• Journal of KIISE
• /
• v.44 no.7
• /
• pp.742-747
• /
• 2017

In this paper, we propose a novel approach to improve the performance of the Convolutional Neural Network(CNN) word embedding model on top of word2vec with the result of performing like doc2vec in conducting a document classification task. The Word Piece Model(WPM) is empirically proven to outperform other tokenization methods such as the phrase unit, a part-of-speech tagger with substantial experimental evidence (classification rate: 79.5%). Further, we conducted an experiment to classify ten categories of news articles written in Korean by feeding words and document vectors generated by an application of WPM to the baseline and the proposed model. From the results of the experiment, we report the model we proposed showed a higher classification rate (89.88%) than its counterpart model (86.89%), achieving a 22.80% improvement. Throughout this research, it is demonstrated that applying doc2vec in the document classification task yields more effective results because doc2vec generates similar document vector representation for documents belonging to the same category.

• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.979-993
• /
• 2021

Acoustic scene classification (ASC) categorizes an audio file based on the environment in which it has been recorded. This has long been studied in the detection and classification of acoustic scenes and events (DCASE). In this study, we considered the problem that ASC faces in real-world applications that the model used should have low-complexity. We compared several models that apply light-weight techniques. First, a base CNN model was proposed using log mel-spectrogram, deltas, and delta-deltas features. Second, depthwise separable convolution, linear bottleneck inverted residual block was applied to the convolutional layer, and Quantization was applied to the models to develop a low-complexity model. The model considering low-complexity was similar or slightly inferior to the performance of the base model, but the model size was significantly reduced from 503 KB to 42.76 KB.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1407-1422
• /
• 2022

This study was conducted to classify tree species and assess the classification accuracy, using SE-Inception, a classification-based deep learning model. The input images of the dataset used Worldview-3 and GeoEye-1 images, and the size of the input images was divided into 10 × 10 m, 30 × 30 m, and 50 × 50 m to compare and evaluate the accuracy of classification of tree species. The label data was divided into five tree species (Pinus densiflora, Pinus koraiensis, Larix kaempferi, Abies holophylla Maxim. and Quercus) by visually interpreting the divided image, and then labeling was performed manually. The dataset constructed a total of 2,429 images, of which about 85% was used as learning data and about 15% as verification data. As a result of classification using the deep learning model, the overall accuracy of up to 78% was achieved when using the Worldview-3 image, the accuracy of up to 84% when using the GeoEye-1 image, and the classification accuracy was high performance. In particular, Quercus showed high accuracy of more than 85% in F₁ regardless of the input image size, but trees with similar spectral characteristics such as Pinus densiflora and Pinus koraiensis had many errors. Therefore, there may be limitations in extracting feature amount only with spectral information of satellite images, and classification accuracy may be improved by using images containing various pattern information such as vegetation index and Gray-Level Co-occurrence Matrix (GLCM).

• Journal of the Korean Geotechnical Society
• /
• v.24 no.10
• /
• pp.57-70
• /
• 2008

The advantage of piezocone penetration test is a guarantee of continuous data, which is a source of reliable interpretation of target soil layer. Many researches have been carried out f3r several decades and several classification charts have been developed to classify in-situ soil from the cone penetration test result. Since most present classification charts or methods were developed based on the data which were compiled over the world except Korea, they should be verified to be feasible for Korean soil. Furthermore, sometimes their charts provide different soil classification results according to the different input parameters. However, unfortunately, revision of those charts is quite difficult or almost impossible. In this research a new soil classification model is proposed by using fuzzy C-mean clustering and neuro-fuzzy theory based on the 5371 CPT results and soil logging results compiled from 17 local sites around Korea. Proposed neuro-fuzzy soil classification model was verified by comparing the classification results f3r new data, which were not used during learning process of neuro-fuzzy model, with real soil log. Efficiency of proposed neuro-fuzzy model was compared with other soft computing classification models and Robertson method for new data.

