• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.1
• /
• pp.56-61
• /
• 2022

In this paper, we classify and present AI algorithms and natural language processing methods used in chatbots. A framework that can be used to implement a chatbot is also described. A chatbot is a system with a structure that interprets the input string by constructing the user interface in a conversational manner and selects an appropriate answer to the input string from the learned data and outputs it. However, training is required to generate an appropriate set of answers to a question and hardware with considerable computational power is required. Therefore, there is a limit to the practice of not only developing companies but also students learning AI development. Currently, chatbots are replacing the existing traditional tasks, and a practice course to understand and implement the system is required. RNN and Char-CNN are used to increase the accuracy of answering questions by learning unstructured data by applying technologies such as deep learning beyond the level of responding only to standardized data. In order to implement a chatbot, it is necessary to understand such a theory. In addition, the students presented examples of implementation of the entire system by utilizing the methods that can be used for coding education and the platform where existing developers and students can implement chatbots.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.1-17
• /
• 2019

Because stock price forecasting is an important issue both academically and practically, research in stock price prediction has been actively conducted. The stock price forecasting research is classified into using structured data and using unstructured data. With structured data such as historical stock price and financial statements, past studies usually used technical analysis approach and fundamental analysis. In the big data era, the amount of information has rapidly increased, and the artificial intelligence methodology that can find meaning by quantifying string information, which is an unstructured data that takes up a large amount of information, has developed rapidly. With these developments, many attempts with unstructured data are being made to predict stock prices through online news by applying text mining to stock price forecasts. The stock price prediction methodology adopted in many papers is to forecast stock prices with the news of the target companies to be forecasted. However, according to previous research, not only news of a target company affects its stock price, but news of companies that are related to the company can also affect the stock price. However, finding a highly relevant company is not easy because of the market-wide impact and random signs. Thus, existing studies have found highly relevant companies based primarily on pre-determined international industry classification standards. However, according to recent research, global industry classification standard has different homogeneity within the sectors, and it leads to a limitation that forecasting stock prices by taking them all together without considering only relevant companies can adversely affect predictive performance. To overcome the limitation, we first used random matrix theory with text mining for stock prediction. Wherever the dimension of data is large, the classical limit theorems are no longer suitable, because the statistical efficiency will be reduced. Therefore, a simple correlation analysis in the financial market does not mean the true correlation. To solve the issue, we adopt random matrix theory, which is mainly used in econophysics, to remove market-wide effects and random signals and find a true correlation between companies. With the true correlation, we perform cluster analysis to find relevant companies. Also, based on the clustering analysis, we used multiple kernel learning algorithm, which is an ensemble of support vector machine to incorporate the effects of the target firm and its relevant firms simultaneously. Each kernel was assigned to predict stock prices with features of financial news of the target firm and its relevant firms. The results of this study are as follows. The results of this paper are as follows. (1) Following the existing research flow, we confirmed that it is an effective way to forecast stock prices using news from relevant companies. (2) When looking for a relevant company, looking for it in the wrong way can lower AI prediction performance. (3) The proposed approach with random matrix theory shows better performance than previous studies if cluster analysis is performed based on the true correlation by removing market-wide effects and random signals. The contribution of this study is as follows. First, this study shows that random matrix theory, which is used mainly in economic physics, can be combined with artificial intelligence to produce good methodologies. This suggests that it is important not only to develop AI algorithms but also to adopt physics theory. This extends the existing research that presented the methodology by integrating artificial intelligence with complex system theory through transfer entropy. Second, this study stressed that finding the right companies in the stock market is an important issue. This suggests that it is not only important to study artificial intelligence algorithms, but how to theoretically adjust the input values. Third, we confirmed that firms classified as Global Industrial Classification Standard (GICS) might have low relevance and suggested it is necessary to theoretically define the relevance rather than simply finding it in the GICS.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.23 no.10
• /
• pp.869-877
• /
• 2013

Tension force of hanger ropes has been recognized and utilized as an important parameter for health monitoring of suspension bridges. Conventional vibration method based on string theory has been utilized to estimate tension forces of relatively long hanger ropes without any problem, however it is convinced that the vibration method is not applicable for shorter hanger ropes in which the influence of flexural stiffness is not ignorable. Therefore, as an alternative of vibration method, a number of feasibility studies of system identification(SI) technique considering flexural stiffness of the hanger ropes are recently performed. In this study, the influence of support condition of the finite element model utilized for the SI method is investigated with numerical examples. The numerical examples are prepared with the specification of the Kwang-Ahn bridge hanger ropes, and it is revealed that the estimation result of the tension force can be varied from -21.6 % to +35.3 % of the exact value according to the consideration of the support condition of FE model. Therefore, it is concluded that the rotational stiffness of the support spring should be included to the list of the identification parameters of the FE model to improve the result of tension estimation.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2002.11a
• /
• pp.95-100
• /
• 2002

The wing in ground effect (WIG) ship is an energy saying vessel that uses the lift from its air-wing along with the lift increase from the ground effect by flying low above the sea surface. The WIG Ship should consist of thin plate in order to float on the sea and to fly in the air. Therefore, the structure of WIG, Ship has very thin and light shell plate and stiffener like stringer and frame has comparatively large cross section area. This structure makes shell plate nearly pure shear field when shell plate is pressed by in-plane load. This complex thin plate structure of WIG Ship can he considered as a closed section beam which makes it possible to analyze structure response of WIG Ship affected by shear load and bending load. In this respect, the present study will show basic theory for analysing shear stress and focus on the analysis of structure strength of model WIC Ship's wing.

