• 생물환경조절학회지
• /
• 제20권2호
• /
• pp.63-71
• /
• 2011

영상처리는 정확한 오이의 형상 및 위치를 인식하기 위하여 형상인식 알고리즘에 대한 연구를 수행하였다. 다양한 오이형상을 인식하기 위한 방법으로는 신경회로망의 연상 메모리 알고리즘을 이용하여 오이의 특정형상을 인식하였다. 형상인식은 실제영상에서 오이의 형상과 위치를 판정할 수 있도록 알고리즘을 개발한 결과, 다음과 같은 결론을 얻었다. 본 알고리즘에서는 일정한 학습패턴의 수를 2개, 3개, 4개를 각각 기억시켜 샘플패턴 20개를 실험하여 연상시킨 결과, 학습패턴으로 복원된 출력패턴의 비율은 각각 65.0%, 45.0%, 12.5%로 나타났다. 이는 학습패턴의 수가 많을수록 수렴할 때, 다른 출력패턴으로 많이 검출되었다. 오이의 특정형상 검출은 $30{\times}30$간격으로 자동검출 되도록 처리하였다. 실제영상에서 자동 검출로 처리한 결과, 오이인식의 처리시간은 약 0.5~1초/1개(패턴) 빠르게 검출되었다. 또한, 다섯 개의 실제 영상에서 실험한 결과, 학습패턴에 대한 다른 출력패턴은 96~99%의 제거율을 나타내었다. 오이로 인식된 출력패턴 중에서, 오검출된 출력패턴의 비율은 0.1~4.2%를 나타내었다. 본 연구에서는 신경회로망을 이용하여 오이의 형상 및 위치를 인식할 수 있도록 알고리즘을 개발하였다. 오이의 위치측정은 실제영상에서 학습패턴과 유사한 출력패턴의 좌표를 가지고, 오이의 위치좌표를 추정할 수 있었다.

• 영어어문교육
• /
• 제6호
• /
• pp.35-47
• /
• 2000

This study questions the validity of Crosslinguistic Influence (CLI) in EFL Learning. A ten-minute grammaticality judgement test involving resumptive pronouns in English relative clauses was given to 15 female subjects. The research results, which were analysed in terns of language transfer and universalist arguments, support the existence of a universal process that guides L2 learning, and some common developmental patterns between the two processes of L1 and L2 learning. Hence, the universalist view should be given at least equal Weight as the CLI approach.

• 전자공학회논문지B
• /
• 제32B권3호
• /
• pp.522-532
• /
• 1995

Multilayer neural networks with backpropagation learning algorithm are widely used for pattern classification problems. For many real applications, it is more important to reduce the misclassification rate than to increase the rate of successful classification. But multilayer perceptrons(MLP's) have drawbacks of slow learning speed and false convergence to local minima. In this paper, we propose a new method for character recognition problems with a single-layer network and double rejection mechanisms, which guarantees a very low misclassification rate. Comparing to the MLP's, it yields fast learning and requires a simple hardware architecture. We also introduce a new coding scheme to reduce the misclassification rate. We have prepared two databases: one with 135,000 digit patterns and the other with 117,000 letter patterns, and have applied the proposed method for printed character recognition, which shows that the method reduces the misclassification rate significantly without sacrificing the correct recognition rate.

• The Korean Journal of Ecology
• /
• 제28권6호
• /
• pp.405-415
• /
• 2005

This paper considers the advantage of Machine Learning (ML) implemented to freshwater ecosystem research. Currently, many studies have been carried out to find the patterns of environmental impact on dynamics of communities in aquatic ecosystems. Ecological models popularly adapted by many researchers have been a means of information processing in dealing with dynamics in various ecosystems. The up-to-date trend in ecological modelling partially turns to the application of ML to explain specific ecological events in complex ecosystems and to overcome the necessity of complicated data manipulation. This paper briefly introduces ML techniques applied to freshwater ecosystems in Korea. The manuscript provides promising information for the ecologists who utilize ML for elucidating complex ecological patterns and undertaking modelling of spatial and temporal dynamics of communities.

• 한국산업융합학회 논문집
• /
• 제27권1호
• /
• pp.183-193
• /
• 2024

Recurrent Neural Network technology that learns past patterns and predicts future patterns using technology for recognizing and classifying objects is being applied to various industries, economies, and languages. And research for practical use is making a lot of progress. However, research on the application of Recurrent Neural Networks for evaluating and predicting the safety of mechanical structures is insufficient. Accurate detection of external load applied to the outside is required to evaluate the safety of mechanical structures. Learning of Recurrent Neural Networks for this requires a large amount of load data. This study applied the Gated Recurrent Unit technique to examine the possibility of load learning and investigated the possibility of applying a stacked Auto Encoder as a way to secure load data. In addition, the usefulness of learning mechanical loads was analyzed with the Gated Recurrent Unit technique, and the basic setting of related functions and parameters was proposed to secure accuracy in the recognition and prediction of loads.

