• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.3
• /
• pp.33-38
• /
• 2021

Recently, transfer learning techniques with a base convolutional neural network (CNN) model have widely gained acceptance in early detection and classification of crop diseases to increase agricultural productivity with reducing disease spread. The transfer learning techniques based classifiers generally achieve over 90% of classification accuracy for crop diseases using dataset of crop leaf images (e.g., PlantVillage dataset), but they have ability to classify only the pre-trained diseases. This paper provides with an evaluation scheme on selecting an effective base CNN model for crop disease transfer learning with regard to the accuracy of trained target crops as well as of untrained target crops. First, we present transfer learning models called CDC (crop disease classification) architecture including widely used base (pre-trained) CNN models. We evaluate each performance of seven base CNN models for four untrained crops. The results of performance evaluation show that the DenseNet201 is one of the best base CNN models.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.5
• /
• pp.1590-1609
• /
• 2021

Laughter is one of the most important nonverbal sound that human generates. It is a means for expressing his emotions. The acoustic and contextual features of this specific sound are different from those of speech and many difficulties arise during their modeling process. During this work, we propose an audio laughter generation system based on unsupervised generative models: the autoencoder (AE) and its variants. This procedure is the association of three main sub-process, (1) the analysis which consist of extracting the log magnitude spectrogram from the laughter database, (2) the generative models training, (3) the synthesis stage which incorporate the involvement of an intermediate mechanism: the vocoder. To improve the synthesis quality, we suggest two hybrid models (LSTM-VAE, GRU-VAE and CNN-VAE) that combine the representation learning capacity of variational autoencoder (VAE) with the temporal modelling ability of a long short-term memory RNN (LSTM) and the CNN ability to learn invariant features. To figure out the performance of our proposed audio laughter generation process, objective evaluation (RMSE) and a perceptual audio quality test (listening test) were conducted. According to these evaluation metrics, we can show that the GRU-VAE outperforms the other VAE models.

• Korean Journal of Radiology
• /
• v.22 no.10
• /
• pp.1697-1707
• /
• 2021

The recent introduction of various high-dimensional modeling methods, such as radiomics and deep learning, has created a much greater diversity in modeling approaches for survival prediction (or, more generally, time-to-event prediction). The newness of the recent modeling approaches and unfamiliarity with the model outputs may confuse some researchers and practitioners about the evaluation of the performance of such models. Methodological literacy to critically appraise the performance evaluation of the models and, ideally, the ability to conduct such an evaluation would be needed for those who want to develop models or apply them in practice. This article intends to provide intuitive, conceptual, and practical explanations of the statistical methods for evaluating the performance of survival prediction models with minimal usage of mathematical descriptions. It covers from conventional to deep learning methods, and emphasis has been placed on recent modeling approaches. This review article includes straightforward explanations of C indices (Harrell's C index, etc.), time-dependent receiver operating characteristic curve analysis, calibration plot, other methods for evaluating the calibration performance, and Brier score.

• Journal of Fisheries and Marine Sciences Education
• /
• v.26 no.5
• /
• pp.1108-1118
• /
• 2014

The purpose of this study was to spread the education of human creativity program and its possibility of consistency, and its application to school system by exploring the evaluation system that determines the valuable and useful decisions. This study was based on the evaluation system according to CIPP model regarding the program's general ideas and its role. The interest of the education of human creativity was increasing as well as its progress of program, and we searched 13 figures and 55 evaluation index according to increase the program's effectiveness and nation-wise situation of evaluation model. The conclusion of this study was that the human creativity program's evaluation was composed of six compartments. And the context evaluation was composed of three evaluation models and eleven index, and input evaluation was composed of three evaluation model and 15 index, while process evaluation is composed of three evaluation models and thirteen evaluation index. Lastly, output evaluation had four evaluation model and sixteen index.