• Journal of the Korean Society of Costume
• /
• v.43
• /
• pp.187-202
• /
• 1999

The objectives of this study were to analyze the effect of dress(Korean traditional dress and suit) of elderly Women and situation on impression formation. The experimental design was $10\times{2}\times{2(dress}\times{perceiver's age}\times{situation)}$ factorial design by 3 independent variables. The stimuli of color photographs of female in her 60's model and the semantic differential scale were used. Six variables of impression formation were used: preference: elegance: potency: activity: feminine: and modernity. Samples were 400 women 200 were in their twenties and 200 in their forties and fifties. The data were analyzed by $\alpha$-reliability t-test ANOVA and duncan's multiple range test. The Korean traditional dress with the combination of Korean traditional color(light blue upper dress with dark red purple collar and string.dark blue skit) had the most positive effect on impression of elegance. Pink traditional dress and light blue traditional dress had a negative effect on impression of potency activity and modernity. Red purple suit had a positive effect on potency and modernity. The interaction between dress perceiver's age and stituation was significant for the impression of activity. Women in their 40's and 50's perceived the activity of red purple suit positively in the situation of alumnae meeting more than in the wedding ceremony. The perceived age of the stimulus person was different according to dresses. Traditional dresses was perceived older than suits were. Women in their 40's and 50's evaluated preferences of the dresses positively more than 20's did. This means that 40's and 50's feel similarity with the stimulus person more than 20's as the age of model was in their 60's The result supports the theory that similarity is basic factor in interpersonal attraction.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.2
• /
• pp.87-93
• /
• 2002

A new approach of constructing a suffix tree $T_s$for the given string S is to construct recursively a suffix tree $ T_0$ for odd positions construct a suffix tree $T_e$ for even positions from $ T_o$ and then merge $ T_o$ and $T_e$ into $T_s$ To construct suffix trees for integer alphabets in linear time had been a major open problem on index data structures. Farach used this approach and gave the first linear-time algorithm for integer alphabets The hardest part of Farachs algorithm is the merging step. In this paper we present a new and simpler merging algorithm based on a coupled BFS (breadth-first search) Our merging algorithm is more intuitive than Farachs coupled DFS (depth-first search ) merging and thus it can be easily extended to other applications.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.6 no.3
• /
• pp.14-24
• /
• 1996

A fuzzy classifier which needs various analyses of features using genetic algorithms is proposed. The fuzzy classifier has a simple structure, which contains a classification part based on fuzzy logic theory and a rule generation ation padptu sing genetic algorithms. The rule generation part determines optimal fuzzy membership functions and inclusior~ or exclusion of each feature in fuzzy classification rules. We analyzed recognition rate of a specific object, then added finer features repetitively, if necessary, to the object which has large misclassification rate. And we introduce repetitive analyses method for the minimum size of string and population, and for the improvement of recognition rates. This classifier is applied to three examples of the classification of iris data, the discrimination of thyroid gland cancer cells and the recognition of confusing handwritten and printed numerals. In the recognition of confusing handwritten and printed numerals, each sample numeral is classified into one of the groups which are divided according to the sample structure. The fuzzy classifier proposed in this paper has recognition rates of 98. 67% for iris data, 98.25% for thyroid gland cancer cells and 96.3% for confusing handwritten and printed numeral!;.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.5
• /
• pp.969-976
• /
• 2007

A fuzzy classifier which needs various analyses of features using genetic algorithms is proposed. The fuzzy classifier has a simple structure, which contains a classification part based on fuzzy logic theory and a rule generation part using genetic algorithms. The rule generation part determines optimal fuzzy membership functions and inclusion or exclusion of each feature in fuzzy classification rules. We analyzed recognition rate of a specific object, then added finer features repetitively, if necessary, to the object which has large misclassification rate. And we introduce repetitive analyses method for the minimum size of string and population, and for the improvement of recognition rates. This classifier is applied to two examples of the recognition of iris data and the recognition of Thyroid Gland cancer cells. The fuzzy classifier proposed in this paper has recognition rates of 98.67% for iris data and 98.25% for Thyroid Gland cancer cells.

