• Journal of the Economic Geographical Society of Korea
• /
• v.25 no.1
• /
• pp.183-201
• /
• 2022

Estimating the implicit value of housing assets is a very important task for participants in the housing market. Until now, such estimations were usually carried out using multiple regression analysis based on the inherent characteristics of the estate. However, in this paper, we examine the estimation capabilities of the Artificial Neural Network(ANN) and its 'Deep Learning' faculty. To make use of the strength of the neural network model, which allows the recognition of patterns in data by modeling non-linear and complex relationships between variables, this study utilizes geographic coordinates (i.e. longitudinal/latitudinal points) as the locational factor of housing prices. Specifically, we built a dataset including structural and spatiotemporal factors based on the hedonic price model and compared the estimation performance of the models with and without geographic coordinate variables. The results show that high estimation performance can be achieved in ANN by explaining the spatial effect on housing prices through the geographic location.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.1
• /
• pp.71-82
• /
• 2024

With the advancement of information and communication technology (ICT), the use of electronic document types such as PDF, MS Office, and HWP files has increased. Such trend has led the cyber attackers increasingly try to spread malicious documents through e-mails and messengers. To counter such attacks, AI-based methodologies have been actively employed in order to detect malicious document files. The main challenge in detecting malicious HWP(Hangul Word Processor) files is the lack of quality dataset due to its usage is limited in Korea, compared to PDF and MS-Office files that are highly being utilized worldwide. To address this limitation, data augmentation have been proposed to diversify training data by transforming existing dataset, but as the usefulness of the augmented data is not evaluated, augmented data could end up harming model's performance. In this paper, we propose an effective semi-supervised learning technique in detecting malicious HWP document files, which improves overall AI model performance via quantifying the utility of augmented data and filtering out useless training data.

• Journal of the Korea Convergence Society
• /
• v.8 no.4
• /
• pp.1-8
• /
• 2017

In the modern society, the size of the fashion market is continuously increasing both overseas and domestic. When purchasing a product through e-commerce, the evaluation data for the product created by other consumers has an effect on the consumer's decision to purchase the product. By analysing the consumer's evaluation data on the product the company can reflect consumer's opinion which can leads to positive affect of performance to company. In this paper, we propose a method to construct a model to analyze user's sentiment using word embedding space formed by learning review data of amazon fashion products. Experiments were conducted by learning three SVM classifiers according to the number of positive and negative review data using the formed word embedding space which is formed by learning 5.7 million Amazon review data.. Experimental results showed the highest accuracy of 88.0% when learning SVM classifier using 50,000 positive review data and 50,000 negative review data.

• Geomechanics and Engineering
• /
• v.37 no.4
• /
• pp.307-321
• /
• 2024

This paper delves into the critical assessment of predicting sidewall displacement in underground caverns through the application of nine distinct machine learning techniques. The accurate prediction of sidewall displacement is essential for ensuring the structural safety and stability of underground caverns, which are prone to various geological challenges. The dataset utilized in this study comprises a total of 310 data points, each containing 13 relevant parameters extracted from 10 underground cavern projects located in Iran and other regions. To facilitate a comprehensive evaluation, the dataset is evenly divided into training and testing subset. The study employs a diverse array of machine learning models, including recurrent neural network, back-propagation neural network, K-nearest neighbors, normalized and ordinary radial basis function, support vector machine, weight estimation, feed-forward stepwise regression, and fuzzy inference system. These models are leveraged to develop predictive models that can accurately forecast sidewall displacement in underground caverns. The training phase involves utilizing 80% of the dataset (248 data points) to train the models, while the remaining 20% (62 data points) are used for testing and validation purposes. The findings of the study highlight the back-propagation neural network (BPNN) model as the most effective in providing accurate predictions. The BPNN model demonstrates a remarkably high correlation coefficient (R2 = 0.99) and a low error rate (RMSE = 4.27E-05), indicating its superior performance in predicting sidewall displacement in underground caverns. This research contributes valuable insights into the application of machine learning techniques for enhancing the safety and stability of underground structures.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.3
• /
• pp.1202-1211
• /
• 2018

In this paper, induction machine operation efficiency and torque is improved using Machine Learning based Gene Optimization (ML-GO) Technique is introduced. Optimized Genetic Algorithm (OGA) is used to select the optimal induction machine data. In OGA, selection, crossover and mutation process is carried out to find the optimal electrical machine data for induction machine design. Initially, many number of induction machine data are given as input for OGA. Then, fitness value is calculated for all induction machine data to find whether the criterion is satisfied or not through fitness function (i.e., objective function such as starting to full load torque ratio, rotor current, power factor and maximum flux density of stator and rotor teeth). When the criterion is not satisfied, annealed selection approach in OGA is used to move the selection criteria from exploration to exploitation to attain the optimal solution (i.e., efficient machine data). After the selection process, two point crossovers is carried out to select two crossover points within a chromosomes (i.e., design variables) and then swaps two parent's chromosomes for producing two new offspring. Finally, Adaptive Levy Mutation is used in OGA to select any value in random manner and gets mutated to obtain the optimal value. This process gets iterated till finding the optimal value for induction machine design. Experimental evaluation of ML-GO technique is carried out with performance metrics such as torque, rotor current, induction machine operation efficiency and rotor power factor compared to the state-of-the-art works.

