• Journal of Intelligence and Information Systems
• /
• v.14 no.2
• /
• pp.193-206
• /
• 2008

Medical diagnosis can be considered a classification task which classifies disease types from patient's condition data represented by a set of pre-defined attributes. This study proposes a hybrid genetic algorithm based classification method to develop classifiers for multidimensional pattern classification problems related with medical decision making. The classification problem can be solved by identifying separation boundaries which distinguish the various classes in the data pattern. The proposed method fits a finite number of regional agents to the data pattern by combining genetic algorithms and local adaptive operations. The local adaptive operations of an agent include expansion, avoidance and relocation, one of which is performed according to the agent's fitness value. The classifier system has been tested with well-known medical data sets from the UCI machine learning database, showing superior performance to other methods such as the nearest neighbor, decision tree, and neural networks.

• Spatial Information Research
• /
• v.20 no.3
• /
• pp.67-81
• /
• 2012

Due to the recent advancement of cloud computing, the research on database outsourcing has been actively done. Moreover, the number of users who utilize Location-based Services(LBS) has been increasing with the development in w ireless communication technology and mobile devices. Therefore, LBS providers attempt to outsource their spatial database to service provider, in order to reduce costs for data storage and management. However, because unauthorized access to sensitive data is possible in spatial database outsourcing, it is necessary to study on the preservation of a user's privacy. Thus, we, in this paper, propose a spatial data encryption scheme to produce outsourced database from an original database. We also propose a k-Nearest Neighbor(k-NN) query processing algorithm that efficiently performs k-NN by using the outsourced database. Finally, we show from performance analysis that our algorithm outperforms the existing one.

• ETRI Journal
• /
• v.43 no.6
• /
• pp.966-977
• /
• 2021

Filterbank multicarrier with offset quadrature amplitude modulation (FBMC-OQAM) is an attractive alternative to the orthogonal frequency division multiplexing (OFDM) modulation technique. In comparison with OFDM, the FBMC-OQAM signal has better spectral confinement and higher spectral efficiency and tolerance to synchronization errors, primarily due to per-subcarrier filtering using a frequency-time localized prototype filter. However, the filtering process introduces intrinsic interference among the symbols and complicates channel estimation (CE). An efficient way to improve the CE in FBMC-OQAM is using a technique known as windowed frequency domain averaging (FDA); however, it requires a priori knowledge of the window length parameter which is set based on the channel's frequency selectivity (FS). As the channel's FS is not fixed and not a priori known, we propose a k-nearest neighbor-based machine learning algorithm to classify the FS and decide on the FDA's window length. A comparative theoretical analysis of the mean-squared error (MSE) is performed to prove the proposed CE scheme's effectiveness, validated through extensive simulations. The adaptive CE scheme is shown to yield a reduction in CE-MSE and improved bit error rates compared with the popular preamble-based CE schemes for FBMC-OQAM, without a priori knowledge of channel's frequency selectivity.

• The Korean Journal of Applied Statistics
• /
• v.37 no.4
• /
• pp.395-410
• /
• 2024

Text data is usually made up of a wide variety of unique words. Even in standard text data, it is common to find tens of thousands of different words. In text data analysis, usually, each unique word is treated as a variable. Thus, text data can be regarded as a dataset with a large number of variables. On the other hand, in text data classification, we often encounter class label imbalance problems. In the cases of substantial imbalances, the performance of conventional classification models can be severely degraded. To improve the classification performance of support vector machines (SVM) for imbalanced data, algorithms such as the Synthetic Minority Over-sampling Technique (SMOTE) can be used. The SMOTE algorithm synthetically generates new observations for the minority class based on the k-Nearest Neighbors (kNN) algorithm. However, in datasets with a large number of variables, such as text data, errors may accumulate. This can potentially impact the performance of the kNN algorithm. In this study, we propose a method for enhancing prediction performance for the minority class of imbalanced text data. Our approach involves employing variable selection to generate new synthetic observations in a reduced space, thereby improving the overall classification performance of SVM.

