• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.2
• /
• pp.742-762
• /
• 2015

In this study, we propose a distance metric learning approach called discriminant metric learning (DML) for face verification, which addresses a binary-class problem for classifying whether or not two input images are of the same subject. The critical issue for solving this problem is determining the method to be used for measuring the distance between two images. Among various methods, the large margin nearest neighbor (LMNN) method is a state-of-the-art algorithm. However, to compensate the LMNN's entangled data distribution due to high levels of appearance variations in unconstrained environments, DML's goal is to penalize violations of the negative pair distance relationship, i.e., the images with different labels, while being integrated with LMNN to model the distance relation between positive pairs, i.e., the images with the same label. The likelihoods of the input images, estimated using DML and LMNN metrics, are then weighted and combined for further analysis. Additionally, rather than using the k-nearest neighbor (k-NN) classification mechanism, we propose a verification mechanism that measures the correlation of the class label distribution of neighbors to reduce the false negative rate of positive pairs. From the experimental results, we see that DML can modify the relation of negative pairs in the original LMNN space and compensate for LMNN's performance on faces with large variances, such as pose and expression.

• Journal of KIISE:Software and Applications
• /
• v.31 no.6
• /
• pp.784-791
• /
• 2004

In this study, we deal with face recognition using fuzzy-based Fisherface method. The well-known Fisherface method is more insensitive to large variation in light direction, face pose, and facial expression than Principal Component Analysis method. Usually, the various methods of face recognition including Fisherface method give equal importance in determining the face to be recognized, regardless of typicalness. The main point here is that the proposed method assigns a feature vector transformed by PCA to fuzzy membership rather than assigning the vector to particular class. In this method, fuzzy membership degrees are obtained from FKNN(Fuzzy K-Nearest Neighbor) initialization. Experimental results show better recognition performance than other methods for ORL and Yale face databases.

• Asia pacific journal of information systems
• /
• v.16 no.1
• /
• pp.71-84
• /
• 2006

Case-based reasoning (CBR) is a reasoning technique that reuses past cases to find a solution to the new problem. It often shows significant promise for improving effectiveness of complex and unstructured decision making. It has been applied to various problem-solving areas including manufacturing, finance and marketing for the reason. However, the design of appropriate case indexing and retrieval mechanisms to improve the performance of CBR is still a challenging issue. Most of the previous studies on CBR have focused on the similarity function or optimization of case features and their weights. According to some of the prior research, however, finding the optimal k parameter for the k-nearest neighbor (k-NN) is also crucial for improving the performance of the CBR system. In spite of the fact, there have been few attempts to optimize the number of neighbors, especially using artificial intelligence (AI) techniques. In this study, we introduce a genetic algorithm (GA) to optimize the number of neighbors to combine. This study applies the novel approach to Korean stock market. Experimental results show that the GA-optimized k-NN approach outperforms other AI techniques for stock market prediction.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.6
• /
• pp.11-17
• /
• 2017

We proposed malware detection method, which use the feature vector that consist of Opcode(operation code) and Windows API Calls extracted from executable files. And, we implemented our feature vector and measured the performance of it by using Bernoulli Naïve Bayes and K-Nearest Neighbor classifier. In experimental result, when using the K-NN classifier with the proposed method, we obtain 95.21% malware detection accuracy. It was better than existing methods using only either Opcode or Windows API Calls.

• Journal of Korea Multimedia Society
• /
• v.15 no.9
• /
• pp.1112-1125
• /
• 2012

To realize LBS (Location Based Service), typically GPS is mostly used. However, this system can be only used in out-sides. Furthermore, the use of the GPS in sensor networks is not efficient due to the low power consumption. Hence, we propose methods for the location positioning which is runnable at indoor in this paper. The proposed methods elaborate the location positioning system via applying K-NN(K-Nearest Neighbour) Algorithm with its intermediate values based on IEEE 802.15.4 technology; which is mostly used for the sensor networks. Logically the accuracy of the location positioning is proportional to the number of sampling sensor nodes' RSS according to the K-NN. By the way, numerous sampling uses a lot of sensor networks' resources. In order to reduce the number of samplings, we, instead, attempt to use the intermediate values of K-NN's signal boundaries, so that our proposed methods are able to positioning almost two times as accurate as the general ways of K-NN's result.

