• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3218-3241
• /
• 2023

Financial fraud undermines the sustainable development of financial markets. Financial statements can be regarded as the key source of information to obtain the operating conditions of listed companies. Current research focuses more on mining financial digital data instead of looking into text data. However, text data can reveal emotional information, which is an important basis for detecting financial fraud. The audit opinion of the financial statement is especially the fair opinion of a certified public accountant on the quality of enterprise financial reports. Therefore, this research was carried out by using the data features of 4,153 listed companies' financial annual reports and audits of text opinions in the past six years, and the paper puts forward a financial fraud detection model integrating audit opinions. First, the financial data index database and audit opinion text database were built. Second, digitized audit opinions with deep learning Bert model was employed. Finally, both the extracted audit numerical characteristics and the financial numerical indicators were used as the training data of the LightGBM model. What is worth paying attention to is that the imbalanced distribution of sample labels is also one of the focuses of financial fraud research. To solve this problem, data enhancement and Focal Loss feature learning functions were used in data processing and model training respectively. The experimental results show that compared with the conventional financial fraud detection model, the performance of the proposed model is improved greatly, with Area Under the Curve (AUC) and Accuracy reaching 81.42% and 78.15%, respectively.

• Korean Journal of Veterinary Research
• /
• v.39 no.1
• /
• pp.138-147
• /
• 1999

Detection and distribution of bovine viral diarrhea virus(BVDV) was studied in formalin-fixed, paraffin-embedded tissues from two naturally infected cattle by in situ hybridization with a non-radioactive biotinylated probe. A 600 base pair cDNA probe from BVDV B-25 strain was used for probe. The whole procedure of ISH to diagnose was carried out within 1~2 hours in $Microprobe^{TM}$ capillary action system. The biotin-labelled probe was demonstrated after hybridization under standard conditions by the application of streptoavidin and biotinylated alkaline phosphatase. Alkaline phosphatase was visualized using a fast red TR/naphthol phosphatase and the sections were counterstained with hematoxylin. We have obtained the result of positive reactions in digestive tract(sm1.all intestine and colon) and epidermis of tongue in the state of the intact tissues. The result suggested that in situ hybridization method can be considered as a useful diagnostic technique for detection of specific nucleic acid sequences of BVDV.

• Journal of Environmental Science International
• /
• v.22 no.12
• /
• pp.1615-1624
• /
• 2013

The aims of this study were to investigate and confirm the occurrence and distribution patterns of blood lipid lower agents (BLLAs) in Nakdong river basin (mainstream and its tributaries). 4 (atorvastatin, lovastatin, mevastatin and simvastatin) out of 5 statins and 2 (clofibric acid and zemfibrozil) out of 3 fibrates were detected in 29 sampling sites and simvastatin (>50%) was predominant compound followed by atorvastatin, lovastatin and clofibric acid. The total concentration levels of BLLAs on April, August and November 2009 in surface water samples ranged from ND~25.7 ng/L, ND~18.8 and ND to 38.8 ng/L, respectively. The highest concentration level of BLLAs in the mainstream and tributaries in Nakdong river were Goryeong and Jincheon-cheon, respectively. The sewage treatment plants (STPs) along the river affect the BLLAs levels in river and the BLLAs levels decreased with downstream because of dilution effects.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.4
• /
• pp.197-203
• /
• 2015

Nuclear power plant has increased continuously for power production in all over the world and the interest about nuclear accident and the dismantling of aging nuclear power plant has been a growing. The leaked radioactive source that is generated by radiation accidents must detect and remove to minimized the damage as soon as possible. Gamma-ray detection system that have been developed until now cannot provide the precise position of radioactive sources because they detect and imaging the position of radiation sources in just two dimensions. In this paper, stereo gamma ray detection system has developed and the algorithm for calculation of the distance has implemented to be able to measure the distribution of the leakage gamma ray source for the system. Stereo camera calibration for distance detection was conducted with the correction pattern and LED light and we carried out performance test of the system for the LED light source and a gamma ray source. In both experiments the results of the performance test, it was confirmed to have a 5% error. The results of this paper is used as a material for the development of gamma-ray imaging device.