• Journal of the Semiconductor & Display Technology
• /
• v.12 no.3
• /
• pp.53-57
• /
• 2013

Human emotions are expressed in various ways. It can be expressed through language, facial expression and gestures. In particular, the facial expression contains many information about human emotion. These vague human emotion appear not in single emotion, but in combination of various emotion. This paper proposes a emotional expression algorithm using Active Appearance Model(AAM) and Fuzz k- Nearest Neighbor which give facial expression in similar with vague human emotion. Applying Mahalanobis distance on the center class, determine inclusion level between center class and each class. Also following inclusion level, appear intensity of emotion. Our emotion recognition system can recognize a complex emotion using Fuzzy k-NN classifier.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.4
• /
• pp.670-691
• /
• 2013

In RSSI-based RFID(Radio Frequency IDentification) indoor localization system, the signal path loss model of each sub-region is different from others in the whole localization area due to the influence of the multi-path phenomenon and other environmental factors. Therefore, this paper divides the localization area into many sub-regions and constructs separately the signal path loss model of each sub-region. Then an improved LANDMARC method is proposed. Firstly, the deployment principle of RFID readers and tags is presented for constructing localization sub-region. Secondly, the virtual reference tags are introduced to create a virtual signal strength space with RFID readers and real reference tags in every sub-region. Lastly, k nearest neighbor (KNN) algorithm is used to locate the target object and an error compensating algorithm is proposed for correcting localization result. The results in real application show that the new method enhances the positioning accuracy to 18.2% and reduces the time cost to 30% of the original LANDMARC method without additional tags and readers.

• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.5
• /
• pp.327-330
• /
• 2016

Currently, the world's elderly population continues to grow at a dramatic rate. As the number of senior citizens increases, detection of someone falling has attracted increasing attention for visual surveillance systems. This paper presents a novel fall-detection algorithm using motion estimation and an integrated spatiotemporal energy map of the object region. The proposed method first extracts a human region using a background subtraction method. Next, we applied an optical flow algorithm to estimate motion vectors, and an energy map is generated by accumulating the detected human region for a certain period of time. We can then detect a fall using k-nearest neighbor (kNN) classification with the previously estimated motion information and energy map. The experimental results show that the proposed algorithm can effectively detect someone falling in any direction, including at an angle parallel to the camera's optical axis.

• Journal of Intelligence and Information Systems
• /
• v.15 no.2
• /
• pp.1-13
• /
• 2009

Scene classification and concept-based procedures have been the great interest for image categorization applications for large database. Knowing the category to which scene belongs, we can filter out uninterested images when we try to search a specific scene category such as beach, mountain, forest and field from database. In this paper, we propose an adaptive segmentation method for real-world natural scene classification based on a semantic modeling. Semantic modeling stands for the classification of sub-regions into semantic concepts such as grass, water and sky. Our adaptive segmentation method utilizes the edge detection to split an image into sub-regions. Frequency of occurrences of these semantic concepts represents the information of the image and classifies it to the scene categories. K-Nearest Neighbor (k-NN) algorithm is also applied as a classifier. The empirical results demonstrate that the proposed adaptive segmentation method outperforms the Vogel and Schiele's method in terms of accuracy.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.04a
• /
• pp.947-949
• /
• 2015

본 논문은 기존의 시간적으로 이웃한 프레임 사이의 움직임을 추정 보상하여 새로운 프레임을 생성하는 프레임률 향상 기법 (frame rate up conversion)을 제안한다. 움직임 추정(Motion Estimation)을 통하여 계산된 움직임 벡터를 이용하여 프레임을 생성하며, 생성된 프레임에서 발생되는 구명 (hole)과 중첩 (overlap) 영역을 처리하는 기법을 제안한다. 특히 k-NN 보간법(k-nearest neighbor interpolation)[3]과 중간값을 적응적으로 활용하여 향상된 화질의 영상을 생성한다. 실험 결과를 통하여 제안하는 기술의 우수성을 입증하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1520-1522
• /
• 2015

본 논문은 기존의 시간적으로 이웃한 프레임 사이의 움직임을 추정 보상하여 새로운 프레임을 생성하는 프레임률 향상 기법 (frame rate up conversion)을 제안한다. 움직임 추정(Motion Estimation)을 통하여 계산된 움직임 벡터를 이용하여 프레임을 생성하며, 생성된 프레임에서 발생되는 구멍 (hole)과 중첩 (overlap) 영역을 처리하는 기법을 제안한다. 특히 k-NN 보간법(k-nearest neighbor interpolation)[3]과 중간값을 적응적으로 활용하여 향상된 화질의 영상을 생성한다. 실험 결과를 통하여 제안하는 기술의 우수성을 입증하였다.