Location difference of multiple distances based k-nearest neighbors algorithm

Xia, Shuyin; Xiong, Zenggang; Luo, Yuling; Dong, Limei; Zhang, Guanghua

doi:10.1016/j.knosys.2015.09.028

Cited by 38 publications

(25 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…18, the proposed method introduces a new measurement to measure the difference of location between different points and only needs n (i.e. the number of data points) times to assign the value of the measurement to each point.…”

Section: Location Difference Of Double Distances Based Neighbors Algomentioning

confidence: 99%

“…Therefore, time complexity of step 4 in Algorithm 1 is smaller than the case of step 2 when ε is set as: log log m m n (18) Time complexity of Algorithm 1 is determined by step 2 and can be controlled to be O(log2 dn log2n) ~ O(log2 dn). The smaller the m, will result in larger value of Expression (18) and the length of the range in Algorithm 1; thus, the m can be set at 2.…”

Section: Value Of Parameter εmentioning

confidence: 99%

“…The smaller the m, will result in larger value of Expression (18) and the length of the range in Algorithm 1; thus, the m can be set at 2.…”

Section: Value Of Parameter εmentioning

confidence: 99%

“…The time complexity of original model is only quadratic. Moreover, by introducing the ideal of location difference, 18 our accelerated algorithm can have a time complexity of O(n log n) and not rely on any tree structure, which makes it more efficient than most other algorithms especially on high dimensional datasets. …”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Vibration-Based Outlier Detection on High Dimensional Data

Xia

Wang

et al. 2016

Int. J. Artif. Intell. Tools

Self Cite

View full text Add to dashboard Cite

Outlier detection refers to find patterns that do not fit in to normal behaviour. Outlier detection plays an important role in data mining. Most of real world datasets are using outlier detection. Outlier detection is useful in many fields like Network intrusion, Credit card fraud detection, stock market, and wireless sensor network data. The distance based outlier detection techniques be unsuccessful to will increases the dimensionality of data because in high dimensionality distance between the two points is less. In existing system using Anti-hub method in reverse nearest neighbors. Anti-hubs are few points are regularly comes in K-nearest neighbors list of another points and few points are an irregularly comes in k-nearest neighbors list of different points. In Anti-hub method propose high computational cost and time requirements for finding outliers. To overcome this problem we can use new method in this paper that is the advanced variety of Anti-hub is Anti-hub2.which is for reconsider the outlier score of a data point obtained by the Anti-hub method. The goal of this paper is locate the inconsistent objects in data which has high dimension through reduced computation time, cost and increase the accuracy. We apply logistic regression rule on the results of Anti-hub dataset then obtained combination of data, prevention measures and Anti-hub calculation. It increase the efficiency of remove out irrelevant, redundant feature. I. INTRODUCTION Data Mining means the process of extracting the knowledge for huge data sources. The general objective of the data mining process is to extract information from a data set and transform it into a reasonable structure for further utilize. In the Data Mining have four techniques that are Clustering, Classification, Association, and Outliers. In this paper we discuss the Outlier detection techniques definition of Outlier location (otherwise called inconsistency discovery) is the way toward discovering data objects with practices that are altogether different from desire. Such objects are called outliers or anomalies. Outlier detection is important in many applications such as medical care, public safety and security, industry damage detection, image processing, sensor/video network surveillance, and intrusion detection. In general, outliers is classified into 3 varieties, those are global outliers, conditional outliers, and collective outliers. To discover global outliers, an essential issue is to search for out an acceptable mensuration of deviation with relation to the appliance in question. Global outlier detection is very significant in several applications. Take into account intrusion detection in laptop networks, as an example. If the statement behaviour of a laptop is extremely completely changed from the conventional designs (e.g., an oversized range of packages is broadcast in a very short time), this behaviour is also thought-about as a worldwide outlier and also the consistent laptop could be a suspected target of hacking. Next one is conditional outlier the item ...

