Hebbian Self-Organizing Integrate-and-Fire Networks for Data Clustering

We propose a Hebbian learning-based data clustering algorithm using spiking neurons. The algorithm is capable of distinguishing between clusters and noisy background data and finds an arbitrary number of clusters of arbitrary shape. These properties render the approach particularly useful for visual scene segmentation into arbitrarily shaped homogeneous regions. We present several application examples, and in order to highlight the advantages and the weaknesses of our method, we systematically compare the results with those from standard methods such as the k-means and Ward's linkage clustering. The analysis demonstrates that not only the clustering ability of the proposed algorithm is more powerful than those of the two concurrent methods, the time complexity of the method is also more modest than that of its generally used strongest competitor.

Download Full-text

Improved Bidirectional CABOSFV Based on Multi-Adjustment Clustering and Simulated Annealing

Cybernetics and Information Technologies ◽

10.1515/cait-2016-0075 ◽

2016 ◽

Vol 16 (6) ◽

pp. 27-42 ◽

Cited By ~ 1

Author(s):

Minghan Yang ◽

Xuedong Gao ◽

Ling Li

Keyword(s):

Simulated Annealing ◽

Data Clustering ◽

Time Complexity ◽

Clustering Algorithm ◽

Feature Vector ◽

Parameter Determination ◽

Data Sets ◽

Parameter Vector ◽

Clustering Validity

Abstract Although Clustering Algorithm Based on Sparse Feature Vector (CABOSFV) and its related algorithms are efficient for high dimensional sparse data clustering, there exist several imperfections. Such imperfections as subjective parameter designation and order sensibility of clustering process would eventually aggravate the time complexity and quality of the algorithm. This paper proposes a parameter adjustment method of Bidirectional CABOSFV for optimization purpose. By optimizing Parameter Vector (PV) and Parameter Selection Vector (PSV) with the objective function of clustering validity, an improved Bidirectional CABOSFV algorithm using simulated annealing is proposed, which circumvents the requirement of initial parameter determination. The experiments on UCI data sets show that the proposed algorithm, which can perform multi-adjustment clustering, has a higher accurateness than single adjustment clustering, along with a decreased time complexity through iterations.

Download Full-text

An Ordinal Data Clustering Algorithm with Automated Distance Learning

Proceedings of the AAAI Conference on Artificial Intelligence ◽

10.1609/aaai.v34i04.6168 ◽

2020 ◽

Vol 34 (04) ◽

pp. 6869-6876

Author(s):

Yiqun Zhang ◽

Yiu-ming Cheung

Keyword(s):

Machine Learning ◽

Data Mining ◽

Categorical Data ◽

Data Clustering ◽

Time Complexity ◽

Ordinal Data ◽

Clustering Algorithm ◽

Major Type ◽

Common Task ◽

Consecutive Integers

Clustering ordinal data is a common task in data mining and machine learning fields. As a major type of categorical data, ordinal data is composed of attributes with naturally ordered possible values (also called categories interchangeably in this paper). However, due to the lack of dedicated distance metric, ordinal categories are usually treated as nominal ones, or coded as consecutive integers and treated as numerical ones. Both these two common ways will roughly define the distances between ordinal categories because the former way ignores the order relationship and the latter way simply assigns identical distances to different pairs of adjacent categories that may have intrinsically unequal distances. As a result, they may produce unsatisfactory ordinal data clustering results. This paper, therefore, proposes a novel ordinal data clustering algorithm, which iteratively learns: 1) The partition of ordinal dataset, and 2) the inter-category distances. To the best of our knowledge, this is the first attempt to dynamically adjust inter-category distances during the clustering process to search for a better partition of ordinal data. The proposed algorithm features superior clustering accuracy, low time complexity, fast convergence, and is parameter-free. Extensive experiments show its efficacy.

Download Full-text

Balanced Data Clustering Algorithm for Both Hard and Soft Clustering

International Journal of Computer Sciences and Engineering ◽

10.26438/ijcse/v6i2.176183 ◽

2018 ◽

Vol 6 (2) ◽

pp. 176-183

Author(s):

Purnendu Das ◽

◽

Bishwa Ranjan Roy ◽

Saptarshi Paul ◽

◽

...

Keyword(s):

Data Clustering ◽

Clustering Algorithm ◽

Soft Clustering

Download Full-text

Tree-ART2 Learning Model for Spatial Clustering in Second Dimension

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.543-547.1934 ◽

2014 ◽

Vol 543-547 ◽

pp. 1934-1938

Author(s):

Ming Xiao

Keyword(s):

Network Model ◽

Spatial Data ◽

Data Clustering ◽

Clustering Algorithm ◽

Spatial Clustering ◽

Adaptive Resonance Theory ◽

Spatial Distance ◽

Resonance Theory ◽

Adaptive Resonance ◽

Vector Module

For a clustering algorithm in two-dimension spatial data, the Adaptive Resonance Theory exists not only the shortcomings of pattern drift and vector module of information missing, but also difficultly adapts to spatial data clustering which is irregular distribution. A Tree-ART2 network model was proposed based on the above situation. It retains the memory of old model which maintains the constraint of spatial distance by learning and adjusting LTM pattern and amplitude information of vector. Meanwhile, introducing tree structure to the model can reduce the subjective requirement of vigilance parameter and decrease the occurrence of pattern mixing. It is showed that TART2 network has higher plasticity and adaptability through compared experiments.

Download Full-text

An improved OPTICS clustering algorithm for discovering clusters with uneven densities

Intelligent Data Analysis ◽

10.3233/ida-205497 ◽

2021 ◽

Vol 25 (6) ◽

pp. 1453-1471

Author(s):

Chunhua Tang ◽

Han Wang ◽

Zhiwen Wang ◽

Xiangkun Zeng ◽

Huaran Yan ◽

...

