes of mass and energy between the atmosphere and the biosphere. In this method, scalar concentration is measured with either open-path or closed-path sensors. For the latter, fluctuations of scalar concentration are attenuated as the sample travels through a long tube, resulting in flux loss. To quantify this tube attenuation, water vapor concentrations measured with both closed-path and open-path sensors were analyzed. Our statistical analysis showed that the power spectral density obtained from the closed-path sensor was different from that from the open-path sensor in the frequency range of > 0.5 Hz. The loss of water vapor flux due to tube attenuation was < 5% during midday. At nighttime, however, the flux loss increased significantly because of the low wind speeds and the weak turbulence sources. Theoretical calculation for the tube attenuation showed a small bias in high frequency range probably because of the interaction of sticky water vapor with a tube wall.

• The Korean Journal of Food And Nutrition
• /
• v.27 no.2
• /
• pp.156-163
• /
• 2014

In order to develop tea by using the leaves of stevia, which is a herbal plant, and to solve the disadvantages of stevia dried leaf tea, we have manufactured the steamed tea, stir-fried tea and fermented tea by changing the manufacturing processes. As a result of the sensory tests, the steamed tea, stir-fried tea and fermented tea received higher evaluations than the dried leaf tea. In terms of efficiency, it is desired that the total number of steaming and stir-frying is only once, but the fermentation is found to be the most desirable for 2 days. There are no trends for changes in the general ingredients, mineral and free amino acid contents of stevia leaf teas by different manufacturing processes. As a result for the measurement of antioxidant activities, the steamed tea and dried leaf tea did not show significant differences, but the stir-fried tea and the fermented tea show significantly low antioxidant activities as compared to the steamed tea. The contents of stevioside in both the stir-fried tea and the fermented tea were less than that in the dried leaf tea, but in the steamed tea, there was no significant difference in the content of stevioside. Base on the present observations, this study supports high potentials of steaming process in order to produce new stevia leaf tea.

• Smart Structures and Systems
• /
• v.6 no.3
• /
• pp.183-196
• /
• 2010

Monitoring and economical design of alternative energy generators such as wind turbines is becoming increasingly critical; however acquisition of the dynamic output data can be a time-consuming and costly process. In recent years, low-cost wireless sensors have emerged as an enabling technology for structural monitoring applications. In this study, wireless sensor networks are installed in three operational turbines in order to demonstrate their efficacy in this unique operational environment. The objectives of the first installation are to verify that vibrational (acceleration) data can be collected and transmitted within a turbine tower and that it is comparable to data collected using a traditional tethered system. In the second instrumentation, the wireless network includes strain gauges at the base of the structure. Also, data is collected regarding the performance of the wireless communication channels within the tower. In both turbines, collected wireless sensor data is used for off-line, output-only modal analysis of the ambiently (wind) excited turbine towers. The final installation is on a turbine with embedded braking capabilities within the nacelle to generate an "impulse-like" load at the top of the tower. This ability to apply such a load improves the modal analysis results obtained in cases where ambient excitation fails to be sufficiently broad-band or white. The improved loading allows for computation of true mode shapes, a necessary precursor to many conditional monitoring techniques.

• Journal of Information Technology Services
• /
• v.21 no.2
• /
• pp.109-126
• /
• 2022

The number and variety of products and services offered by companies have increased dramatically, providing customers with more choices to meet their needs. As a solution to this information overload problem, the provision of tailored services to individuals has become increasingly important, and the personalized recommender systems have been widely studied and used in both academia and industry. Existing recommender systems face important problems in practical applications. The most important problem is that it cannot clearly explain why it recommends these products. In recent years, some researchers have found that the explanation of recommender systems may be very useful. As a result, users are generally increasing conversion rates, satisfaction, and trust in the recommender system if it is explained why those particular items are recommended. Therefore, this study presents a methodology of providing an explanatory function of a recommender system using a review text left by a user. The basic idea is not to use all of the user's reviews, but to provide them in a summarized form using only reviews left by similar users or neighbors involved in recommending the item as an explanation when providing the recommended item to the user. To achieve this research goal, this study aims to provide a product recommendation list using user-based collaborative filtering techniques, combine reviews left by neighboring users with each product to build a model that combines text summary methods among deep learning-based natural language processing methods. Using the IMDb movie database, text reviews of all target user neighbors' movies are collected and summarized to present descriptions of recommended movies. There are several text summary methods, but this study aims to evaluate whether the review summary is well performed by training the Sequence-to-sequence+attention model, which is a representative generation summary method, and the BertSum model, which is an extraction summary model.

• Journal of Digital Convergence
• /
• v.20 no.2
• /
• pp.269-275
• /
• 2022

As artificial intelligence services targeting humans increase, social demands are increasing that artificial intelligence should also be made on an ethical basis. Following this trend, the government and businesses are preparing policies and norms related to artificial intelligence ethics. In order to establish reasonable policies and norms, the first step is to understand the public's perceptions. In this paper, social data and news comments were collected and analyzed to understand the public's perception related to artificial intelligence and ethics. Interest analysis, emotional analysis, and discourse analysis were performed and visualized on the collected datasets. As a result of the analysis, interest in "artificial intelligence ethics" and "artificial intelligence" favorability showed an inversely proportional correlation. As a result of discourse analysis, the biggest issue was "personal information leakage," and it also showed a discourse on contamination and deflection of learning data and whether computer-made artificial intelligence should be given a legal personality. This study can be used as data to grasp the public's perception when preparing artificial intelligence ethical norms and policies.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.11
• /
• pp.501-512
• /
• 2021

Recently, a breakthrough has been made in the NLP area by Transformer techniques based on encoder-decoder. However, this only can be used in mainstream languages where millions of dataset are well-equipped, such as English and Chinese, and there is a limitation that it cannot be used in non-mainstream languages where dataset are not established. In addition, there is a deflection problem that focuses on the beginning of the document in mechanical summarization. Therefore, these methods are not suitable for documents with flows such as fairy tales and novels. In this paper, we propose a hybrid summarization method that does not require a dataset and improves the deflection problem using GAN with two adaptive discriminators. We evaluate our model on the CNN/Daily Mail dataset to verify an objective validity. Also, we proved that the model has valid performance in Korean, one of the non-mainstream languages.

• The Journal of Asian Finance, Economics and Business
• /
• v.9 no.5
• /
• pp.273-283
• /
• 2022

Banking and finance is a broad term that incorporates a variety of smaller, more specialized subjects such as corporate finance, tax finance, and insurance finance. A virtual assistant that assists users in searching for information about banking and finance terms might be an extremely beneficial tool for users. In this study, we explored the process of searching for information, seeking opportunities, and developing a virtual assistant in the first stages of starting learning and understanding Vietnamese to increase effectiveness and save time, which is also an innovative business practice in Use-case Vietnam. We built the FIBA2020 dataset and proposed a pipeline that used Natural Language Processing (NLP) inclusive of Natural Language Understanding (NLU) algorithms to build chatbot applications. The open-source framework RASA is used to implement the system in our study. We aim to improve our model performance by replacing parts of RASA's default tokenizers with Vietnamese tokenizers and experimenting with various language models. The best accuracy we achieved is 86.48% and 70.04% in the ideal condition and worst condition, respectively. Finally, we put our findings into practice by creating an Android virtual assistant application using the model trained using Whitespace tokenizer and the pre-trained language m-BERT.