• 한국BIM학회 논문집
• /
• 제6권4호
• /
• pp.35-41
• /
• 2016

Learning effect is an observation that the more times a task is performed, the less time is required to produce the same amount of outcomes. The construction industry heavily relies on repeated tasks where the learning effect is an important measure to be used. However, most construction durations are calculated and applied in real projects without considering the learning effects in each of the repeated activities. This paper applied the learning effect to the repeated activities in a small sized apartment construction project. The result showed that there was about 10 percent of difference in duration (one approach of the total duration with learning effects in 41 days while the other without learning effect in 36.5 days). To make the comparison between the two approaches, a large number of BIM based computer simulations were generated and useful patterns were recognized using machine learning algorithm named Decision Tree (See5). Machine learning is a data-driven approach for pattern recognition based on observational evidence.

• Educational Technology International
• /
• 제24권2호
• /
• pp.295-329
• /
• 2023

The recent development of metaverse-related technology has led to efforts to overcome the limitations of time and space in education by creating a virtual educational environment. To make use of this platform efficiently, applying learning analytics has been proposed as an optimal instructional and learning decision support approach to address these issues by identifying specific rules and patterns generated from learning data, and providing a systematic framework as a guideline to instructors. To achieve this, we employed an inductive, bottom-up approach for framework modeling. During the modeling process, based on the activity system model, we specifically derived the fundamental components of the learning analytics framework centered on learning activities and their contexts. We developed a prototype of the framework through deduplication, categorization, and proceduralization from the components, and refined the learning analytics framework into a 7-stage framework suitable for application in the metaverse through 3 steps of Delphi surveys. Lastly, through a framework model evaluation consisting of seven items, we validated the metaverse learning analytics framework, ensuring its validity.

• 전자공학회논문지B
• /
• 제33B권3호
• /
• pp.186-192
• /
• 1996

A modified learning rule is introduced for the implementation of feedforward artificial neural networks with on-chip learning circuitry using standard analog CMOS technology. Learning rule, is modified form the EBP (error back propagation) rule which is one of the well-known learning rules for the feedforward rtificial neural nets(FANNs). The employed MEBP ( modified EBP) rule is well - suited for the hardware implementation of FANNs with on-chip learning rule. As a ynapse circuit, a four-quadrant vector-product linear multiplier is employed, whose input/output signals are given with voltage units. Two $2{\times}2{\times}1$ FANNs are implemented with the learning circuitry. The implemented FANN circuits have been simulatied with learning test patterns using the PSPICE circuit simulator and their results show correct learning functions.

• 전자공학회논문지B
• /
• 제33B권5호
• /
• pp.102-108
• /
• 1996

Adaptive resonance theory (ART2) characterized by its built-in mechanism of handling the stability-plasticity switching and by the adaptive learning without forgetting informations learned in the past, is based on an unsupervised template matching. We propose an improved tow-stage learning algorithm for aRT2: the original unsupervised learning followed by a new supervised learning. Each of the output nodes, after the unsupervised learning, is labeled according to the category informations to reinforce the template pattern associated with the target output node belonging to the same category some dominant classes from exhausting a finite number of template patterns in ART2 inefficiently. Experimental results on a set of 2545 FLIR images show that the ART2 trained by the two-stage learning algorithm yields better accuracy than the original ART2, regardless of th esize of the network and the methods of evaluating the accuracy. This improvement shows the effectiveness of the two-stage learning process.

• 정보교육학회논문지
• /
• 제18권4호
• /
• pp.559-566
• /
• 2014

본 연구는 인지적 발달단계에 대한 신경학적 역동 발달주기를 규명하기 위하여 추상적 발달단계인 추상적 맵핑, 추상적 체계, 단일원리의 각 학습단계별 뇌파의 변화와 역동적 학습발달 간의 관계를 규명하였다. 컴퓨터 수학학습에서 일어나는 자발적 학습은 수학과제를 수행할 때 적은 학습지원 으로 나타나는 학습효과에 중점을 두었으며 이해적 학습은 적절한 학습지원을 통해 나타내는 학습효과를 중심으로 인지적 변화와 뇌파와의 관계성을 통해 뇌와 뇌신경의 발달관점에서 파악한 것이다. 연구 결과, 추상적 맵핑과 추상적 시스템 단계에서 지원을 통한 이해적 학습이 두정엽과 전두엽에서 의미 있는 뇌 활동성을 가져왔으며 추상적 개념학습의 마지막 단계인 단일원리에서는 피험자의 발달단계가 적정나이보다 작아 오히려 지원을 통한 이해적 학습이 더 적은 뇌 활동성을 가져왔다.