• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.3
• /
• pp.183-192
• /
• 2017

k-nearest neighbor (K-NN) is a well-known classification algorithm, being feature space-based on nearest-neighbor training examples in machine learning. However, K-NN, as we know, is a lazy learning method. Therefore, if a K-NN-based system very much depends on a huge amount of history data to achieve an accurate prediction result for a particular task, it gradually faces a processing-time performance-degradation problem. We have noticed that many researchers usually contemplate only classification accuracy. But estimation speed also plays an essential role in real-time prediction systems. To compensate for this weakness, this paper proposes correlation coefficient-based clustering (CCC) aimed at upgrading the performance of K-NN by leveraging processing-time speed and plurality rule-based density (PRD) to improve estimation accuracy. For experiments, we used real datasets (on breast cancer, breast tissue, heart, and the iris) from the University of California, Irvine (UCI) machine learning repository. Moreover, real traffic data collected from Ojana Junction, Route 58, Okinawa, Japan, was also utilized to lay bare the efficiency of this method. By using these datasets, we proved better processing-time performance with the new approach by comparing it with classical K-NN. Besides, via experiments on real-world datasets, we compared the prediction accuracy of our approach with density peaks clustering based on K-NN and principal component analysis (DPC-KNN-PCA).

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2573-2575
• /
• 2003

In this paper, we discuss a fuzzy neural network classifier with immune algorithm. The fuzzy neural network classifier is constructed with the fuzzy classifier and the neural network classifier based on fuzzy rules. To maximize performance of classifier, the immune algorithm and the back propagation algorithm are used. For the generalized classification ability, the simulation results from the iris data demonstrate superiority of the proposed classifier in comparison with other classifier.

• Korean Journal of Environmental Biology
• /
• v.19 no.2
• /
• pp.113-118
• /
• 2001

This study was carried out to provide basic data for establishment of Natural Ecology Center through vegetation research in Nodong valley, Pyungchanggun of Kangwondo. The vascular plants consisted of total 178 taxa such as 56 families, 121 genera, 152 species, 24 varieties and two forms. In the study area, we found some plants designated as protect plants by the Ministry of Environment such as Dryopteris crassirhizoma and Rodgersia podophylla that is a vulnerable species and Iris odaesanensis that is an endemic species. Communities by characteristic in species composition of Nodong valley forest were classified into Quercus mongolica - Acer pseudosieboldianum community as a natural forest community and Larix leptolepis and Pinus koraiensis community as artificial forest. The Quercus mongolica-Acer pseudosieboldianum community was further classified as Pinus densiflora subcommunity and Acer mono subcommunity. A protection program will be needed for Iris odaesanensis classified as an endemic species to make a Natural Ecology Center.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.800-804
• /
• 2005

This paper presents a fuzzy learning rule which is the fuzzified version of LVQ(Learning Vector Quantization). This fuzzy learning rule 3 uses fuzzy learning rates. instead of the traditional learning rates. LVQ uses the same learning rate regardless of correctness of classification. But, the new fuzzy learning rule uses the different learning rates depending on whether classification is correct or not. The new fuzzy learning rule is integrated into the improved IAFC(Integrated Adaptive Fuzzy Clustering) neural network. The improved IAFC neural network is both stable and plastic. The iris data set is used to compare the performance of the supervised IAFC neural network 3 with the performance of backprogation neural network. The results show that the supervised IAFC neural network 3 is better than backpropagation neural network.

• Journal of KIISE:Software and Applications
• /
• v.26 no.11
• /
• pp.1314-1323
• /
• 1999

컴퓨터의 사용이 일반화됨에 따라 데이타를 생성하고 수집하는 것이 용이해졌다. 이에 따라 데이타로부터 자동적으로 유용한 지식을 얻는 기술이 필요하게 되었다. 데이타 마이닝에서 얻어진 지식은 정확성과 이해성을 충족해야 한다. 본 논문에서는 데이타 마이닝을 위하여 퍼지 결정트리에 기반한 효율적인 퍼지 규칙을 생성하는 알고리즘을 제안한다. 퍼지 결정트리는 ID3와 C4.5의 이해성과 퍼지이론의 추론과 표현력을 결합한 방법이다. 특히, 퍼지 규칙은 속성 축에 평행하게 판단 경계선을 결정하는 방법으로는 어려운 속성 축에 평행하지 않는 경계선을 갖는 패턴을 효율적으로 분류한다. 제안된 알고리즘은 첫째, 각 속성 데이타의 히스토그램 분석을 통해 적절한 소속함수를 생성한다. 둘째, 주어진 소속함수를 바탕으로 ID3와 C4.5와 유사한 방법으로 퍼지 결정트리를 생성한다. 또한, 유전자 알고리즘을 이용하여 소속함수를 조율한다. IRIS 데이타, Wisconsin breast cancer 데이타, credit screening 데이타 등 벤치마크 데이타들에 대한 실험 결과 제안된 방법이 C4.5 방법을 포함한 다른 방법보다 성능과 규칙의 이해성에서 보다 효율적임을 보인다.Abstract With an extended use of computers, we can easily generate and collect data. There is a need to acquire useful knowledge from data automatically. In data mining the acquired knowledge needs to be both accurate and comprehensible. In this paper, we propose an efficient fuzzy rule generation algorithm based on fuzzy decision tree for data mining. We combine the comprehensibility of rules generated based on decision tree such as ID3 and C4.5 and the expressive power of fuzzy sets. Particularly, fuzzy rules allow us to effectively classify patterns of non-axis-parallel decision boundaries, which are difficult to do using attribute-based classification methods.In our algorithm we first determine an appropriate set of membership functions for each attribute of data using histogram analysis. Given a set of membership functions then we construct a fuzzy decision tree in a similar way to that of ID3 and C4.5. We also apply genetic algorithm to tune the initial set of membership functions. We have experimented our algorithm with several benchmark data sets including the IRIS data, the Wisconsin breast cancer data, and the credit screening data. The experiment results show that our method is more efficient in performance and comprehensibility of rules compared with other methods including C4.5.

