• Title/Summary/Keyword: Dual classification method

Search Result 39, Processing Time 0.023 seconds

A SIMPLEX TYPE METHOD USING MULTIPLE UPDATINGS FOR SPECIALLY STRUCTURED LINEAR PROGRAMS

  • Oh, Seyoung;Chung, Seiyoung
    • Journal of applied mathematics & informatics
    • /
    • v.3 no.1
    • /
    • pp.25-34
    • /
    • 1996
  • Many important large-scale linear programs with special structures lead to special computational procedures which are more ef-ficient than the ordinary procedure of the generalized methods. Prob-lems and their solvers taking advantage of multiple updatings of the basis in the dual simplex type method are presented. Computational results run by the efficient algorithm and a presented. Computational results run by the efficient algorithm and a standard code MINOS for the test problems are compared and analyzed. It is shown that the amount of work for the optimal solution for the prblem can be re-duced by the new algorithm.

Anomaly-based Alzheimer's disease detection using entropy-based probability Positron Emission Tomography images

  • Husnu Baris Baydargil;Jangsik Park;Ibrahim Furkan Ince
    • ETRI Journal
    • /
    • v.46 no.3
    • /
    • pp.513-525
    • /
    • 2024
  • Deep neural networks trained on labeled medical data face major challenges owing to the economic costs of data acquisition through expensive medical imaging devices, expert labor for data annotation, and large datasets to achieve optimal model performance. The heterogeneity of diseases, such as Alzheimer's disease, further complicates deep learning because the test cases may substantially differ from the training data, possibly increasing the rate of false positives. We propose a reconstruction-based self-supervised anomaly detection model to overcome these challenges. It has a dual-subnetwork encoder that enhances feature encoding augmented by skip connections to the decoder for improving the gradient flow. The novel encoder captures local and global features to improve image reconstruction. In addition, we introduce an entropy-based image conversion method. Extensive evaluations show that the proposed model outperforms benchmark models in anomaly detection and classification using an encoder. The supervised and unsupervised models show improved performances when trained with data preprocessed using the proposed image conversion method.

Classification of Piperazinylalkylisoxazole Library by Recursive Partitioning

  • Kim, Hye-Jung;Park, Woo-Kyu;Cho, Yong-Seo;No, Kyoung-Tai;Koh, Hun-Yeong;Choo, Hyun-Ah;Pae, Ae-Nim
    • Bulletin of the Korean Chemical Society
    • /
    • v.29 no.1
    • /
    • pp.111-116
    • /
    • 2008
  • A piperazinylalkylisoxazole library containing 86 compounds was constructed and evaluated for the binding affinities to dopamine (D3) and serotonin (5-HT2A/2C) receptor to develop antipsychotics. Dopamine antagonists (DA) showing selectivity for D3 receptor over the D2 receptor, serotonin antagonists (SA), and serotonin-dopamine dual antagonists (SDA) were identified based on their binding affinity and selectivity. The analogues were divided into three groups of 7 DAs (D3), 33 SAs (5-HT2A/2C), and 46 SDAs (D3 and 5-HT2A/2C). A classification model was generated for identifying structural characteristics of those antagonists with different affinity profiles. On the basis of the results from our previous study, we conducted the generation of the decision trees by the recursive-partitioning (RP) method using Cerius2 2D descriptors, and identified and interpreted the descriptors that discriminate in-house antipsychotic compounds.

Classification of Urinary Stone into Uric Acid & Non-uric Acid by Dual-Energy (이중에너지 전산화단층촬영을 이용한 요로결석의 성분 분석에 관한 연구)