• The Journal of the Korea Contents Association
• /
• v.9 no.4
• /
• pp.19-26
• /
• 2009

The microwave exposure device for microwave breast cancer detection consists of RF transceiver and several antennas. The microwave information of object acquired from the microwave exposure device can be calculated permittivity and conductivity by using the inverse scattered analysis. In this paper, we have developed the software for detecting breast cancers based on microwave tomography, by which users not only can check out the existence of breast cancers through the permittivity and conductivity information analysis of the object's internal, but also can analysis easily information for distribution of breast cancers. The developed software provides the function for visualizing the captured permittivity and conductivity information as 2D or 3D color images on which users can easily detect the existence of breast cancers. For more detailed analysis of tomography images, the proposed software also has provided the functions for displaying their cutting profiles as well as position and size information of special area in them.

• Research in Plant Disease
• /
• v.7 no.2
• /
• pp.80-85
• /
• 2001

This study was carried out to examine tne incidences of virus diseases in lily plants cultivated in highland areas, and to develop an effective detection method. Viral symptoms on lilies in the highland areas were differentiated into mosaic, crinkle, mottle, stripe and line pattern. The distribution of symptoms on infected plants was 43.8% of mosaic, 29.2% of crinkle, and 10.9% of mottle symptoms. Six viruses such as Lily symptomless vires(LSV), Cucumber mosaic virus (CMV), Lily mottle virus (LMoV), Lily virus X (LVX, Potexvirus), Tabacco mosaic virus (TMV,Tobamovirus), and Tabacco rattle virus (TRV,Tobravirus) were detected from the infected lilies. Infection rate of Lilium oriental (cvs. Casablanca and Marcopolo) was 2~4 times higher than that of L. asiatic (cvs. Solemio and Prato). Virus detection on lilies by one-step RT-PCR (by using reverse transcription and polymerase chain reaction simultaneously) was more rapid rapid and reliable than by the conventional RT-PCR method.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.58 no.2
• /
• pp.141-152
• /
• 2022

As a method to understand the ecological habits around the artificial reef, various reports such as fishing gear survey, diving, sound survey, underwater CCTV and camera, etc. are reported. Among them, the sound survey method is carried out by installing an acoustic system on the ship and can be investigated regardless of the marine environment such as time constraints and turbidity. Such method, however, takes a lot of manpower and time as the ship travels at a constant speed. Investigations around artificial reefs are being conducted in an artificial way, and a lot of time and labor are consumed as such. Maritime buoys have been operated for various purposes such as route signs, weather observation, marine environment monitoring and defense monitoring for navigation safety in the past, but studies on monitoring systems for ecological habits and distribution of fish using marine buoys are remarkably insufficient. Therefore, this study aims to develop a system that allows users to directly monitor fish group detector data by estimating the distribution of fish groups around artificial reefs and using wireless communication at sea. In order to confirm the suitability of the maritime buoy used in this study, it was operated to compare data using LTE-equipped buoys capable of wireless communication and a data logger-type system buoy. Data transmission of buoys capable of LTE communication was carried out in a 10-minute ON, 10-minute OFF method due to the limitation of the power supply capacity, and data of the data logger-type buoy received full data. We compared and analyzed the data received from the two fish detectors. It is expected that real-time monitoring of the wireless buoy detection device using LTE will be possible through future research.

• Journal of Korean Society of Water and Wastewater
• /
• v.26 no.6
• /
• pp.767-777
• /
• 2012

Annular Biofilm Reactor (ABR) equipped with coupons of three different pipe materials (STS 304, PVC, PE) was used to generate drinking water biofilm samples. The level of assimilable organic carbon (AOC) during the sample generation period was $37.3{\mu}g/L$, and this level did not seem to be low enough to limit the formation of biofilm in this study. Terminal-restriction fragment length polymorphism (T-RFLP) analyses determined T-RF profile as early as 3 h of exposure on PVC coupons. Average surface roughness ($R_a$) measured by atomic force microscopic analyses was 125.7 nm for PVC, and this value was higher than for STS (71.6 nm) and PE (74.0 nm). However, biofilm formation was faster on STS (6 h) than on PE (12 h), which indicated that surface roughness might not be the only factor that controlled the initiation of biofilm development. Upon detection of the T-RF peaks, richness (S) and diversity indices such as Shannon (H) and Simpson (1/D) demonstrated a rather slow increase until 48 h followed by rapid increase regardless of the pipe materials. Differences of microbial community structures among the biofilm samples were determined based on the cluster analysis using Jaccard coefficients (Sj). Biofilm communities could be divided into two distinct groups according to the exposure time regardless of the pipe materials. First group contained a young (< 48 h) biofilm samples (10 out of 11) but second group contained a mature (${\geq}$ 48 h) samples (11 out of 14). Results suggested that, due to the complexity of biofilm, the targeting of the first group of cluster was crucial for optimizing the management of drinking water distribution systems and controlling microbial growth.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.183-203
• /
• 2018

