• Journal of Digital Contents Society
• /
• v.18 no.4
• /
• pp.641-647
• /
• 2017

In this paper, we propose a method to automatically generate a configured song with melodies composed of front/middle/last parts by using hierarchical artificial neural networks in automatic composition. In the first layer, an artificial neural network is used to learn an existing song or a random melody and outputs a song after performing rhythm post-processing. In the second layer, the melody created by the artificial neural network in the first layer is learned by three artificial neural networks of front/middle/last parts in the second layer in order to make a configured song. In the artificial neural network of the second layer, we applied a method to generate repeatability using measure identity in order to make song with repeatability and after that the song is completed after rhythm, chord, tonality post-processing. It was confirmed from experiments that our proposed method produced configured songs well.

• Earthquakes and Structures
• /
• v.4 no.5
• /
• pp.509-525
• /
• 2013

In this paper, learning capabilities of two types of Arterial Neural Networks, namely hierarchical neural networks and Generalized Regression Neural Network were used in a two-stage approach to develop a method for generating spatial varying accelerograms from acceleration response spectra and a distance parameter in which generated accelerogram is desired. Data collected from closely spaced arrays of seismographs in SMART-1 array were used to train neural networks. The generated accelerograms from the proposed method can be used for multiple support excitations analysis of structures that their supports undergo different motions during an earthquake.

• International Journal of Advanced Culture Technology
• /
• v.6 no.4
• /
• pp.262-265
• /
• 2018

Deep learning, a sub-field of machine learning changing the prospects of artificial intelligence (AI) because of its recent advancements and application in various field. Deep learning deals with algorithms inspired by the structure and function of the brain called artificial neural networks. This works reviews basic architecture and recent advancement of deep structured learning. It also describes contemporary applications of deep structured learning and its advantages over the treditional learning in artificial interlligence. This study is useful for the general readers and students who are in the early stage of deep learning studies.

• Proceedings of the Zoological Society Korea Conference
• /
• 1996.10a
• /
• pp.108.1-108
• /
• 1996

No Abstract, See Full Text

• Structural Engineering and Mechanics
• /
• v.70 no.4
• /
• pp.457-466
• /
• 2019

This study develops a damage detection method based on neural networks. The performance of the method is numerically and experimentally verified using a three-story shear building model. The framework is mainly composed of two hierarchical stages to identify damage location and extent using artificial neural network (ANN). The normalized damage signature index, that is a normalized ratio of the changes in the natural frequency and mode shape caused by the damage, is used to identify the damage location. The modal parameters extracted from the numerically developed structure for multiple damage scenarios are used to train the ANN. The positive alarm from the first stage of damage detection activates the second stage of ANN to assess the damage extent. The difference in mode shape vectors between the intact and damaged structures is used to determine the extent of the related damage. The entire procedure is verified using laboratory experiments. The damage is artificially modeled by replacing the column element with a narrow section, and a stochastic subspace identification method is used to identify the modal parameters. The results verify that the proposed method can accurately detect the damage location and extent.

• International Journal of Computer Science & Network Security
• /
• v.23 no.9
• /
• pp.120-128
• /
• 2023

In this work, we have presented ATSA, a hierarchical attention deep learning model for Arabic sentiment analysis. ATSA was proposed by addressing several challenges and limitations that arise when applying the classical models to perform opinion mining in Arabic. Arabic-specific challenges including the morphological complexity and language sparsity were addressed by modeling semantic composition at the Arabic morphological analysis after performing tokenization. ATSA proposed to perform phrase-chunks sentiment embedding to provide a broader set of features that cover syntactic, semantic, and sentiment information. We used phrase structure parser to generate syntactic parse trees that are used as a reference for ATSA. This allowed modeling semantic and sentiment composition following the natural order in which words and phrase-chunks are combined in a sentence. The proposed model was evaluated on three Arabic corpora that correspond to different genres (newswire, online comments, and tweets) and different writing styles (MSA and dialectal Arabic). Experiments showed that each of the proposed contributions in ATSA was able to achieve significant improvement. The combination of all contributions, which makes up for the complete ATSA model, was able to improve the classification accuracy by 3% and 2% on Tweets and Hotel reviews datasets, respectively, compared to the existing models.

• ETRI Journal
• /
• v.16 no.1
• /
• pp.35-43
• /
• 1994

In this paper, the robustness of the artificial neural networks to noise is demonstrated with a multilayer perceptron, and the reason of robustness is due to the statistical orthogonality among hidden nodes and its hierarchical information extraction capability. Also, the misclassification probability of a well-trained multilayer perceptron is derived without any linear approximations when the inputs are contaminated with random noises. The misclassification probability for a noisy pattern is shown to be a function of the input pattern, noise variances, the weight matrices, and the nonlinear transformations. The result is verified with a handwritten digit recognition problem, which shows better result than that using linear approximations.

• Management Science and Financial Engineering
• /
• v.19 no.1
• /
• pp.1-10
• /
• 2013

Hierarchical forecasting strategy does not always outperform direct forecasting strategy. The performance generally depends on demand features. This research guides the use of the alternative forecasting strategies according to demand features. This paper developed and evaluated various classification models such as logistic regression (LR), artificial neural networks (ANN), decision trees (DT), boosted trees (BT), and random forests (RF) for predicting the relative performance of the alternative forecasting strategies for the South Korean navy's spare parts demand which has non-normal characteristics. ANN minimized classification errors and inventory costs, whereas LR minimized the Brier scores and the sum of forecasting errors.

• ETRI Journal
• /
• v.41 no.6
• /
• pp.771-781
• /
• 2019

We propose the VS³-NET model to solve the task of question answering questions with machine-reading comprehension that searches for an appropriate answer in a given context. VS³-NET is a model that trains latent variables for each question using variational inferences based on a model of a simple recurrent unit-based sentences and self-matching networks. The types of questions vary, and the answers depend on the type of question. To perform efficient inference and learning, we introduce neural question-type models to approximate the prior and posterior distributions of the latent variables, and we use these approximated distributions to optimize a reparameterized variational lower bound. The context given in machine-reading comprehension usually comprises several sentences, leading to performance degradation caused by context length. Therefore, we model a hierarchical structure using sentence encoding, in which as the context becomes longer, the performance degrades. Experimental results show that the proposed VS³-NET model has an exact-match score of 76.8% and an F1 score of 84.5% on the SQuAD test set.

• The KIPS Transactions:PartB
• /
• v.10B no.1
• /
• pp.33-40
• /
• 2003

Data mining plays an important role in a knowledge discovery process and various algorithms of data mining can be selected for the specific purpose. Most of traditional hierachical clustering methode are suitable for processing small data sets, so they difficulties in handling large data sets because of limited resources and insufficient efficiency. In this study we propose a hybrid neural networks clustering technique, called PPC for Pre-Post Clustering that can be applied to large data sets and find unknown patterns. PPC combinds an artificial intelligence method, SOM and a statistical method, hierarchical clustering technique, and clusters data through two processes. In pre-clustering process, PPC digests large data sets using SOM. Then in post-clustering, PPC measures Similarity values according to cohesive distances which show inner features, and adjacent distances which show external distances between clusters. At last PPC clusters large data sets using the simularity values. Experiment with UCI repository data showed that PPC had better cohensive values than the other clustering techniques.