• Journal of information and communication convergence engineering
• /
• v.20 no.2
• /
• pp.103-112
• /
• 2022

The goal of pattern mining is to identify novel patterns in a database. High utility itemset mining (HUIM) is a research direction for pattern mining. This is different from frequent itemset mining (FIM), which additionally considers the quantity and profit of the commodity. Several algorithms have been used to mine high utility itemsets (HUIs). The original BPSO algorithm lacks local search capabilities in the subsequent stage, resulting in insufficient HUIs to be mined. Compared to the transfer function used in the original PSO algorithm, the V-shaped transfer function more sufficiently reflects the probability between the velocity and position change of the particles. Considering the influence of the acceleration factor on the particle motion mode and trajectory, a nonlinear acceleration strategy was used to enhance the search ability of the particles. Experiments show that the number of mined HUIs is 73% higher than that of the original BPSO algorithm, which indicates better performance of the proposed algorithm.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.313-315
• /
• 2021

Timely and accurate diagnosis of lung diseases using Chest X-ray images has been gained much attention from the computer vision and medical imaging communities. Although previous studies have presented the capability of deep convolutional neural networks by achieving competitive binary classification results, their models were seemingly unreliable to effectively distinguish multiple disease groups using a large number of x-ray images. In this paper, we aim to build an advanced approach, so-called Ensemble Knowledge Distillation (EKD), to significantly boost the classification accuracies, compared to traditional KD methods by distilling knowledge from a cumbersome teacher model into an ensemble of lightweight student models with parallel branches trained with ground truth labels. Therefore, learning features at different branches of the student models could enable the network to learn diverse patterns and improve the qualify of final predictions through an ensemble learning solution. Although we observed that experiments on the well-established ChestX-ray14 dataset showed the classification improvements of traditional KD compared to the base transfer learning approach, the EKD performance would be expected to potentially enhance classification accuracy and model generalization, especially in situations of the imbalanced dataset and the interdependency of 14 weakly annotated thorax diseases.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.2
• /
• pp.6-10
• /
• 2023

High-potential data can be predicted and stored in the cache to prevent cache misses, thus reducing the processor's request and wait times. As a result, the processor can work non-stop, hiding memory latency. By utilizing the temporal/spatial locality of memory access, the prefetcher introduced to improve the performance of these computers predicts the following memory address will be accessed. We propose a prefetcher that applies the GRU model, which is advantageous for handling time series data. Display the currently accessed address in binary and use it as training data to train the Gated Recurrent Unit model based on the difference (delta) between consecutive memory accesses. Finally, using a GRU model with learned memory access patterns, the proposed data prefetcher predicts the memory address to be accessed next. We have compared the model with the multi-layer perceptron, but our prefetcher showed better results than the Multi-Layer Perceptron.

• Proceedings of the Plant Resources Society of Korea Conference
• /
• 2022.09a
• /
• pp.100-100
• /
• 2022

Hovenia dulcis is a herbal plant, which belongs to the Rhamnaceae family and is a native of Japan, China and Korea. Its fruit stalk is called 'Jiguja' in Korea. It has been traditionally used as a medicinal plant in East Asia. It was reported to have detoxification effects on alcohol poisoning, and antioxidant, antidiabetic etc. Sample of 5 g was extracted with 50 mL of 70% EtOH. The supernatant was filtered by 0.45 ㎛ membrane filter before analysis. The UPLC system was performed on Waters alliance UPLC HSS T3 column (2.1 × 100 mm, 1.7 ㎛) with a UV detector. The gradient system was a binary eluent of 0.1% formic acid in water(A) and 0.1% formic acid in acetonitrile(B) with gradient conditions as follows: Initial, 10% B; 1 min, 10% B; 4 min, 20% B; 10 min, 25% B; 12 min, 30% B; 14 min, 90% B; 17 min, 90% B; flow rate of 0.2 mL/min. The samples were injected by 2 µL and were detected at UV 355 nm. As a result of analysis, chromatographic patterns appeared in two cases: samples analyzed for ampelopsin and myricetin, and samples analyzed for taxifolin and quercetin. Among the four compounds, the largest regional difference was found to be taxifolin.

• Korean Journal of Materials Research
• /
• v.34 no.5
• /
• pp.235-241
• /
• 2024

Gas identification techniques using pattern recognition methods were developed from four micro-electronic gas sensors for noxious gas mixture analysis. The target gases for the air quality monitoring inside vehicles were two exhaust gases, carbon monoxide (CO) and nitrogen oxides (NO_x), and two odor gases, ammonia (NH₃) and formaldehyde (HCHO). Four MEMS gas sensors with sensing materials of Pd-SnO₂ for CO, In₂O₃ for NO_X, Ru-WO₃ for NH₃, and hybridized SnO₂-ZnO material for HCHO were fabricated. In six binary mixed gas systems with oxidizing and reducing gases, the gas sensing behaviors and the sensor responses of these methods were examined for the discrimination of gas species. The gas sensitivity data was extracted and their patterns were determined using principal component analysis (PCA) techniques. The PCA plot results showed good separation among the mixed gas systems, suggesting that the gas mixture tests for noxious gases and their mixtures could be well classified and discriminated changes.

• ETRI Journal
• /
• v.46 no.4
• /
• pp.648-659
• /
• 2024

Author verification is challenging because of the diversity in writing styles. We propose an enhanced handwriting verification method that combines handcrafted and automatically extracted features. The method uses a genetic algorithm to reduce the dimensionality of the feature set. We consider offline Bangla handwriting content and evaluate the proposed method using handcrafted features with a simple logistic regression, radial basis function network, and sequential minimal optimization as well as automatically extracted features using a convolutional neural network. The handcrafted features outperform the automatically extracted ones, achieving an average verification accuracy of 94.54% for 100 writers. The handcrafted features include Radon transform, histogram of oriented gradients, local phase quantization, and local binary patterns from interwriter and intrawriter content. The genetic algorithm reduces the feature dimensionality and selects salient features using a support vector machine. The top five experimental results are obtained from the optimal feature set selected using a consensus strategy. Comparisons with other methods and features confirm the satisfactory results.