• Journal of Distribution Science
• /
• v.17 no.4
• /
• pp.27-32
• /
• 2019

Purpose - The business model is a method of creating corporate value, in existing "classification of business model", limitations and redundancy phenomena are applied when a new type flows in, and as consumer's purchasing decision of consumer behavior 5 steps. The classification schemes can be used for more accurate data analysis by proposing a new mapping technique in the fourth industry. Research design, data, and methodology - It was far more classified on the business model (BMS by BDS), and so on. Designing the new horizons of logistics, marketing, methodology by reclassifying these existing data to new useful data with the old methods, in order to analyze the areas where the problem has been raised for the point that the existing methods are not suitable configured. This will be applicable to the system of quaternary industry from the perspective of the buyer. Results - The mapping results of the consumer purchase decision were as follows,the 1^st stage (interest) was 23.73%, 2^nd stages (publicity) 33.90%, 3^rd stages (sales) 13.56%, 4^th stages (decision) 11.86%, 5^th stages (repurchaser) 16.95%. This verified that "the business model can be classified through "BMS by BDS". Conclusions - This structural classification is the basis of logistics marketing in the 4th industry, and proposes a innovative and effective model of constructing theory.

• Journal of Information Processing Systems
• /
• v.15 no.5
• /
• pp.1179-1191
• /
• 2019

We design an ingenious view-pooling method named learning-based multiple pooling fusion (LMPF), and apply it to multi-view convolutional neural network (MVCNN) for 3D model classification or retrieval. By this means, multi-view feature maps projected from a 3D model can be compiled as a simple and effective feature descriptor. The LMPF method fuses the max pooling method and the mean pooling method by learning a set of optimal weights. Compared with the hand-crafted approaches such as max pooling and mean pooling, the LMPF method can decrease the information loss effectively because of its "learning" ability. Experiments on ModelNet40 dataset and McGill dataset are presented and the results verify that LMPF can outperform those previous methods to a great extent.

• Communications for Statistical Applications and Methods
• /
• v.26 no.6
• /
• pp.635-644
• /
• 2019

Class probability is a fundamental target in classification that contains complete classification information. In this article, we propose a class probability estimation method when the predictor is functional. Motivated by Wang et al. (Biometrika, 95, 149-167, 2007), our estimator is obtained by training a sequence of functional weighted support vector machines (FWSVM) with different weights, which can be justified by the Fisher consistency of the hinge loss. The proposed method can be extended to multiclass classification via pairwise coupling proposed by Wu et al. (Journal of Machine Learning Research, 5, 975-1005, 2004). The use of FWSVM makes our method model-free as well as computationally efficient due to the piecewise linearity of the FWSVM solutions as functions of the weight. Numerical investigation to both synthetic and real data show the advantageous performance of the proposed method.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.301-304
• /
• 2005

For detecting an intrusion based on the anomaly of a user's activities, previous works are concentrated on statistical techniques or frequent episode mining in order to analyze an audit data. But, since they mainly analyze the average behaviour of user's activities, some anomalies can be detected inaccurately. Therefore, we propose an anomaly detection method that utilizes an associative classification for modelling intrusion detection. Finally, we proof that a prediction model built from associative classification method yields better accuracy than a prediction model built from a traditional methods by experimental results.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.178-180
• /
• 2003

EEG-based brain computer interface(BCI) provides a new communication channel between human brain and computer. EEG data is a multivariate time series so that hidden Markov model (HMM) might be a good choice for classification. However EEG is very noisy data and contains artifacts, so useful features mr expected to improve the performance of HMM. In this paper we addresses the usefulness of principal component features with Hidden Markov model (HHM). We show that some selected principal component features can suppress small noises and artifacts, hence improves classification performance. Experimental study for the classification of EEG data during imagination of a left, right up or down hand movement confirms the validity of our proposed method.

• Asia pacific journal of information systems
• /
• v.14 no.2
• /
• pp.115-130
• /
• 2004

This paper proposes how to build an expert profile database in KMS, which provides the information of expertise that each expert possesses in the organization. To manage tacit knowledge in a knowledge management system, recent researches in this field have shown that it is more applicable in many ways to provide expert search mechanisms in KMS to pinpoint experts in the organizations with searched expertise so that users can contact them for help. In this paper, we develop a framework to automate expert classification using a text categorization technique called Vector Space Model, through which an expert database composed of all the compiled profile information is built. This approach minimizes the maintenance cost of manual expert profiling while eliminating the possibility of incorrectness and obsolescence resulted from subjective manual processing. Also, we define the structure of expertise so that we can implement the expert classification framework to build an expert database in KMS. The developed prototype system, "Knowledge Portal for Researchers in Science and Technology," is introduced to show the applicability of the proposed framework.