• Journal of the Korea Society of Computer and Information
• /
• v.26 no.12
• /
• pp.45-52
• /
• 2021

In this paper, we propose a prediction model for osteopenia and osteoporosis based on a convolutional neural network(CNN) using computed tomography(CT) images. In a single CT image, CNN had a limitation in utilizing important local features for diagnosis. So we propose a compound model which has two identical structures. As an input, two different texture images are used, which are converted from a single normalized CT image. The two networks train different information by using dissimilarity loss function. As a result, our model trains various features in a single CT image which includes important local features, then we ensemble them to improve the accuracy of predicting osteopenia and osteoporosis. In experiment results, our method shows an accuracy of 77.11% and the feature visualize of this model is confirmed by using Grad-CAM.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.3991-4010
• /
• 2021

The Question Similarity Measurement of Chinese Crop Diseases and Insect Pests (QSM-CCD&IP) aims to judge the user's tendency to ask questions regarding input problems. The measurement is the basis of the Agricultural Knowledge Question and Answering (Q & A) system, information retrieval, and other tasks. However, the corpus and measurement methods available in this field have some deficiencies. In addition, error propagation may occur when the word boundary features and local context information are ignored when the general method embeds sentences. Hence, these factors make the task challenging. To solve the above problems and tackle the Question Similarity Measurement task in this work, a corpus on Chinese crop diseases and insect pests(CCDIP), which contains 13 categories, was established. Then, taking the CCDIP as the research object, this study proposes a Chinese agricultural text similarity matching model, namely, the AgrCQS. This model is based on mixed information extraction. Specifically, the hybrid embedding layer can enrich character information and improve the recognition ability of the model on the word boundary. The multi-scale local information can be extracted by multi-core convolutional neural network based on multi-weight (MM-CNN). The self-attention mechanism can enhance the fusion ability of the model on global information. In this research, the performance of the AgrCQS on the CCDIP is verified, and three benchmark datasets, namely, AFQMC, LCQMC, and BQ, are used. The accuracy rates are 93.92%, 74.42%, 86.35%, and 83.05%, respectively, which are higher than that of baseline systems without using any external knowledge. Additionally, the proposed method module can be extracted separately and applied to other models, thus providing reference for related research.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1998.10a
• /
• pp.385-391
• /
• 1998

As the increase in the use of computers and rapid development of information technology exacerbate auditors' judgmental process, achieving objectives(effectiveness) of control and auditing with limited resources requires auditors to identify relative importance and priority of controls. Consequently, auditors' review and evaluation of internal control is been highlighted in modrn auditing process. Especially, as more organizations adopted LAN(Local Area Network) systems. LAN environment is facing with problems arising from characteristics of theLAN and the evaluation of LAN controls. The modern in this study is LAN control evaluation model with information system auditing objectives. This study concludes that auditors' proper evalution of the importance of LAN controls contributes to the effective achievement of IS auditing objectives.

• Proceedings of the KIEE Conference
• /
• 1999.07b
• /
• pp.627-629
• /
• 1999

This paper presents a new FDI scheme based on dynamic fuzzy model(DFM) for the nonlinear system. The dynamic behavior of a nonlinear system is represented by a set of local linear models. The parameters of the DFM are identified in on-line and aggregated to generate a residual vector by the approximate reasoning. The neural network classifer learns the relationship between the residual vector and fault type and used both for the detection and isolation of process faults We apply the proposed FDI scheme to the FDI system design for a two-tank system and show the usefulness of the proposed scheme.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.05a
• /
• pp.213-216
• /
• 2005

In this paper, we propose the redirect streaming service model to enhance resurgent ability. The system consists of a redirect server, local streaming servers and clients. A redirect server searches the best streaming server. And streaming servers starts a requested service or resumes it when in a trouble. RTSP/RTP is one of the effective solutions to improve QOS in VOD, however a service can be broken off by an overloaded server or network especially in live. We designed and implemented not only a distributed streaming system that solves the broken-off service to enhance a resurgent system, but also DSSP, distributed streaming service protocol, which is adopted to implement this model. Those will improve the performance of streaming service using RTP/RTSP and be contributed to a real time streaming research to solve a service trouble.