  • Myung-Jin Jung;Sung-Gil Kim
    • Journal of the Korean Society of Radiology
    • /
    • v.17 no.6
    • /
    • pp.835-841
    • /
    • 2023
  • The aim of this study is to evaluate the diagnostic ability of dual-energy computed tomography (DECT) for Composition determination of urinary stones in phantom model. Seventeen cases with urinary stones who underwent DECT were enrolled in the study. The composition of the urinary stones was extracted from the seventeen patients were analyzed with DECT in phantom model with fresh pork. The volume scan method using Dual-energy software was used and the scanned image sets were assessed. All 17 urinary stones of the phantom model were analyzed according to the stone composition using DE stone Analysis were divided into uric acid stones (n=6, 35.29%) and non-uric acid stones (n=11, 64.71%). These urinary stones were pathologically confirmed. The mean attenuation values of uric acid stones at 135 kV, 100 kV and 80 kV was 348.87 ± 166.37 HU, 345.33 ± 151.18 HU and 337.94 ± 172.77 HU, respectively. The mean attenuation values of non-uric acid stones at 135 kV, 100 kV and 80 kV was 551.93 ± 297.09 HU, 747.04 ± 351.31 HU and 958.19 ± 424.72 HU, respectively. At 80 kV, uric acid stones and non-uric acid stones showed significant difference in the attenuation values(P<0.05). The attenuation values of DECT could differentiate the compositions of urinary stones between uric acid and non-uric acid stones at 80 kV in phantom model.

Diagnostic Method for Inborn Metabolic Disorders using differentiation between D- and R- Isomers on GC-MS (D체와 R체 이성질체 판별과 GC-MS를 이용한 유전성 대사이상질환의 진단법 개발)

  • Yoon, Hye-Ran
    • Journal of The Korean Society of Inherited Metabolic disease
    • /
    • v.15 no.2
    • /
    • pp.65-71
    • /
    • 2015
  • Since the secretion of specific chiral isomers in urine (or plasma) is very crucial to diagnose some inborn metabolic disorders, clinical application of dual column achiral differential method has been performed for the absolute configuration of chiral compounds. Extracted from the acidified urine with diethyl ether, carboxylic functional group of organic acid (stereoisomers of the volatile) was derivatized with (-)-menthylation or (S)-(+)-3-methyl-2-butylation and followed by O-trifluoroacylation. Each of the enantiomers was accurately separated from the library matched double column (achiral) with a retention index (I). In various inborn metabolic disease urines, absolute chirality was identified correctly in the urine (10 patients) with inborn metabolic disease (including secretion of D, L- lactic acid, D, L-3-hydroxybutyric acid, and D, L-2-hydroxyglutaric acid). In this study, we identified and isolated the volatile diastereomer as a useful diagnostic marker, this successful application to urine specimens may be useful for diagnostic classification of inherited metabolic disorders.

A study on Defect Diagnosis of Gas Turbine Engine Using Hybrid SVM-ANN in Off-Design Region

  • Seo, Dong-Hyuck;Choi, Won-Jun;Roh, Tae-Seong;Choi, Dong-Whan
    • Proceedings of the Korean Society of Propulsion Engineers Conference
    • /
    • 2008.03a
    • /
    • pp.72-79
    • /
    • 2008
  • The weak point of the artificial neural network(ANN) is that it is easy to fall in local minima when it learns too much nonlinear data. Accordingly, the classification ratio must be low. To overcome this weakness, the hybrid method has been proposed. That is, the ANN learns data selectively after detecting the defect position by the support vector machine(SVM). First, the SVM has been used for determination of the defect position and then the magnitude of the defect has been measured by the ANN. In off-design condition, the operation region of the engine is wide and the nonlinearity of learning data increases. The module system, dividing the whole operating region into reasonably small-size sections, has been suggested to solve this problem. In this study, the proposed algorithm has diagnosed the defects of triple components as well as single and dual components of the gas turbine engine in off-design condition.

  • PDF

The Records and Archives Administrative Reform in China in 1930s (1930년대 중국 문서당안 행정개혁론의 이해)

