• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.4
• /
• pp.173-178
• /
• 2023

As Korean literature spreads around the world, its position in the overseas publishing market has become important. As demand in the overseas publishing market continues to grow, it is essential to predict future book sales and analyze the characteristics of books that have been highly favored by overseas readers in the past. In this study, we proposed ensemble learning based prediction model and analyzed characteristics of the cumulative sales of more than 5,000 copies classified as good sellers published overseas over the past 5 years. We applied the five ensemble learning models, i.e., XGBoost, Gradient Boosting, Adaboost, LightGBM, and Random Forest, and compared them with other machine learning algorithms, i.e., Support Vector Machine, Logistic Regression, and Deep Learning. Our experimental results showed that the ensemble algorithm outperforms other approaches in troubleshooting imbalanced data. In particular, the LightGBM model obtained an AUC value of 99.86% which is the best prediction performance. Among the features used for prediction, the most important feature is the author's number of overseas publications, and the second important feature is publication in countries with the largest publication market size. The number of evaluation participants is also an important feature. In addition, text mining was performed on the four book reviews that sold the most among good-selling books. Many reviews were interested in stories, characters, and writers and it seems that support for translation is needed as many of the keywords of "translation" appear in low-rated reviews.

• The Journal of Bigdata
• /
• v.8 no.2
• /
• pp.149-163
• /
• 2023

Due to the continuous growth of the E-commerce market, the volume of orders that fulfillment centers have to process has increased, and various customer requirements have increased the complexity of order processing. Along with this trend, the operational efficiency of fulfillment centers due to increased labor costs is becoming more important from a corporate management perspective. Using historical performance data as training data, this study focused on real-time box recommendations applicable to packaging areas during fulfillment center shipping. Four types of data, such as product information, order information, packaging information, and delivery information, were applied to the machine learning model through pre-processing and feature-engineering processes. As an input vector, three characteristics were used as product specification information: width, length, and height, the characteristics of the input vector were extracted through a feature engineering process that converts product information from real numbers to an integer system for each section. As a result of comparing the performance of each model, it was confirmed that when the Gradient Boosting model was applied, the prediction was performed with the highest accuracy at 95.2% when the product specification information was converted into integers in 21 sections. This study proposes a machine learning model as a way to reduce the increase in costs and inefficiency of box packaging time caused by incorrect box selection in the fulfillment center, and also proposes a feature engineering method to effectively extract the characteristics of product specification information.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.1
• /
• pp.142-149
• /
• 2016

This paper proposes a method to detect LP(License Plate) of vehicles in indoor and outdoor parking lots. In restricted environment, there are many conventional methods for detecting LP. But, it is difficult to detect LP in natural and complex scenes with background clutters because several patterns similar with text or LP always exist in complicated backgrounds. To verify the performance of LP text detection in natural images, we apply MB-LGP feature by combining with ensemble machine learning algorithm in purpose of selecting optimal features of small number in huge pool. The feature selection is performed by adaptive boosting algorithm that shows great performance in minimum false positive detection ratio and in computing time when combined with cascade approach. MSER is used to provide initial text regions of vehicle LP. Throughout the experiment using real images, the proposed method functions robustly extracting LP in natural scene as well as the controlled environment.

• Proceedings of the KOSOMBE Conference
• /
• v.1997 no.11
• /
• pp.541-545
• /
• 1997

This paper presents a modified geometric active contour model or edge detection and segmentation of computed tomography(CT) scan images. The method is based on the level setup approach developed by Osher and Sethian and the modeling of propagation fronts with curvature dependent speeds by Malladi. Based on above algorithms, the geometric active contour is obtained through a particular level set of hypersurface lowing along its gradient force and curvature force. This technique retains the attractive feature which is topological and geometric flexibility of the contour in recovering objects with complex shapes and unknown topologies. But there are limitations in this algorithm which are being not able to separate the object with weak difference from neighbor object. So we use speed limitation filter to overcome those problems. We apply a 2D model to various synthetic cases and the three cases of real CT scan images in order to segment objects with complicated shapes and topologies. From the results, the presented model confirms that it attracts very naturally and efficiently to the desired feature of CT scan images.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.10a
• /
• pp.683-686
• /
• 2002

It is fundamental to count and pack the pills in the medicine manufacture field but those tasks are time and labor consuming. Thus, the need fur automation of those tasks is necessarily getting increased in order to get effective mass production. It Is significant to perceive pills quickly and precisely. There were many trials for this processing but the performance of the existing counting machines varies about size, shape and dispersion tendency of pills. In this paper, the authors try to improve the counting performance of a pill counting machine that has optical sensors with the neural network. The passing signal of pill is acquired with optical sensor and the passage signal of the pill is extracted as input patterns. The gradient and integration of signal during passing time and the time keeping the pill interrupt the light from the LED are used as characteristic feature. The back propagation and perception algorithm are used for training. Experimental results with several pills show that the designed algorithm is a little bit effective to reduce the noise effect which is generated from interference among the machine components and unreliable environment.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.1
• /
• pp.147-165
• /
• 2021