• Korean Journal of Geomatics
• /
• v.5 no.1
• /
• pp.21-26
• /
• 2005

According to revisions of survey law taking effect on January 1, 2003, the Korean geodetic datum has been changed from a local geodetic to a world geodetic system. In this study, the datum transformation parameters especially for the maritime geographical data are determined. From database constructed through MGIS, a total of 492 coordinate pairs were selected and used in the parameter determination after outlier testing. Based on the parameter estimation, the Molodensky model is selected for datum transformation. For higher accuracy, Application of network optimization and a least squares collocation with Gaussian model has resulted in the accuracy better than 15 cm in coordinate transformation.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.35 no.8
• /
• pp.309-315
• /
• 1986

We proposed a preliminary Data Flow Machine Model(DNL1) operating on the basis of Node Label. In this model, all the PMs(Processing Modules) were synchronized with the content of LC(Level Counter) and were not implemented dy the processing cability on conditional nodes. This paper presents an architecture of a concurrent multiprocessor system which was developed from DNL1 with two additional types of memories, CF(Control Flag) and ETF (Enabled Token Flag). The CF memory holds the control condition flag ('1' or '0') to be referenced to when a node is fired and the ETF represents the firability of a certain node. Firable nodes are fetched to the PU(Processing Unit) and processed. This Data Flow system can be extended hierarchically by a network of simple modules. The principle working elements of the machine are a set of PMs, each of which performs the execution of the data flow procedures held in a local memory, NTM(Node Token Memory) within the PM.

• Journal of Communications and Networks
• /
• v.16 no.4
• /
• pp.385-396
• /
• 2014

In this paper, we present a formal method for modeling and checking an enhanced version of the carrier sense multiple access with collision avoidance protocol related to the IEEE 802.11 MAC layer, which has been proposed as the standard protocol for wireless local area networks. We deal mainly with the distributed coordination function (DCF) procedure of this protocol throughout a sequence of transformation steps. First, we use the unified modeling language state machines to thoroughly capture the behavior of wireless stations implementing a DCF, and then translate them into the input language of the UPPAAL model checking tool, which is a network of communicating timed automata. Finally, we proceed by checking of some of the safety and liveness properties, such as deadlock-freedom, using this tool.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.719-722
• /
• 2000

It is essential to develop the internetworking strategies between the OBP satellite B-ISDN and the terrestrial B-ISDN for the economical and efficient implementation of the future network infrastructure. We have considered the implementation issues of the emerging B-ISDN, especially interworking strategies of the terrestrial B-ISDN with the OBP satellite B-ISBN. We proposed interworking model between terrestrial B-ISDN and OBP satellite B-ISDN NNI signaling protocol, and describes the design and verification of the interworking protocol. For the verification, interworking protocol is modeled by Predicate/Action Net derived from Petri Net and the designed model is analyzed by reachability tree. Using Unix socket, the verified interworking protocol is implemented in two workstations and implemented interworking protocol is tested by local test method which is outstanding in error detection and recommended by ISO.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.8 no.3
• /
• pp.157-169
• /
• 2012

As of 2006. 3. the twitter offered in the USA has been one of the propaganda instrument used with ads and politics functioning speedy information diffusion on SNS communicated with others through 140 letters of short messages. and while twitter is using propaganda instrument, it keeps on trying to verify how it has an effect on. So, on the paper, I suggest new simulation model of information diffusion based on probability being able to predict the range of proliferation after it analyze the existing influence and the diffusion force on verification methods. It designed algorithm of verification and algorithm of prediction to use twitter's Open API with Python basement. It proved effectiveness on the model through the analysis to operate the twitter of practical local autonomous entity.