• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.79-81
• /
• 2014

People increasingly rely on sources of information from E-commerce reviews. Product reviews is an important determinant of potential customers' buying choices. They are also utilized by product manufacturers to find problems of their products and to collect competitive intelligence information about their competitors. Unfortunately, it is well-known that many online product reviews are not made by genuine costumers of products. Reviewers could write some undeserving positive reviews to promote or fake negative reviews to defame some certain product, and we call them fake product reviews. Fake product review detection makes an attempt to detect fake reviews and removes them to restore the truthful ones for readers. To the best of our knowledge, there is still less published study on this problem. In this paper, we make a survey and an attempt to give a brief overview on fake product review detection. The related work of fake product review detection is presented including web spam and spam email. Then some methods to detect fake reviews are introduced and summarized. The trend of fake product review detection is concluded finally.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.10 no.1
• /
• pp.139-148
• /
• 2015

In this paper, a new algorithm is proposed for detecting the product number of each thick plate and extracting each character of the product number from a image which contains several thick plates. In general, a image of thick plates contains several steal plates. To obtain the product number from the image, we first need to separate each plate. To do so, we use the line edges of thick plates and a clustering algorithm. After separating each plate, background parts are eliminated from the image of each plate. Background parts of an individual thick plate image consist of the dark part of steel and the white part of paint which is used for printing the product number. We propose a two-tiered method where dark background parts are first eliminated and then white parts are eliminated. Finally, each character is extracted from the product number image using the characteristics of product number. The results of the experiments on the various steal plates images emphasize that the proposed algorithm detects each thick plate and extracts the product number from a image effectively.

• Journal of IKEEE
• /
• v.19 no.4
• /
• pp.541-547
• /
• 2015

In current semiconductor manufacturing, as the feature size of integrated circuit (IC) devices continuously shrinks, detecting endpoint in plasma etching process is more difficult than before. For endpoint detection, various kinds of sensors are installed in semiconductor manufacturing equipments, and sensor data are gathered with predefined sampling rate. Generally, detecting endpoint is performed using OES data of by-product. In this study, OES data of both by-product and etchant gas are used to improve reliability of endpoint detection. For the OES data pre-processing, a combination of Signal to Noise Ratio (SNR) and Principal Component Analysis (PCA),are used. Polynomial Regression and Expanded Hidden Markov model (eHMM) technique are applied to pre-processed OES data to detect endpoint.

• Journal of Microbiology and Biotechnology
• /
• v.23 no.12
• /
• pp.1717-1725
• /
• 2013

Traditional soybean paste from Shandong Liangshan and Tianyuan Jiangyuan commercial soybean paste were chosen for analysis and comparison of their bacterial and fungal dynamics using denaturing gel gradient electrophoresis and 16S rRNA gene clone libraries. The bacterial diversity results showed that more than 20 types of bacteria were present in traditional Shandong soybean paste during its fermentation process, whereas only six types of bacteria were present in the commercial soybean paste. The predominant bacteria in the Shandong soybean paste were most closely related to Leuconostoc spp., an uncultured bacterium, Lactococcus lactis, Bacillus licheniformis, Bacillus spp., and Citrobacter freundii. The predominant bacteria in the Tianyuan Jiangyuan soybean paste were most closely related to an uncultured bacterium, Bacillus licheniformis, and an uncultured Leuconostoc spp. The fungal diversity results showed that 10 types of fungi were present in the Shandong soybean paste during the fermentation process, with the predominant fungi being most closely related to Geotrichum spp., an uncultured fungal clone, Aspergillus oryzae, and yeast species. The predominant fungus in the commercial soybean paste was Aspergillus oryzae.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.108-111
• /
• 2022

Applying deep learning to machine vision systems for defect detection of products requires vast amounts of training data about various defect cases. However, since data imbalance occurs according to the type of defect in the actual manufacturing industry, it takes a lot of time to collect product images enough to generalize defect cases. In this paper, we apply a Siamese neural network that can be learned with even a small amount of data to product defect detection, and modify the image pairing method and contrastive loss function by properties the situation of product defect image data. We indirectly evaluated the embedding performance of Siamese neural networks using AUC-ROC, and it showed good performance when the images only paired among same products, not paired among defective products, and learned with exponential contrastive loss.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.2
• /
• pp.9-16
• /
• 2018

Recently, as various IP set-top boxes based on Android OS have been widely used in general households and public facilities, complaints about services and set-top boxes have continued to increase as much as other smart devices. In order to reduce this problem, the manufacturer performs the testing work before the product is commercialized. However, the testing can reduce potential defects in the product, but it can not prove that the product is free of defects. Therefore, the quality of the product can vary depending on how effective testing techniques are introduced. In this paper, we propose a new exploratory testing method that minimizes test case creation time and makes it easier to plan and execute test while simultaneously learning how to run the product under test. Using the first proposed method, the test time is reduced by about 16.7 hours and the defect detection rate is 25.4% higher than the formal specification-based testing method. Informally, the test time was shortened by about 4.7 hours and the defect detection rate was 13% higher than the informal experience-based testing method.