• Composites Research
• /
• v.33 no.3
• /
• pp.108-114
• /
• 2020

Piecewise Integrated Composite (PIC) beam is composed of different stacking against loading type depending upon location. The aim of current study is to assign robust stacking sequences against external loading to every corresponding part of the PIC beam based on the value of stress triaxiality at generated reference points using the k-NN (k-Nearest Neighbor) classification, which is one of representative machine learning techniques, in order to excellent superior bending characteristics. The stress triaxiality at reference points is obtained by three-point bending analysis of the Al beam with training data categorizing the type of external loading, i.e., tension, compression or shear. Loading types of each plane of the beam were classified by independent plane scheme as well as total beam scheme. Also, loading fidelities were calibrated for each case with the variation of hyper-parameters. Most effective stacking sequences were mapped into the PIC beam based on the k-NN classification model with the highest loading fidelity. FE analysis result shows the PIC beam has superior external loading resistance and energy absorption compared to conventional beam.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2005.11a
• /
• pp.588-594
• /
• 2005

사례기반추론(Case-Based Reasoning)은 다양한 예측 문제에 있어서 성공적으로 활용되고 있는 데이터마이닝 기법 중 하나이다. 사례기반추론 시스템의 예측 성능은 예측에 사용되는 최근접이웃(Nearest Neighbor)을 어떻게 설정하느냐에 따라 영향을 받게 된다. 따라서 최근접 이웃을 결정짓는 k 값의 설정은 성공적인 사례기반추론 시스템을 구축하기 위한 중요 요인 중 하나가 된다. 최근접 이웃의 설정에 있어서 대부분의 선행 연구들은 고정된 k 값을 사용하는 사례기반추론 시스템은 k 값을 크게 설정할 경우 최근접 이웃 안에 주어진 오류를 일으킬 수 있으며, k 값이 작게 설정된 경우에는 유사 사례 중 일부만을 예측에 사용하기 때문에 예측 결과의 왜곡을 초래할 수 있다. 본 이웃을 결정함에 있어서 Similarity Threshold를 이용하는 s-NN 방법을 제안하였다. 본 연구의 실험을 위해 UCI(University of california, Irvine) Machine Learning Repository에서 제공하는 두 개의 신용 데이터 셋을 사용하였으며, 실험 결과 s-NN 적용한 CBR 모델이 고정된 k 값을 적용한 전통적인 CBR 모델보다 더 우수한 성능을 보여주었다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.1
• /
• pp.126-132
• /
• 2011

In the application areas of LBS(Location Based Service) and ITS(Intelligent Transportation System), continuous k-nearest neighbor query(CkNN) which is defined as a query to find the nearest points of interest to all the points on a given path is widely used. It is necessary to acquire results quickly in the above applications and be applicable to spatial network databases. It is also able to cope successfully with frequent updates of POI objects. This paper proposes a new method to search nearest POIs for moving query objects on the spatial networks. The method produces a set of split points and their corresponding k-POIs as results with partial order among k-POIs. The results obtained from experiments with real dataset show that the proposed method outperforms the existing methods. The proposed method achieves very short processing time(15%) compared with the existing method.

• Journal of Korean Society of Forest Science
• /
• v.101 no.4
• /
• pp.547-557
• /
• 2012

This study purposed to estimate the forest biomass using k-Nearest Neighbor (k-NN) algorithm. Multiple data sources were used for the analysis such as forest type map, field survey data and Landsat TM data. The accuracy of forest biomass was evaluated with the forest stratification, horizontal reference area (HRA) and spatial filtering. Forests were divided into 3 types such as conifers, broadleaved, and Korean pine (Pinus koriansis) forests. The applied radii of HRA were 4 km, 5 km and 10 km, respectively. The estimated biomass and mean bias for conifers forest was 222 t/ha and 1.8 t/ha when the value of k=8, the radius of HRA was 4 km, and $5{\times}5$ modal was filtered. The estimated forest biomass of Korean pine was 245 t/ha when the value of k=8, the radius of HRA was 4km. The estimated mean biomass and mean bias for broadleaved forests were 251 t/ha and -1.6 t/ha, respectively, when the value of k=6, the radius of HRA was 10 km. The estimated total forest biomass by k-NN method was 799,000t and 237 t/ha. The estimated mean biomass by ${\kappa}NN$method was about 1t/ha more than that of filed survey data.

• Journal of KIISE:Databases
• /
• v.29 no.5
• /
• pp.358-366
• /
• 2002

Feature-based similarity retrieval become an important research issue in image database systems. The features of image data are useful to discrimination of images. In this paper, we propose the high speed k-Nearest Neighbor search algorithm based on Self-Organizing Maps. Self-Organizing Maps(SOM) provides a mapping from high dimensional feature vectors onto a two-dimensional space and generates a topological feature map. A topological feature map preserves the mutual relations (similarities) in feature spaces of input data, and clusters mutually similar feature vectors in a neighboring nodes. Therefore each node of the topological feature map holds a node vector and similar images that is closest to each node vector. We implemented a k-NN search for similar image classification as to (1) access to topological feature map, and (2) apply to pruning strategy of high speed search. We experiment on the performance of our algorithm using color feature vectors extracted from images. Promising results have been obtained in experiments.