News articles are the most suitable medium for examining the events occurring at home and abroad. Especially, as the development of information and communication technology has brought various kinds of online news media, the news about the events occurring in society has increased greatly. So automatically summarizing key events from massive amounts of news data will help users to look at many of the events at a glance. In addition, if we build and provide an event network based on the relevance of events, it will be able to greatly help the reader in understanding the current events. In this study, we propose a method for extracting event networks from large news text data. To this end, we first collected Korean political and social articles from March 2016 to March 2017, and integrated the synonyms by leaving only meaningful words through preprocessing using NPMI and Word2Vec. Latent Dirichlet allocation (LDA) topic modeling was used to calculate the subject distribution by date and to find the peak of the subject distribution and to detect the event. A total of 32 topics were extracted from the topic modeling, and the point of occurrence of the event was deduced by looking at the point at which each subject distribution surged. As a result, a total of 85 events were detected, but the final 16 events were filtered and presented using the Gaussian smoothing technique. We also calculated the relevance score between events detected to construct the event network. Using the cosine coefficient between the co-occurred events, we calculated the relevance between the events and connected the events to construct the event network. Finally, we set up the event network by setting each event to each vertex and the relevance score between events to the vertices connecting the vertices. The event network constructed in our methods helped us to sort out major events in the political and social fields in Korea that occurred in the last one year in chronological order and at the same time identify which events are related to certain events. Our approach differs from existing event detection methods in that LDA topic modeling makes it possible to easily analyze large amounts of data and to identify the relevance of events that were difficult to detect in existing event detection. We applied various text mining techniques and Word2vec technique in the text preprocessing to improve the accuracy of the extraction of proper nouns and synthetic nouns, which have been difficult in analyzing existing Korean texts, can be found. In this study, the detection and network configuration techniques of the event have the following advantages in practical application. First, LDA topic modeling, which is unsupervised learning, can easily analyze subject and topic words and distribution from huge amount of data. Also, by using the date information of the collected news articles, it is possible to express the distribution by topic in a time series. Second, we can find out the connection of events in the form of present and summarized form by calculating relevance score and constructing event network by using simultaneous occurrence of topics that are difficult to grasp in existing event detection. It can be seen from the fact that the inter-event relevance-based event network proposed in this study was actually constructed in order of occurrence time. It is also possible to identify what happened as a starting point for a series of events through the event network. The limitation of this study is that the characteristics of LDA topic modeling have different results according to the initial parameters and the number of subjects, and the subject and event name of the analysis result should be given by the subjective judgment of the researcher. Also, since each topic is assumed to be exclusive and independent, it does not take into account the relevance between themes. Subsequent studies need to calculate the relevance between events that are not covered in this study or those that belong to the same subject.

• Korean Journal of Veterinary Research
• /
• v.36 no.2
• /
• pp.405-415
• /
• 1996

This study was carried out to investigate the distribution of inclusion bodies in the tissues as well as to observe the general histopathological lesions of dogs infected with canine distemper. And also, the reliability of diagnostic values of inclusion bodies and the distribution of viral antigen in tissues were inspected by immunohistochemistry with monoclonal antibody. The results obtained were as follows; 1. Pneumonia observed in dogs infected with canine distemper virus was classified into interstitial, broncho-, and broncho-interstitial pneumonia histopathologically. Each occurring ratio was 35, 45 and 20%. 2. Histopathological classification of the canine distemper encephalitis was 20% in acute, 60% in subacute, and 20% in chronic encephalitis, respectively. 3. The organs in which inclusion bodies were predominantly distributed were stomach(82.6%), cerebellum(62.9%), lung(62.1%), cerebrum(50.0%), urinary bladder (46.1%), kidney(36.0%) and pancreas(25.0%). Intracytoplasmic inclusion bodies were mainly observed in the organs except the brain. 4. Canine distemper virus antigens were detected in the numerous tissues as well as in the inclusion bodies observed in the various organs. Antigen detection ratios in the lung, cerebellum and cerebrum were 68.9, 70.4 and 52.2%, respectively. These ratios were somewhat higher than those of inclusion bodies observed in the organs. 5. Canine distemper virus was mainly distributed in astrocytes and ependymal cells in the brain. These results suggested that the histopathologic diagnosis of canine distemper was reliable, and the spread of canine distemper virus in the brain was related with cerebrospinal fluid pathway.