• The Plant Pathology Journal
• /
• v.19 no.3
• /
• pp.159-161
• /
• 2003

The coat protein (CP) gene of Apple mosaic virus (ApMV), a member of the genus Ilarvirus, was selected for the design of virus-specific primers for amplification and molecular detection of the virus in cultivated apple. A combined assay of reverse transcription and polymerase chain reaction (RT-PCR) was performed with a single pair of ApMV-specific primers and crude nucleic acid extracts from virus-infected apple for rapid detection of the virus. The PCR product was verified by restriction mapping analysis and by sequence determination. The lowest concentration of template viral RNA required for detection was 100 fg. This indicates that the RT-PCR for detection of the virus is a 10$^3$times more sensitive, reproducible and time-saving method than the enzyme-linked immunosorbent assay. The specificity of the primers was verified using other unrelated viral RNAs. No PCR product was observed when Cucumber mosaic virus (Cucumovirus) or a crude extract of healthy apple was used as a template in RT-PCR with the same primers. The PCR product (669 bp) of the CP gene of the virus was cloned into the plasmid vector and result-ant recombinant (pAPCP1) was selected for molecule of apple transformation to breed virus-resistant transgenic apple plants as the next step. This method can be useful for early stage screening of in vitro plantlet and genetic resources of resistant cultivar of apple plants.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.2
• /
• pp.77-83
• /
• 2022

In this paper, the two-stage object detection approach is proposed to implement a deep learning-based label inspection system on edge computing environments. Since the label printed on the products during the production process contains important information related to the product, it is significantly to check the label information is correct. The proposed system uses the lightweight deep learning model that able to employ in the low-performance edge computing devices, and the two-stage object detection approach is applied to compensate for the low accuracy relatively. The proposed Two-Stage object detection approach consists of two object detection networks, Label Area Detection Network and Character Detection Network. Label Area Detection Network finds the label area in the product image, and Character Detection Network detects the words in the label area. Using this approach, we can detect characters precise even with a lightweight deep learning models. The SF-YOLO model applied in the proposed system is the YOLO-based lightweight object detection network designed for edge computing devices. This model showed up to 2 times faster processing time and a considerable improvement in accuracy, compared to other YOLO-based lightweight models such as YOLOv3-tiny and YOLOv4-tiny. Also since the amount of computation is low, it can be easily applied in edge computing environments.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.2
• /
• pp.95-101
• /
• 2020

In this paper, we consider a group testing (GT) framework which is to find a set of defective samples out of a large number of samples. To handle this framework, we propose a maximum product detection algorithm (MPDA) which is based on maximum a posteriori probability (MAP). The key idea of this algorithm exploits iterative detection to propagate belief to neighbor samples by exchanging marginal probabilities between samples and output results. The belief propagation algorithm as a conventional approach has been used to detect defective samples, but it has computational complexity to obtain the marginal probability in the output nodes which combine other marginal probabilities from the sample nodes. We show that the our proposed MPDA provides a benefit to reduce computational complexity up to 12% in runtime, while its performance is only slightly degraded compared to the belief propagation algorithm. And we verify the simulations to compare the difference of performance.

• Food Science of Animal Resources
• /
• v.31 no.5
• /
• pp.658-662
• /
• 2011

In this study, 543 samples of press hams, sausages, processed ground meat and processed cheese acquired from retail markets in Seoul and Gyeonggi province in Korea from 2005 to 2010 were monitored using a one-step multiplex polymerase chain reaction (PCR) method that involves the amplification of specific soya or maize endogenous genes and the amplification of 35S promoter (p35S) and nopaline synthase terminator (tNOS) for GMO detection. Among the 543 samples, 477 samples were amplified for maize and/or soybean endogenous genes. Although one sausage sample collected in 2008 showed amplification of tNOS, the result was assumed to be false positive based on the results from further tests of other sausage samples of the same brand. Our results demonstrate the absence of GM soya and/or maze of livestock products in the Korean market during 2005-2010. In addition, the one-step multiplex PCR using previously constructed primer sets appears to be useful as a screening method for the detection of GMOs in processed livestock products. However, more specific methods should be established and employed to detect the event-specific GM gene for positive reaction samples by screening tests in processed livestock products.