An improved KNN text classification algorithm based on density

Shi, Kansheng; Li, Lemin; Liu, Haitao; He, Jie; Zhang, Naitong; Song, Wentao

doi:10.1109/ccis.2011.6045043

Cited by 35 publications

(21 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this method, although the outlier or the effect of noises is reduced, the problems arising from extreme closeness of the test element to any training element were not avoided completely. Because this method is very sensitive to the selection of neighbour number [7].…”

Section: Introductionmentioning

confidence: 99%

Density Weighted K-Nearest Neighbors Algorithm for Outliers in the Training Set Are So Close to the Test Element

Akben¹,

Alkan²

2015

JEE

View full text Add to dashboard Cite

Abstract:In KNN (K-Nearest Neighbour) method, the distance-weighted algorithm is applied in order to reduce the effect of noisy data. However, as this weighting algorithm will be insufficient when the outliers in the training set are so close to the test element, a new weighting algorithm is required for KNN. In this study, instead of distance-weighting, a new density-weighted KNN algorithm is proposed for reducing the effect of noisy data. In the first stage of the proposed method, the coefficient of density of each element in the training set was obtained by Parzen window method. And then, the membership of each test element was determined according to the total of density coefficients (weights) of neighbours belonging to the same class. As for the last stage, the performance results of the frequently used KNN methods and the proposed method (Density-weighted KNN, Classical KNN and Distance-weighted KNN) were compared. The obtained results have shown that the proposed method is more successful by almost 1% than classical KNN method, by 9% than distance weighted KNN method. Moreover, it is more successful than other KNN methods when the test element is so close to the training set elements.

show abstract

Section: Introductionmentioning

confidence: 99%

Density Weighted K-Nearest Neighbors Algorithm for Outliers in the Training Set Are So Close to the Test Element

Akben¹,

Alkan²

2015

JEE

View full text Add to dashboard Cite

show abstract

“…7, No. 1;2014 Amato and Falchi (2013) addressed the image recognition issue by combining the KNN and local feature search method with similarity functions.…”

Section: Related Workmentioning

confidence: 99%

3N-Q: Natural Nearest Neighbor with Quality

Shu¹,

Mouhoub²,

Sadaoui³

2014

CIS

View full text Add to dashboard Cite

In this paper, a novel algorithm for enhancing the performance of classification is proposed. This new method provides rich information for clustering and outlier detection. We call it Natural Nearest Neighbor with Quality (3N-Q). Comparing to K-nearest neighbor and E-nearest neighbor, 3N-Q employs a completely different concept to find the nearest neighbors passively, which can adaptively and automatically get the K value. This value as well as distribution of neighbors and frequency of being neighbors of others offer precious foundation not only in classification but also in clustering and outlier detection. Subsequently, we propose a fitness function that reflects the quality of each training sample, retaining the good ones while eliminating the bad ones according to the quality threshold. From the experiment results we report in this paper, it is observed that 3N-Q is efficient and accurate for solving data mining problems.

show abstract

“…They are supposed to reduce the number of variables and use distribution of R and S to imitate that of X and Y. PSL uses relevance coefficient as in (14) and is optimized by (15). Due to the existence of large quantity of complex numbers, coordinate values in each dimension of the subspace are converted to their polar form:…”

Section: Cca Representationmentioning

confidence: 99%

“…Based on previous research [12,13,14,15,16,17,18], a novel personalized subject learning (PSL) system is created based on the above ideas. PSL system is a computer aided education system using TD technologies and CCA methodology.…”

Section: Introductionmentioning

confidence: 99%

Personalized Subject Learning Based on Topic Detection and Canonical Correlation Analysis

Shi¹,

Shi²,

Shi³

2015

ijacsa

View full text Add to dashboard Cite

Abstract-To keep pace with the time, learning from printed medium alone is no longer a comprehensive approach. Fresh digital contents can definitely be the complement of printed education medium. Although timely access to fresh contents is becoming increasingly important for education and gaining such access is no longer a problem, the capacity for human teachers to assimilate such huge amounts of contents is limited. Topic Detection (TD) is then a promising research area that addresses speedy access of desired contents based on topic or subject. On the other hand, personalized education is getting more attention because it facilitates the improvement of creativity and subject learning of the students. This paper reveals a patented Personalized Subject Learning (PSL) system that caters for the need of personalized education and efficiently provides subject based contents. An efficient topic detection algorithm for providing subject content is presented. Moreover, since education contents are multimedia based ones with multimodal, PSL introduces Canonical Correlation Analysis (CCA) method to detect multimodal correlations across different types of media. Due to its novelty, PSL has been used as the key engine in a real world application of personalized education system as the smart education module sponsored by a Smart City project.

show abstract

An improved KNN text classification algorithm based on density

Cited by 35 publications

References 5 publications

Density Weighted K-Nearest Neighbors Algorithm for Outliers in the Training Set Are So Close to the Test Element

Density Weighted K-Nearest Neighbors Algorithm for Outliers in the Training Set Are So Close to the Test Element

3N-Q: Natural Nearest Neighbor with Quality

Personalized Subject Learning Based on Topic Detection and Canonical Correlation Analysis

Contact Info

Product

Resources

About