• Journal of Internet Computing and Services
• /
• v.16 no.6
• /
• pp.89-95
• /
• 2015

Traditional frequent pattern mining discovers valid patterns with no smaller frequency than a user-defined minimum threshold from databases. In this framework, an enormous number of patterns may be extracted by a too low threshold, which makes result analysis difficult, and a too high one may generate no valid pattern. Setting an appropriate threshold is not an easy task since it requires the prior knowledge for its domain. Therefore, a pattern mining approach that is not based on the domain knowledge became needed due to inability of the framework to predict and control mining results precisely according to the given threshold. Top-k frequent pattern mining was proposed to solve the problem, and it mines top-k important patterns without any threshold setting. Through this method, users can find patterns from ones with the highest frequency to ones with the k-th highest frequency regardless of databases. In this paper, we provide knowledge both on frequent and top-k pattern mining. Although top-k frequent pattern mining extracts top-k significant patterns without the setting, it cannot consider both item quantities in transactions and relative importance of items in databases, and this is why the method cannot meet requirements of many real-world applications. That is, patterns with low frequency can be meaningful, and vice versa, in the applications. High utility pattern mining was proposed to reflect the characteristics of non-binary databases and requires a minimum threshold. Recently, top-k high utility pattern mining has been developed, through which users can mine the desired number of high utility patterns without the prior knowledge. In this paper, we analyze two algorithms related to top-k high utility pattern mining in detail. We also conduct various experiments for the algorithms on real datasets and study improvement point and development direction of top-k high utility pattern mining through performance analysis with respect to the experimental results.

• The KIPS Transactions:PartB
• /
• v.15B no.4
• /
• pp.275-284
• /
• 2008

With increases in recent security requirements, biometric technology such as fingerprints, faces and iris recognitions have been widely used in many applications including door access control, personal authentication for computers, internet banking, automatic teller machines and border-crossing controls. Finger vein recognition uses the unique patterns of finger veins in order to identify individuals at a high level of accuracy. This paper proposes new device and methods for touchless finger vein recognition. This research presents the following five advantages compared to previous works. First, by using a minimal guiding structure for the finger tip, side and the back of finger, we were able to obtain touchless finger vein images without causing much inconvenience to user. Second, by using a hot mirror, which was slanted at the angle of 45 degrees in front of the camera, we were able to reduce the depth of the capturing device. Consequently, it would be possible to use the device in many applications having size limitations such as mobile phones. Third, we used the holistic texture information of the finger veins based on a LBP (Local Binary Pattern) without needing to extract accurate finger vein regions. By using this method, we were able to reduce the effect of non-uniform illumination including shaded and highly saturated areas. Fourth, we enhanced recognition performance by excluding non-finger vein regions. Fifth, when matching the extracted finger vein code with the enrolled one, by using the bit-shift in both the horizontal and vertical directions, we could reduce the authentic variations caused by the translation and rotation of finger. Experimental results showed that the EER (Equal Error Rate) was 0.07423% and the total processing time was 91.4ms.

• Journal of Korean Society of Transportation
• /
• v.23 no.6 s.84
• /
• pp.91-102
• /
• 2005

When it comes to the process of information storage, people are likely to organize individual information into the forms of groups rather than independent attributes, and put them together in their brains. Likewise, in case of finding the shortest path, this study suggests that a Hierarchical Road Network(HRN) model should be selected to browse the most desirable route, since the HRN model takes the process mentioned above into account. Moreover, most of drivers make a decision to select a route from origin to destination by road hierarchy. It says that the drivers feel difference between the link travel tine which was measured by driving and the theoretical link travel time. There is a different solution which has predicted the link travel time to solve this problem. By using this solution, the link travel time is predicted based on link conditions from time to time. The predicated link travel time is used to search the shortest path. Stochastic Process model uses the historical patterns of travel time conditions on links. The HRN model has compared favorably with the conventional shortest path finding model in tern of calculated speeds. Even more, the result of the shortest path using the HRN model has more similar to the survey results which was conducted to the taxi drivers. Taxi drivers have a strong knowledge of road conditions on the road networks and they are more likely to select a shortest path according to the real common sense.

• Analytical Science and Technology
• /
• v.11 no.3
• /
• pp.189-193
• /
• 1998

Binary alloys, $MoRu_3$ and $MoRh_3$, have been prepared using arc melting furnace. Mo and the noble metals Ru and Rh are the constituents of metallic insoluble residues, which were found in the early days of the post-irradiation studies on uranium oxide fuels. Detailed structural informations about these alloys have not been reported on JCPDS files of ICDD (International Centre for Diffraction Data). The results of X-ray diffraction study showed that the alloy was crystallized in hexagonal close-packing, well known as ${\varepsilon}$-phase. The X-ray diffraction patterns of these alloys matched well to that of $WRh_3$ with $P6_3/mmc$ of space group. The lattice parameters, a and c, were calculated using the least squares extrapolation. It was found from X-ray photoelectron spectroscopic measurements that Mo on the surface of the alloy was oxidized to Mo(6+), which could be removed by sputtering with Ar ions for approximately 15 minutes. The changes in binding energy of Mo, Ru, and Rh on the surface of the alloy were not observed. Magnetic susceptibility measurements resulted in the typical Pauli-paramagnetic behavior in the temperature range of 2 to 300 K.