• Journal of Korean Medical classics
• /
• v.5
• /
• pp.200-251
• /
• 1992

As concerned the life and the medical idea of Choo, Tan-Kye(朱丹溪), which it can be summarized as follows by studying. 1. Tan-Kye(丹溪) lived in the end of the won dynasty(元代末期), When the people starved and suffered from a flood-disaster and drought. etc, also the social conditions were in disorder on account of the corrupt ion of politics. And Cheol Kang seong(浙江省), located in the south region of China, has sterile soil and the climate condition humid and heatful. So the south district peoples have very weak constitution. So We can found that his medical idea reflected the phases of the periods and the regional enviornmental situations. 2. For that reason, Tan-Kye(丹溪) rejected the prescription of the "WHa Che Gook Bang(和劑局方)" which was prevalent at that time, in which the the pungent-dried herbs were widly used ; So he persisted in the "Sang Wha Lon(相火論)" and the "Positivity is usually excedeed while the negativity deficient(陽有餘陰不足論)". Then he treated with the drugs to nourish the negativity for the prime object to be applied in the clinic. 3. Tan-Kye(丹溪) refined the follows from the natural law; Heaven is to the positivity(陽) and the Earth is defined the negativity(陰), so the heaven is to the Macro(大) and the earth, micro(小):So the Sun is to the Positivity(陽), the Moon, the Negativity(陰): as to the Sun is always full while the moon always defected too. Therefore the "positivity is always excedeed for that the negativity is deficientalways(陽有餘陰不足)". In Human body, "the negativity energy (陰精) "is hard formed-easily defected(難成易虧)". And the heat(相火) in the body can be moved easily and let the negative energy to leak out. Therefore the more the positivity excedeed, the more the negativity deficient"(陽當有餘陰常不足). 4. He made it expanded the contents of the "Heat(相火)" in the Chapter Woon Chi of the Nae Kyeong(內徑) and discribed, the Life-string of the human body is originated from the movement of the "Heat with unique energy(相火一氣)". And more in human body, it is specifically regulated by the two visceras, Liver and Kidney, and is distributed in the 'Pericardium(心包絡)' 'Tripie Warmer(三焦)' 'Gallbladder(膽)' etc. In the point of his assertion of heat(相火), it is concluded both the physiological and the pathological heat of all. 5. Tan-Kye(丹溪) grew up in the family or the Confucianism. He was instructed the Confucianism(性理學) from Heo-Kyeom(許謙), the fourth diciple of Chu-Ja(朱子), and was received the Yoo Chang Ri(劉 張 李)'s triple doctrine from the La Tae Moo(羅太無), the second disciple of Yoo Wan So(劉完素). So there are much of content of Confucianism(性理學) in his medical thedry, and his theory has succeeded the achievements of the triple study. 6. About the theory of the "positivity is usually excedeed while the negativity deficient"(陽常有餘陰常不足論) of Tan-Kye, it was asserted that the positivity is never sufficient for the vital mainspring, by Chang, Kye-Pin(張介賓) and Lee, Kyoo-Zoon(李奎晙) etc. And for the Heat theory(相火論), eventhough the scholars of postorior generations criicized all of that, there are defect of the content and unification between them. 7. The father of the "Cha Eum Pa(滋陰派), Tan-Kye(丹溪) contributed considerably to the development of the oriental medicine and to the general therapy for the various diseases(一般雜病施治). 8. there are handed down and remained twenty or more of volumes of list of his writings. Among them, except "Kyeok Chi Yeo Ron"(格致餘論), "Kuk Pang Pal Hyeu"(局方發揮), they are reorganized by posteriority. There are Cho, Do-Chin(趙道震). Cho, Ee-Teok(趙以德), Tae, Sa-Gong(戴思恭), Wang Ri(王履) and Yoo, Suk-Yeon(劉淑淵) etc as disciples of his. And Wang Ryoon(王論) and Woo Pak(虞搏) as the admirer of him.

• Journal of Korean Society of Steel Construction
• /
• v.21 no.2
• /
• pp.165-173
• /
• 2009

Hanger cables in suspension bridges are constrained by the horizontal clamp. So, the accuracy of estimated tension of hange cable using existing methods based on the simple mathematical model of singel cable decreases as the length of cable decreases because of the flexural rigidity. Therefore, back analysis and system identification techniques based on the finite element model are proposed recently. In this paper, the applicability of the back analysis and system identification techniques are compared using the hanger cable of Gang-An Bridge. The experimental results show that the back analysis and system identification techniques are more reliable than the existing string theory and linear regression method in the view point of the error of natural frequencies. However, the estimation error of tension can be varied according to the accuracy of finite element model in the model based methods. Especially, the boundary condition is more affective when the length of cable is short, so it is important to identify the boundary condition through experiment if it is possible. The tension estimation method using system identification technique is more attractive because it can easily consider the boundary condition and it is not sensitive to the number of input measured natural frequencies.