show abstract

Section: Location Difference Of Double Distances Based Neighbors Algomentioning

confidence: 99%

Section: Value Of Parameter εmentioning

confidence: 99%

“…The smaller the m, will result in larger value of Expression (18) and the length of the range in Algorithm 1; thus, the m can be set at 2.…”

Section: Value Of Parameter εmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Vibration-Based Outlier Detection on High Dimensional Data

Xia

Wang

et al. 2016

Int. J. Artif. Intell. Tools

Self Cite

View full text Add to dashboard Cite

show abstract

“…Conventionally, VQ conducts a full search to ensure that a codeword is best matched with an arbitrary input vector, but the full search requires an enormous computational load. Thus, a continuous effort has been made to simplify the search complexity of an encoding process in a great volume of published studies [14][15][16][17][18][19][20][21][22][23][24][25]. These approaches are further classified into three types in terms of the way the complexity is simplified, including the tree-structured VQ (TSVQ) techniques [14][15][16], the TIE-based approaches [18][19][20] and the equal-average equal-variance equal-norm nearest neighbor search (EEENNS) based algorithms [21][22][23][24][25].…”

Section: Introductionmentioning

confidence: 99%

A Search Complexity Improvement of Vector Quantization to Immittance Spectral Frequency Coefficients in AMR-WB Speech Codec

2016

View full text Add to dashboard Cite

An adaptive multi-rate wideband (AMR-WB) code is a speech codec developed on the basis of an algebraic code-excited linear-prediction (ACELP) coding technique, and has a double advantage of low bit rates and high speech quality. This coding technique is widely used in modern mobile communication systems for a high speech quality in handheld devices. However, a major disadvantage is that a vector quantization (VQ) of immittance spectral frequency (ISF) coefficients occupies a significant computational load in the AMR-WB encoder. Hence, this paper presents a triangular inequality elimination (TIE) algorithm combined with a dynamic mechanism and an intersection mechanism, abbreviated as the DI-TIE algorithm, to remarkably improve the complexity of ISF coefficient quantization in the AMR-WB speech codec. Both mechanisms are designed in a way that recursively enhances the performance of the TIE algorithm. At the end of this work, this proposal is experimentally validated as a superior search algorithm relative to a conventional TIE, a multiple TIE (MTIE), and an equal-average equal-variance equal-norm nearest neighbor search (EEENNS) approach. With a full search algorithm as a benchmark for search load comparison, this work provides a search load reduction above 77%, a figure far beyond 36% in the TIE, 49% in the MTIE, and 68% in the EEENNS approach.

show abstract

Iterative triangular inequality elimination algorithm for codebook search of vector quantization

Yeh

2018

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

This paper presents an iterative triangular inequality elimination (ITIE) algorithm for vector quantization (VQ) codebook search, which is then applied to the quantization of ISF coefficients in an AMR-WB+ audio codec. Most of the codewords are ruled out as candidates using the ITIE rejection criterion, leading to considerable computational complexity reduction but with a wellmaintained coding quality. With a full search algorithm as the benchmark, this work gives a reduction in the total number of operations above 74%, a figure far beyond the 60% using the TIE algorithm with a dynamic and an intersection mechanism (DI-TIE) and 65% using the equal-average equal-variance equal-norm nearest neighbor search (EEENNS) algorithm. Furthermore, this work is expected to meet the energy-saving requirement when implemented on an AMR-WB+ audio codec of mobile devices.

show abstract

Location difference of multiple distances based k-nearest neighbors algorithm

Cited by 38 publications

References 24 publications

Vibration-Based Outlier Detection on High Dimensional Data

Vibration-Based Outlier Detection on High Dimensional Data

A Search Complexity Improvement of Vector Quantization to Immittance Spectral Frequency Coefficients in AMR-WB Speech Codec

Iterative triangular inequality elimination algorithm for codebook search of vector quantization

Contact Info

Product

Resources

About