• Journal of Korea Multimedia Society
• /
• v.7 no.3
• /
• pp.438-449
• /
• 2004

More and more users are using the query e-mail according to the increment of use of internet. The operator of internet site desires the users to check the FAQ and Q＆A contents first before sending the query e-mail to the operator However the users try to get the solution for a problem easily by simply sending a query e-mail. Therefore the increment of query e-mail is inevitable, and the site operator is suffering from too heavy loads and spending too much time and cost to reply the query e-mail. In this paper, we are proposing an efficient method of classifying the query e-mail of users automatically by using a neural network. To verify the reasonability of our work, the query e-mails of KORNET are used as the test data, which is actually gathered in KT. A total of 210 learning data and 280 test data were used to test the performance of the proposed approach. From the experiments we got the encouraging result from the view point of application in real life. The proposed approach satisfied the request of users who wanted rapid response for their query e-mail.

• International Journal of Contents
• /
• v.13 no.2
• /
• pp.66-74
• /
• 2017

Patent classification is becoming more critical as patent filings have been increasing over the years. Despite comprehensive studies in the area, there remain several issues in classifying patents on IPC hierarchical levels. Not only structural complexity but also shortage of patents in the lower level of the hierarchy causes the decline in classification performance. Therefore, we propose a new method of classification based on different criteria that are categories defined by the domain's experts mentioned in trend analysis reports, i.e. Patent Landscape Report (PLR). Several experiments were conducted with the purpose of identifying type of features and weighting methods that lead to the best classification performance using Support Vector Machine (SVM). Two types of features (noun and noun phrases) and five different weighting schemes (TF-idf, TF-rf, TF-icf, TF-icf-based, and TF-idcef-based) were experimented on.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.103.1-103
• /
• 2002

In variable structure control algorithms, The control law used to realized the desired sliding mode dynamics is discontinuous on the switching manifold. However, due to imperfections in switching, such as time delays, the system trajectory chatters instead of sliding along the switching manifold. This chattering is undesirable because it may excite unmodeled high frequency dynamics in the physical system. In this paper, to overcome this drawback a self-organizing fuzzy sliding mode control algorithm using gradient descent method is proposed. The proposed method has the characteristics which are viewed in conventional VSC, e.g. insensitivity to a class of disturbance, parameter variations and uncertainties ill the sliding mode. To demonstrate its performance, the proposed control algorithm is applied to an inverted pendulum system. The results show that both alleviation of chattering and performance are achieved.

• International Journal of Advanced Culture Technology
• /
• v.6 no.4
• /
• pp.309-316
• /
• 2018

The main purpose of this study is to understand how cognitive factors influence the way people interact with information/information systems, by conducting comprehensive and in-depth literature reviews and a theoretical synthesis of related research. Adaptive systems have been built around an individual user's characteristics, such as interests, preferences, knowledge and goals. Individual differences in the ability to use new information and communication technology have been an important issue in all fields. Performance differences in utilizing new information and communication technology are sufficiently predictable that we can begin to coordinate them. Therefore, it is necessary to understand cognitive mechanisms to explain differences between individuals as well as the levels of performance. The theoretical synthesis from this study can be applied to design intelligent (i.e., human friendly) systems in our everyday lives. Further research should explore optimization design for user, by integrating user's individual traits (such as emotion and intent) and system modules to improve the interactions of human-system in data-driven environments.

• Journal of Multimedia Information System
• /
• v.6 no.2
• /
• pp.87-90
• /
• 2019

Machine learning techniques have been applied in almost all the domains of human life to aid and enhance the problem solving capabilities of the system. The field of medical science has improved to a greater extent with the advent and application of these techniques. Efficient expert systems using various soft computing techniques like artificial neural network, Fuzzy Logic, Genetic algorithm, Hybrid system, etc. are being developed to equip medical practitioner with better and effective diagnosing capabilities. In this paper, a comparative study to evaluate the predictive performance of subtractive clustering based ANFIS hybrid system (SCANFIS) with Fuzzy C-Means (FCM) based ANFIS system (FCMANFIS) for Alzheimer disease (AD) has been taken. To evaluate the performance of these two systems, three parameters i.e. root mean square error (RMSE), prediction accuracy and precision are implemented. Experimental results demonstrated that the FCMANFIS model produce better results when compared to SCANFIS model in predictive analysis of Alzheimer disease (AD).