• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.1
• /
• pp.101-108
• /
• 2010

In this paper, we proposed a new lazy classifier with fuzzy k-nearest neighbors approach and feature selection which is based on reconstruction error. Reconstruction error is the performance index for locally linear reconstruction. When a new query point is given, fuzzy k-nearest neighbors approach defines the local area where the local classifier is available and assigns the weighting values to the data patterns which are involved within the local area. After defining the local area and assigning the weighting value, the feature selection is carried out to reduce the dimension of the feature space. When some features are selected in terms of the reconstruction error, the local classifier which is a sort of polynomial is developed using weighted least square estimation. In addition, the experimental application covers a comparative analysis including several previously commonly encountered methods such as standard neural networks, support vector machine, linear discriminant analysis, and C4.5 trees.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.4
• /
• pp.129-137
• /
• 1994

In this paper, a new texture feature extraction method for texture image classification is proposed. The proposed method is a modified texture spectrum method. It uses local weighting averaged texture unit, that is, the neighbor pixels are weithted and averaged in 4-direction and the calculated values are compared with center pixel to find texture units. The proposed method has only 81 texture units and these units are really good features for texture classification. The proposed method is applied to vegetable images and Blodatz album images and compared with several conventional methods for the feature extraction time and the recognition rate.

• The KIPS Transactions:PartB
• /
• v.17B no.5
• /
• pp.381-388
• /
• 2010

Naive Bayesian learning has been widely used in many data mining applications, and it performs surprisingly well on many applications. However, due to the assumption that all attributes are equally important in naive Bayesian learning, the posterior probabilities estimated by naive Bayesian are sometimes poor. In this paper, we propose more fine-grained weighting methods, called value weighting, in the context of naive Bayesian learning. While the current weighting methods assign a weight to each attribute, we assign a weight to each attribute value. We investigate how the proposed value weighting effects the performance of naive Bayesian learning. We develop new methods, using gradient descent method, for both value weighting and feature weighting in the context of naive Bayesian. The performance of the proposed methods has been compared with the attribute weighting method and general Naive bayesian, and the value weighting method showed better in most cases.

• Journal of KIISE
• /
• v.43 no.11
• /
• pp.1270-1274
• /
• 2016

Logistic regression is widely used for predicting and estimating the relationship among variables. We propose a new logistic regression model, the value weighted logistic regression, which comprises of a fine-grained weighting method, and assigns adapted weights to each feature value. This gradient approach obtains the optimal weights of feature values. Experiments were conducted on several data sets from the UCI machine learning repository, and the results revealed that the proposed method achieves meaningful improvement in the prediction accuracy.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.7
• /
• pp.410-415
• /
• 2005

Recently, most of the speaker verification systems are based on the pattern recognition approach method. And performance of the pattern-classifier depends on how to classify a variety of speakers' feature parameters. In order to classify feature parameters efficiently and effectively, it is of great importance to enlarge variations between speakers and effectively measure distances between feature parameters. Therefore, this paper would suggest the positively mixed model scheme that can enlarge inter-speaker variation by searching the individual model with world model at the same time. During decision procedure, we can maximize inter-speaker variation by using the proposed mixed model scheme. We also make use of a symbol probability weighting function in this system so as to reduce vector quantization errors by measuring symbol probability derived from the distance rate of between the world codebook and individual codebook. As the result of our experiment using this method, we could halve the Detection Cost Function (DCF) of the system from $2.37\%\;to\;1.16\%$.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.199-202
• /
• 2000

본 논문에서는 퍼지-유전자 접근방법을 이용한 무감독 특징 선택방법에 대하여 나타내었다. 이 방법은 각각의 특징들의 중요도에 따라 순서를 정하기 위해 사용되는 weighted distance 를 포함하는 특징 평가 지표 (feature evaluation index)를 최소화시키는데 있다. 또한 특징 평가 지표에서 사용되는 각 패턴들의 쌍에 대하여 근접함의 정도를 퍼지 멤버쉽 함수를 이용하여 결정하고 유전자 알고리즘은 평가 지표를 최소화시킴으로써 각 특징의 중요도를 나타내는 최적의 weighting 계수의 집합을 한기 위하여 적용하였다.

• Journal of the Korean Society for information Management
• /
• v.25 no.1
• /
• pp.211-233
• /
• 2008

This study examines various weighting methods for improving the performance of automatic classification based on Rocchio algorithm on two collections(LISA, Reuters-21578). First, three factors for weighting are identified as document factor, document factor, category factor for each weighting schemes, the performance of each was investigated. Second, the performance of combined weighting methods between the single schemes were examined. As a result, for the single schemes based on each factor, category-factor-based schemes showed the best performance, document set-factor-based schemes the second, and document-factor-based schemes the worst. For the combined weighting schemes, the schemes(idf*cat) which combine document set factor with category factor show better performance than the combined schemes(tf*cat or ltf*cat) which combine document factor with category factor as well as the common schemes (tfidf or ltfidf) that combining document factor with document set factor. However, according to the results of comparing the single weighting schemes with combined weighting schemes in the view of the collections, while category-factor-based schemes(cat only) perform best on LISA, the combined schemes(idf*cat) which combine document set factor with category factor showed best performance on the Reuters-21578. Therefore for the practical application of the weighting methods, it needs careful consideration of the categories in a collection for automatic classification.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.10
• /
• pp.2477-2484
• /
• 1997

Best-first heuristic search algorithm, such as $A^{\ast}$ algorithm, are one of the most important techniques used to solve many problems in artificial intelligence. A common feature of heuristic search is its high computational complexity, which prevents the search from being applied to problems is practical domains such as route-finding in road map with significantly many nodes. In this paper, several heuristic search algorithms are concerned. A new dynamic weighting heuristic method called the pat-sensitive heuristic is proposed. It is based on a dynamic weighting heuristic, which is used to improve search effort in practical domain such as admissible heuristic is not available or heuristic accuracy is poor. It's distinctive feature compared with other dynamic weighting heuristic algorithms is path-sensitive, which means that ${\omega}$(weight) is adjusted dynamically during search process in state-space search domain. For finding an optimal path, randomly scattered road-map is used as an application area.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.1
• /
• pp.36-42
• /
• 2006

A Principal component analysis (PCA)-based Mel-Frequency Discrete Wavelet Coefficients (MFDWC) feature Parameters for speaker verification system is Presented in this Paper In this method, we used the 1st-eigenvector obtained from PCA to calculate the energy of each node of level that was approximated by. met-scale. This eigenvector satisfies the constraint of general weighting function that the squared sum of each component of weighting function is unity and is considered to represent speaker's characteristic closely because the 1st-eigenvector of each speaker is fairly different from the others. For verification. we used Universal Background Model (UBM) approach that compares claimed speaker s model with UBM on frame-level. We performed experiments to test the effectiveness of PCA-based parameter and found that our Proposed Parameters could obtain improved average Performance of $0.80\%$compared to MFCC. $5.14\%$ to LPCC and 6.69 to existing MFDWC.

• MALSORI
• /
• no.44
• /
• pp.83-92
• /
• 2002

Speech recognition performance is severly degraded under noisy envrionments. One approach to cope with this problem is audio-visual speech recognition. In this paper, we discuss the experiment results of bimodal speech recongition based on enhanced speech feature vectors using lip information. We try various kinds of speech features as like linear predicion coefficient, cepstrum, log area ratio and etc for transforming lip information into speech parameters. The experimental results show that the cepstrum parameter is the best feature in the point of reconition rate. Also, we present the desirable weighting values of audio and visual informations depending on signal-to-noiso ratio.