• Korean Journal of Artificial Intelligence
• /
• v.11 no.4
• /
• pp.29-34
• /
• 2023

The flow from developing a machine learning model to deploying it in a production environment suffers challenges. Efficient and reliable deployment is critical for realizing the true value of machine learning models. Bridging this gap between development and publication has become a pivotal concern in the machine learning community. FastAPI, a modern and fast web framework for building APIs with Python, has gained substantial popularity for its speed, ease of use, and asynchronous capabilities. This paper focused on leveraging FastAPI for deploying machine learning models, addressing the potentials associated with integration, scalability, and performance in a production setting. In this work, we explored the seamless integration of machine learning models into FastAPI applications, enabling real-time predictions and showing a possibility of scaling up for a more diverse range of use cases. We discussed the intricacies of integrating popular machine learning frameworks with FastAPI, ensuring smooth interactions between data processing, model inference, and API responses. This study focused on elucidating the integration of machine learning models into production environments using FastAPI, exploring its capabilities, features, and best practices. We delved into the potential of FastAPI in providing a robust and efficient solution for deploying machine learning systems, handling real-time predictions, managing input/output data, and ensuring optimal performance and reliability.

• Journal of KIISE:Software and Applications
• /
• v.33 no.8
• /
• pp.692-702
• /
• 2006

In this study, a weighted fuzzy min-max (WFMM) neural network model for pattern classification is proposed. The model has a modified structure of FMM neural network in which the weight concept is added to represent the frequency factor of feature values in a learning data set. First we present in this paper a new activation function of the network which is defined as a hyperbox membership function. Then we introduce a new learning algorithm for the model that consists of three kinds of processes: hyperbox creation/expansion, hyperbox overlap test, and hyperbox contraction. A weight adaptation rule considering the frequency factors is defined for the learning process. Finally we describe a feature analysis technique using the proposed model. Four kinds of relevance factors among feature values, feature types, hyperboxes and patterns classes are proposed to analyze relative importance of each feature in a given problem. Two types of practical applications, Fisher's Iris data and Cleveland medical data, have been used for the experiments. Through the experimental results, the effectiveness of the proposed method is discussed.

• 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.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.4
• /
• pp.692-700
• /
• 2017

In this paper, we propose the structural design of Interval Type-2 FCM based RBFNN. Proposed model consists of three modules such as condition, conclusion and inference parts. In the condition part, Interval Type-2 FCM clustering which is extended from FCM clustering is used. In the conclusion part, the parameter coefficients of the consequence part are estimated through LSE(Least Square Estimation) and WLSE(Weighted Least Square Estimation). In the inference part, final model outputs are acquired by fuzzy inference method from linear combination of both polynomial and activation level obtained through Interval Type-2 FCM and acquired activation level through Interval Type-2 FCM. Additionally, The several parameters for the proposed model are identified by using differential evolution. Final model outputs obtained through benchmark data are shown and also compared with other already studied models' performance. The proposed algorithm is performed by using Iris and Vehicle data for pattern classification. For the validation of regression problem modeling performance, modeling experiments are carried out by using MPG and Boston Housing data.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.20 no.6
• /
• pp.161-180
• /
• 2017

As the competition for securing cross-border biological sovereignty becomes intensified due to the adoption of the Nagoya Protocol, this study analyzed patent trends only for freshwater plants in order to secure national biological sovereignty. As a result, freshwater plants include a total of 68 genera and 128 species, and a total of 60 genera and 3,256 patents were surveyed. Among them, iris was the most industrialized, 14.71% followed by angelica(8.48%) and Mentha(6.94%). However, unconfirmed eight genera (Aneilema, Artemisia Cabomba, Nymphoides, Pistia stratiotes L., Pseudoraphis Griff., Ruppia) are not patented freshwater plants and it is expected that patent entry is high and barrier is low in the future. Based on patent results, Cooperative Patent Classification analysis was carried out and as a result, a total of 15 industry sectors were derived. And biopharmaceutical(30.24%) was found to be the most industrialized industry sector followed by agricultural chemistry (28.89%), biochemical industry (16.25%). In the biomedical industry, angelica(17.74%) was the most used and Iris (9.55%), Sium(20.56%) and angelica (20.48%) were found to be the most used in agricultural chemistry, biochemical industry and bio food industry, respectively. The analysis of detailed industry fields for 15 industry sectors showed that medicines of unknown structure containing substances from plants (37.77%), raw materials (46.57%) such as insect repellants, attracting agents and preparation of peptides(16.82%) with more than 21 amino acids were most frequently used in biopharmaceutical, agricultural chemistry and biochemical industry, respectively. This study is of significance as a basic data to know which freshwater plants are used in which field in order to secure biological sovereignty and patent analysis is considered necessary to continuously secure the biological sovereignty for freshwater plants.