• The Journal of the Convergence on Culture Technology
• /
• v.7 no.4
• /
• pp.313-323
• /
• 2021

The purpose of this study is to propose rules that can identify the speaker referred by the speaker directive in the text for the realization of a machine learning-based virtual character using a literary text. Through previous studies, we found that when applying literary texts to machine learning, the machine did not properly discriminate the speaker without any specific rules for the analysis of speaker directives such as other names, nicknames, pronouns, and so on. As a way to solve this problem, this study proposes 'nine rules for finding a speaker indicated by speaker directives (including pronouns)': location, distance, pronouns, preparatory subject/preparatory object, quotations, number of speakers, non-characters directives, word compound form, dispersion of speaker names. In order to utilize characters within a literary text as virtual ones, the learning text must be presented in a machine-comprehensible way. We expect that the rules suggested in this study will reduce trial and error that may occur when using literary texts for machine learning, and enable smooth learning to produce qualitatively excellent learning results.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.3
• /
• pp.411-425
• /
• 2013

The importance and necessity of the credit loan are increasing over time. Also, it is a natural consequence that the increase of the risk for borrower increases the risk of non-performing loan. Thus, we need to predict accurately in order to prevent the loss of a credit loan company. Our final goal is to build reliable and accurate prediction model, so we proceed the following steps: At first, we can get an appropriate sample by using several resampling methods. Second, we can consider variety models and tools to fit our resampling data. Finally, in order to find the best model for our real data, various models were compared and assessed.

• Journal of KIISE:Information Networking
• /
• v.32 no.3
• /
• pp.301-309
• /
• 2005

Learning program's behavior using machine learning techniques based on system call audit data is an effective intrusion detection method. Rule teaming, neural network, statistical technique, and hidden Markov model are representative methods for intrusion detection. Among them neural networks are known for its good performance in teaming system call sequences. In order to apply it to real world problems successfully, it is important to determine their structure. However, finding appropriate structure requires very long time because there are no formal solutions for determining the structure of networks. In this paper, a novel intrusion detection technique using evolutionary neural networks is proposed. Evolutionary neural networks have the advantage that superior neural networks can be obtained in shorter time than the conventional neural networks because it leams the structure and weights of neural network simultaneously Experimental results against 1999 DARPA IDEVAL data confirm that evolutionary neural networks are effective for intrusion detection.

• Journal of KIISE
• /
• v.43 no.6
• /
• pp.692-699
• /
• 2016

C compilers for 8-bit MCUs used in washing machines and refrigerators often do not follow the C standard to improve runtime performance. Developers who are unaware of the difference between C compilers following the C standard and the C compilers for 8-bit MCU can cause bugs that do not appear in the standard C environment but appear in the embedded systems using 8-bit MCUs. It is difficult for bug detectors that assume the standard C environment to detect such bugs. In this paper, we introduce integer promotion bugs caused by the different integer promotion rules of the C compilers for 8-bit MCU from the C standard and propose 5 bug patterns where the integer promotion bugs occur. We have developed an integer promotion bug detection tool and applied it to the washing machine control software developed by the LG electronics. The integer promotion bug detection tool successfully detected 27 integer promotion bugs in the washing machine control software.

• Journal of KIISE:Software and Applications
• /
• v.36 no.3
• /
• pp.226-235
• /
• 2009

People use various information for searching POI in the navigation system such as name, category, address, phone number. Most of users use name and category to search their POT. They don't know exact name in POI DB provided by Maker. They use abbreviated or generalized name as key word for searching POI. Because of these reasons, the hit ratio has been very low. In this paper, We suggest a extra DB_construction system for raising the hit ratio. It generates allomorphes DB link to the POI name in original DB. We classified the POI names in original DB into seven types of allomorph by analyzing the gathered patterns from the POI DB which has over 650,000 entries. For auto_generating the allomorphes, we made 577 rules based on the classified types. And we generated the allomorphes manually for the entries which are difficult to make the rule and has low frequency The generated allomorphes account for 35.8% of all original DB. The hit ratio is 89% under suggested system.

• The KIPS Transactions:PartA
• /
• v.18A no.3
• /
• pp.115-122
• /
• 2011

It is very hard to obtain a general algorithm for solution of all the constraint satisfaction optimization problems. However, if the whole problem is separated into subproblems by characteristics of decision variables, we can assume that an algorithm to obtain solutions of these subproblems is easier. Under the assumption, we propose a problem classifying rule which subdivide the whole problem, and develop backtracking algorithms fit for these subproblems. One of the methods of finding a quick solution is efficiently arrange for any order of the search tree nodes. We choose the cluster head positioning problem in wireless sensor networks in which static characteristics is dominant and interference minimization problem of RFID readers that has hybrid mixture of static and dynamic characteristics. For these problems, we develop optimal variable ordering algorithms, and compare with the conventional methods. As a result of classifying the problem into subproblems, we can realize a backtracking framework for systematic search. We also have shown that developed backtracking algorithms have good performance in their quality.

• Journal of architectural history
• /
• v.7 no.4 s.17
• /
• pp.191-198
• /
• 1998

The form of Pantheon in Rome is graphically analyzed in relation to the angle of the Sun that varies through four seasons of the year. These are worked out in the Autocad drawing files for exactitude and efficiency. Some of the results suggest that the Pantheon is carefully designed to predict the equinoxes and the solstices.

• Journal of the Institute of Convergence Signal Processing
• /
• v.10 no.1
• /
• pp.93-99
• /
• 2009

The fuzzy control, neural network and genetic algorithm(GA) are algorithms to make the intelligence of system more higher. In this paper, we optimized the fuzzy controller using a genetic algorithm for desire response. Also a compensated fuzzy controller has dual rules. One control rule used to decrease the overshoot and rise time occurring in transient response region and another fuzzy control rule use to decrease the steady state error and rapildy to converge at the convergence region. GA is necessary to optimal the exchange time of the two fuzzy control rule base. Fuzzy-GA controller have a process of reproduction, crossover and mutation and we experimented by hydraulic servo motor control system We could observe that compensated Fuzzy-GA controller have good control performance compare to the fuzzy control technique have two rule base table.

• Education of Primary School Mathematics
• /
• v.14 no.3
• /
• pp.303-316
• /
• 2011

The purpose of this study was to analyze a first grader's number sense in playing the searching complement of ten game. For this purpose, the researcher conducted the game with a first-year student and analyzed her number sense based on conversations in playing games. The results obtained in this study were as follows: First, the student had a very big interested in the game and it was easily converted into additional situations. Second, she fully understood the rules of the game and made the complement of the given number. However, she was not good at making complement of twelve. Third, she made new rules like using three or four cards. She easily made the number 10 with two cards, but she didn't easily understand how to make 10 with more than three cards.

• Journal of Digital Convergence
• /
• v.11 no.11
• /
• pp.351-357
• /
• 2013

Information retrieval by query keywords have some mismatching problems to fit user's requirement for the retrieved documents due to the varieties of users. These problems are originated from the different situations and characteristics of user's requirement. Also, it has a problem that general correlation coefficients did not display the information relations. In this thesis, it is to suggest knowledge retrieval system to verify feasibility of personnel selection procedure and results supporting selection rules after construction of personnel selection ontologies and rules composed of various concept and knowledge based on the semantic web technology. In the suggested system, it is to clear disadvantages of limited information retrieval providing the suitable information to satisfy user's different situations and characteristics using Spearman's coefficients. Experimental results by this semantic-based information retrieval show 90.3% of accuracy and 71.8% of recall compared with legacy keyword information retrieval.