• Korean Management Science Review
• /
• v.23 no.2
• /
• pp.93-107
• /
• 2006

This paper attempts to improve the typical DEA model that is often used as a performance evaluation method for service firms. The after-sales service system in a Korean electronic company is chosen as a sample case. The performance evaluation method currently practiced in the firm was reviewed and some critical issues of the current method were discussed. DEA model would be considered a solution for solving those Issues. However, typical DEA models has various drawbacks in determining the weights over evaluation criteria. Proposed is the priority-constrained DEA model that accommodates the strategic Importance of each evaluation criterion. The results of the proposed model are compared with those of typical DEA models. It is shown that the priority-constrained DEA model is superior to the others in terms of evaluation quality.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1995.04a
• /
• pp.797-806
• /
• 1995

An ergonomic evaluation model which can properly evaluate the design of a product or workplaces was developed to evaluate a new product in the early stage of design. Although several evaluation models have been developed and are in use in the field of ergonomics, they lack an integrated framework and analysis techniques. In this research, an ergonomic evaluation model was developed according to the framework which accommodates desirable components of the model. The framework represents the integration of four components: man model, workplace modeller, ergonomic evaluation functions and knowledge bases, which interact with a common object-oriented database. Components were developed individually and integrated with the workplace modeller. The generation and manipulation of man models, workplace modelling, and ergonomic evaluation features such as reach and visibility were implemented. The developed evaluation model has demonstrated distinct advantages of the flexibility to adapt another features for the ergonomic evaluation model and the better usability for users.

• Journal of Information Technology Services
• /
• v.8 no.4
• /
• pp.141-150
• /
• 2009

There are no objective standards for selection and improvement of SLA evaluation metrics for IT service. In this study, we analyze the current IT maturity models for selection and improvement of the metrics and then we derive them according to the maturity levels and propose the redesigned maturity model. To verify whether the model is applicable, we execute a case study based on the D company. We apply the proposed evaluation metrics of the maturity models to the D company and evaluate the metrics. We select a proper level of the D company and an improvement line after measuring evaluation metrics in the maturity level 2. We propose improvement guidelines of evaluation metrics which score is less than the improvement line's and derive SLA evaluation metrics. By using the SLA evaluation metrics for a year, we prove that the way of selection and improvement is useful.

• Physical Therapy Korea
• /
• v.8 no.1
• /
• pp.97-106
• /
• 2001

Physical therapists should under stand motor control models and apply various models to evaluation and treatment of neurologically impaired patients. Thus, this paper reviews motor control models and applications in clinical settings. Assumptions and limitations of reflex models, hierarchical models, and systems models are presented. This paper also delineates goals and dissatisfaction of neurologic rehabilitation approaches for neurologically impaired patients. Muscle reeducation approach, neurotherapeutic facilitation approach, and contemporary task-oriented approach are explained.

• Advances in concrete construction
• /
• v.10 no.6
• /
• pp.463-478
• /
• 2020

Multiple mathematical modeling for prediction of slump, compressive strength and depth of water penetration at 28 days were performed using statistical analysis for the concrete containing waste limestone powder as partial replacement of sand obtained from experimental program reported in this research. To extract experimental data, 180 concrete cubic samples with 20 different mix designs were investigated. The twenty non-linear regression models were used to predict each of the concrete properties including slump, compressive strength and water depth penetration of concrete with waste limestone powder. Evaluation of the models using numerical methods showed that the majority of models give acceptable prediction with a high accuracy and trivial error rates. The 15-term regression models for predicting the slump, compressive strength and water depth were found to have the best agreement with the tested concrete specimens.

• Phonetics and Speech Sciences
• /
• v.5 no.4
• /
• pp.109-119
• /
• 2013

The purpose of this paper is to show the tendency of evaluators in the pronunciation evaluation of English utterances. The tendency was visualized using the evaluation model of English pronunciation proposed in [1]. One hundred fifty female university students and four evaluators participated in the study. Students read eight English sentences aloud as evaluators evaluated English pronunciation by their own criteria. The models based on their pronunciation evaluation proved to be efficient in showing their evaluation tendency in terms of the fundamental frequency, intensity, segmental durations, and segmental spectra as compared to those of the five native speakers of English chosen for building the models. However, human evaluators were not always consistent in their evaluation and sometimes gave conflicting scores to the same students.