  • Lee, Won-Kyu
    • The Korean Journal of Archival Studies
    • /
    • no.10
    • /
    • pp.276-322
    • /
    • 2004
  • Historical interest in China in 1930s has been mostly focused on political characteristic of the National Government(國民政府) which was established by the KMT(中國國民黨) as a result of national unification. It is certain that China had a chance to construct a modern country by the establishment of the very unified revolutionary government. But, it was the time of expanding national crises that threatened the existence of the country such as the Manchurian Incident and the Chinese-Japanese War as well as the chaos of the domestic situation, too. So it has a good reason to examine the characteristic and pattern of the response of the political powers of those days. But, as shown in the recent studies, the manifestation method of political power by the revolutionary regime catches our attention through the understanding of internal operating system. Though this writing started from the fact that the Nationalist Government executed the administrative reform which aimed at "administrative efficiency" in the middle of 1930s, but it put stress on the seriousness of the problem and its solution rather than political background or results. "Committee on Administrative Efficiency(行政效率委員會)", the center of administrative reform movement which was established in 1934, examined the plan to execute the reform through legislation by the Executive Council(行政院) on the basis of the results of relevant studies. They claimed that the construction of a modern country should be performed by not political revolution anymore but by gradual improvement and daily reform, and that the operation of the government should become modern, scientific and efficient. There were many fields of administrative reform subjects, but especially, the field of records and archives adminstration(文書檔案行政) was studied intensively from the initial stage because that subject had already been discussed intensively. They recognized that records and archives were the basic tool of work performance and general activity but an inefficient field in spite of many input staff members, and most of all, archival reform bring about less conflicts than the fields of finance, organization and personnel. When it comes to the field of records adminstration, the key subjects that records should be written simply, the process of record treatment should be clear and the delay of that should be prevented were already presented in a records administrative meeting in 1922. That is, the unified law about record management was not established, so each government organization followed a conventional custom or performed independent improvement. It was through the other records administrative workshop of the Nationalist Government in 1933 when the new trend was appeared as the unified system improvement. They decided to unify the format of official records, to use marker and section, to unify the registration of receipt records and dispatch records and to strengthen the examination of records treatment. But, the method of records treatment was not unified yet, so the key point of records administrative reform was to establish a unified and standard record management system for preventing repetition by simplifying the treatment procedure and for intensive treatment by exclusive organizations. From the foundation of the Republic of China to 1930s, there was not big change in the field of archives administration, and archives management methods were prescribed differently even in the same section as well as same department. Therefore, the point at issue was to centralize scattered management systems that were performed in each section, to establish unified standard about filing and retention period allowance and to improve searching system through classification and proper number allowance. Especially, the problem was that each number system and classification system bring about different result due to dual operation of record registration and archives registration, and that strict management through mutual contrast, searching and application are impossible. Besides, various problems such as filing tools, arrangement method, preservation facilities & equipment, lending service and use method were raised also. In the process this study for the system improvement of records and archives management, they recognized that records and archives are the identical thing and reached to create a successive management method of records and archives called "Records and Archives Chain Management Method(文書檔案連鎖法)" as a potential alternative. Several principles that records and archives management should be performed unitedly in each organization by the general record recipient section and the general archives section under the principle of task centralization, a consistent classification system should be used by classification method decided in advance according to organizational constitution and work functions and an identical number system should be used in the process of record management stage and archive management stage by using a card-type register were established. Though, this "Records and Archives Chain Management Method" was developed to the stage of test application in several organizations, but it was not adopted as a regular system and discontinued. That was because the administrative reform of the Nationalist Government was discontinued by the outbreak of the Chinese-Japanese War. Even though the administrative reform in the middle of 1930s didn't produce practical results but merely an experimentation, it was verified that the reform against tradition and custom conducted by the Nationalist Government that aimed for the construction of a modern country was not only a field of politics, but on the other hand, the weak basis of the government operation became the obstacle to the realization of the political power of the revolutionary regime. Though the subject of records and archives administrative reform was postponed to the future, it should be understood that the consciousness of modern records and archives administration and overall studies began through this examination of administrative reform.

Study on the influence of sewer network simplification on urban inundation modelling results (하수관망의 간소화가 도시침수 모의에 미치는 영향 분석에 관한 연구)