In this work, we empirically evaluated the efficiency of the recent EfficientNetB0 model to identify and diagnose malaria parasite infections in blood smears. The dataset used was collected and classified by relevant experts from the Lister Hill National Centre for Biomedical Communications (LHNCBC). We prepared our samples with minimal image transformations as opposed to others, as we focused more on the feature extraction capability of the EfficientNetB0 baseline model. We applied transfer learning to increase the initial feature sets and reduced the training time to train our model. We then fine-tuned it to work with our proposed layers and re-trained the entire model to learn from our prepared dataset. The highest overall accuracy attained from our evaluated results was 94.70% from fifty epochs and followed by 94.68% within just ten. Additional visualization and analysis using the Gradient-weighted Class Activation Mapping (Grad-CAM) algorithm visualized how effectively our fine-tuned EfficientNetB0 detected infections better than other recent state-of-the-art DCNN models. This study, therefore, concludes that when fine-tuned, the recent EfficientNetB0 will generate highly accurate deep learning solutions for the identification of malaria parasites in blood smears without the need for stringent pre-processing, optimization, or data augmentation of images.

• Journal of Biomedical Engineering Research
• /
• v.43 no.5
• /
• pp.353-360
• /
• 2022

Laryngeal disease harms quality of life, and laryngoscopy is critical in identifying causative lesions. This study extracts and analyzes using radiomics quantitative features from the lesion in laryngoscopy images and will fit and validate a classifier for finding meaningful features. Searching the region of interest for lesions not classified by the YOLOv5 model, features are extracted with radionics. Selected the extracted features are through a combination of three feature selectors, and three estimator models. Through the selected features, trained and verified two classification models, Random Forest and Gradient Boosting, and found meaningful features. The combination of SFS, LASSO, and RF shows the highest performance with an accuracy of 0.90 and AUROC 0.96. Model using features to select by SFM, or RIDGE was low lower performance than other things. Classification of larynx lesions through radiomics looks effective. But it should use various feature selection methods and minimize data loss as losing color data.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.3
• /
• pp.19-29
• /
• 2011

Until now, many face recognition methods have been proposed, most of them use a 1-dimensional feature vector which is vectorized the input image without feature extraction process or input image itself is used as a feature matrix. It is known that the face recognition methods using raw image yield deteriorated performance in databases whose have severe illumination changes. In this paper, we propose a face recognition method using local statistics of gradients and correlations which are good for illumination changes. BDIP (block difference of inverse probabilities) is chosen as a local statistics of gradients and two types of BVLC (block variation of local correlation coefficients) is chosen as local statistics of correlations. When a input image enters the system, it extracts the BDIP, BVLC1 and BVLC2 feature images, fuses them, obtaining feature matrix by $(2D)^2$ PCA transformation, and classifies it with training feature matrix by nearest classifier. From experiment results of four face databases, FERET, Weizmann, Yale B, Yale, we can see that the proposed method is more reliable than other six methods in lighting and facial expression.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.3
• /
• pp.433-444
• /
• 2020

This study proposes a method to improve the sleep stage and efficiency estimation of sleep apnea patients using a UWB (Ultra-Wideband) radar. Motion and respiration extracted from the radar signal were used. Respiratory signal disturbances by motion artifacts and irregular respiration patterns of sleep apnea patients are compensated for in the preprocessing stage. Preprocessing calculates the standard deviation of the respiration signal for a shift window of 15 seconds to estimate thresholds for compensation and applies it to the breathing signal. The method for estimating the sleep stage is based on the difference in amplitude of two kinds of smoothed respirations signals. In smoothing, the window size is set to 10 seconds and 34 seconds, respectively. The estimated feature was processed by the k-nearest neighbor classifier and the feature filtering model to discriminate between the sleep periods of the rapid eye movement (REM) and non-rapid eye movement (NREM). The feature filtering model reflects the characteristics of the REM sleep that occur continuously and the characteristics that mainly occur in the latter part of this stage. The sleep efficiency is estimated by using the sleep onset time and motion events. Sleep onset time uses estimated features from the gradient changes of the breathing signal. A motion event was applied based on the estimated energy change in the UWB signal. Sleep efficiency and sleep stage accuracy were assessed with polysomnography. The average sleep efficiency and sleep stage accuracy were estimated respectively to be about 96.3% and 88.8% in 18 sleep apnea subjects.

• Journal of Korea Multimedia Society
• /
• v.15 no.11
• /
• pp.1369-1376
• /
• 2012

In this paper, we propose 2D industrial image registration method for the detection of defects. Proposed method performs preprocessing to smooth the original image with the preservation of the edge for the robust registration against general noise. Then, x-direction gradient magnitude image and corresponding binary image are generated. Density analysis around neighborhood regions per pixel are performed to generate feature image for preventing mis-registration due to moire-like patterns, which frequently happen in industrial images. Finally, 2D image registration based on phase correlation between feature images is performed to calculate translational parameters to align two images rapidly and optimally. Experimental results showed that the registration accuracy of proposed method for the real industrial images was 100% and our method was about twenty times faster than the previous method. Our fast and accurate method could be used for the real industrial applications.