  • Lee, Seung-Soo;Pakdimanivong, Mary;Jung, Kwan-Sue;Kim, Yeonsu
    • Journal of Korea Water Resources Association
    • /
    • v.51 no.4
    • /
    • pp.347-354
    • /
    • 2018
  • In urban areas, runoff flow is drained through sewer networks as well as surface areas. Therefore, it is very important to consider sewer networks as a component of hydrological drainage processes when conducting urban inundation modelling. However, most researchers who have implemented urban inundation/flood modelling, instinctively simplified the sewer networks without the appropriate criteria. In this research, a 1D-2D fully coupled urban inundation model is applied to estimate the influence of sewer network simplification on urban inundation modelling based on the dendritic network classification. The one-dimensional (1D) sewerage system analysis model, which was introduced by Lee et al. (2017), is used to simulate inlet and overflow phenomena by interacting with surface flow. Two-dimensional (2D) unstructured meshes are also applied to simulate surface flow and are combined with the 1D sewerage analysis model. Sewer network pipes are simplified based on the dendritic network classification method, namely the second and third order, and all cases of pipes are conducted as a control group. Each classified network case, including a control group, is evaluated through their application to the 27 July 2011 extreme rainfall event, which caused severe inundation damages in the Sadang area in Seoul, South Korea. All cases are compared together regarding inundation area, inflow discharge and overflow discharge. Finally, relevant criterion for the simplification method is recommended.

A New Bias Scheduling Method for Improving Both Classification Performance and Precision on the Classification and Regression Problems (분류 및 회귀문제에서의 분류 성능과 정확도를 동시에 향상시키기 위한 새로운 바이어스 스케줄링 방법)

  • Kim Eun-Mi;Park Seong-Mi;Kim Kwang-Hee;Lee Bae-Ho
    • Journal of KIISE:Software and Applications
    • /
    • v.32 no.11
    • /
    • pp.1021-1028
    • /
    • 2005
  • The general solution for classification and regression problems can be found by matching and modifying matrices with the information in real world and then these matrices are teaming in neural networks. This paper treats primary space as a real world, and dual space that Primary space matches matrices using kernel. In practical study, there are two kinds of problems, complete system which can get an answer using inverse matrix and ill-posed system or singular system which cannot get an answer directly from inverse of the given matrix. Further more the problems are often given by the latter condition; therefore, it is necessary to find regularization parameter to change ill-posed or singular problems into complete system. This paper compares each performance under both classification and regression problems among GCV, L-Curve, which are well known for getting regularization parameter, and kernel methods. Both GCV and L-Curve have excellent performance to get regularization parameters, and the performances are similar although they show little bit different results from the different condition of problems. However, these methods are two-step solution because both have to calculate the regularization parameters to solve given problems, and then those problems can be applied to other solving methods. Compared with UV and L-Curve, kernel methods are one-step solution which is simultaneously teaming a regularization parameter within the teaming process of pattern weights. This paper also suggests dynamic momentum which is leaning under the limited proportional condition between learning epoch and the performance of given problems to increase performance and precision for regularization. Finally, this paper shows the results that suggested solution can get better or equivalent results compared with GCV and L-Curve through the experiments using Iris data which are used to consider standard data in classification, Gaussian data which are typical data for singular system, and Shaw data which is an one-dimension image restoration problems.

Continuous Query Processing in Data Streams Using Duality of Data and Queries (데이타와 질의의 이원성을 이용한 데이타스트림에서의 연속질의 처리)

  • Lim Hyo-Sang;Lee Jae-Gil;Lee Min-Jae;Whang Kyu-Young
    • Journal of KIISE:Databases
    • /
    • v.33 no.3
    • /
    • pp.310-326
    • /
    • 2006
  • In this paper, we deal with a method of efficiently processing continuous queries in a data stream environment. We classify previous query processing methods into two dual categories - data-initiative and query-initiative - depending on whether query processing is initiated by selecting a data element or a query. This classification stems from the fact that data and queries have been treated asymmetrically. For processing continuous queries, only data-initiative methods have traditionally been employed, and thus, the performance gain that could be obtained by query-initiative methods has been overlooked. To solve this problem, we focus on an observation that data and queries can be treated symmetrically. In this paper, we propose the duality model of data and queries and, based on this model, present a new viewpoint of transforming the continuous query processing problem to a multi-dimensional spatial join problem. We also present a continuous query processing algorithm based on spatial join, named Spatial Join CQ. Spatial Join CQ processes continuous queries by finding the pairs of overlapping regions from a set of data elements and a set of queries defined as regions in the multi-dimensional space. The algorithm achieves the effects of both of the two dual methods by using the spatial join, which is a symmetric operation. Experimental results show that the proposed algorithm outperforms earlier methods by up to 36 times for simple selection continuous queries and by up to 7 times for sliding window join continuous queries.