Keyword(s):

Time Complexity ◽

Clustering Algorithm ◽

Nearest Neighbor ◽

Clustering Algorithms ◽

Substantial Improvement ◽

Experimental Results ◽

High Time ◽

Parameter Setting ◽

K Nearest Neighbor ◽

Density Based Clustering

Most density-based clustering algorithms have the problems of difficult parameter setting, high time complexity, poor noise recognition, and weak clustering for datasets with uneven density. To solve these problems, this paper proposes FOP-OPTICS algorithm (Finding of the Ordering Peaks Based on OPTICS), which is a substantial improvement of OPTICS (Ordering Points To Identify the Clustering Structure). The proposed algorithm finds the demarcation point (DP) from the Augmented Cluster-Ordering generated by OPTICS and uses the reachability-distance of DP as the radius of neighborhood eps of its corresponding cluster. It overcomes the weakness of most algorithms in clustering datasets with uneven densities. By computing the distance of the k-nearest neighbor of each point, it reduces the time complexity of OPTICS; by calculating density-mutation points within the clusters, it can efficiently recognize noise. The experimental results show that FOP-OPTICS has the lowest time complexity, and outperforms other algorithms in parameter setting and noise recognition.

Download Full-text

Improved Fuzzy C-Means Clustering for Transformer Fault Diagnosis Using Dissolved Gas Analysis Data

Energies ◽

10.3390/en11092344 ◽

2018 ◽

Vol 11 (9) ◽

pp. 2344 ◽

Cited By ~ 6

Author(s):

Enwen Li ◽

Linong Wang ◽

Bin Song ◽

Siliang Jian

Keyword(s):

Fault Diagnosis ◽

Membership Function ◽

Data Clustering ◽

Clustering Algorithm ◽

Gas Analysis ◽

Dissolved Gas ◽

Fuzzy C Means ◽

Dissolved Gas Analysis ◽

Fcm Clustering ◽

Transformer Fault

Dissolved gas analysis (DGA) of the oil allows transformer fault diagnosis and status monitoring. Fuzzy c-means (FCM) clustering is an effective pattern recognition method, but exhibits poor clustering accuracy for dissolved gas data and usually fails to subsequently correctly classify transformer faults. The existing feasible approach involves combination of the FCM clustering algorithm with other intelligent algorithms, such as neural networks and support vector machines. This method enables good classification; however, the algorithm complexity is greatly increased. In this paper, the FCM clustering algorithm itself is improved and clustering analysis of DGA data is realized. First, the non-monotonicity of the traditional clustering membership function with respect to the sample distance and its several local extrema are discussed, which mainly explain the poor classification accuracy of DGA data clustering. Then, an exponential form of the membership function is proposed to obtain monotony with respect to distance, thereby improving the dissolved gas data clustering. Likewise, a similarity function to determine the degree of membership is derived. Test results for large datasets show that the improved clustering algorithm can be successfully applied for DGA-data-based transformer fault detection.

Download Full-text

Genetic Algorithm Based Parallel K-Means Data Clustering Algorithm Using MapReduce Programming Paradigm on Hadoop Environment (GAPKCA)

Advances in Intelligent Systems and Computing - Recent Advances on Soft Computing and Data Mining ◽

10.1007/978-3-030-36056-6_10 ◽

2019 ◽

pp. 98-108 ◽

Cited By ~ 1

Author(s):

Sayer Alshammari ◽

Maslina Binti Zolkepli ◽

Rusli Bin Abdullah

Keyword(s):

Genetic Algorithm ◽

Data Clustering ◽

Clustering Algorithm ◽

Programming Paradigm

Download Full-text

Algoritma K-Means Clustering dalam Mengklasifikasi Data Daerah Rawan Tindak Kriminalitas (Polres Kepulauan Mentawai)

Jurnal Sistim Informasi dan Teknologi ◽

10.37034/jsisfotek.v3i4.179 ◽

2021 ◽

pp. 243-248

Author(s):

Yoni Aswan ◽

Sarjon Defit ◽

Gunadi Widi Nurcahyo

Keyword(s):

Data Clustering ◽

Clustering Algorithm ◽

Motor Vehicle ◽

Motor Vehicles ◽

Hierarchical Data ◽

Motor Vehicle Theft ◽

Vehicle Theft ◽

Mentawai Islands ◽

Or Groups ◽

Different Characteristics

Crime is all kinds of actions and actions that are economically and psychologically harmful that violate the laws in force in the State of Indonesia as well as social and religious norms. Ordinary criminal acts affect the security of the community and threaten their inner and outer peace. The research location is the Mentawai Islands Police, which is an agency that can provide security and protection for the community, especially those in the Mentawai Islands Regency. The problem is that it is difficult for the Mentawai Islands Police to classify areas that are prone to crime in the most vulnerable, moderately vulnerable and not vulnerable categories. Especially considering the condition of the Mentawai, there are four large islands consisting of 10 sub-districts, where crime is increasing every year, especially those in the Mentawai Islands Regency area such as motor vehicle theft. Based on the background of the problem above, the researcher is interested in taking research in creating a system to predict the crime rate in the Mentawai Islands Regency in order to anticipate the surge in crime that will come. The method used is the K-Means Clustering Algorithm as a non-hierarchical data clustering method to partition existing data into one or more clusters or groups. This method partitions data into clusters so that data with the same characteristics are grouped into the same cluster and data with different characteristics are grouped into other clusters. Clustering is one of the data mining techniques used to get groups of objects that have common characteristics in large enough data. The data used is data on cases of criminal theft of motor vehicles for the last 5 years from 2016 to 2020. The results of the test show that South Sipora District is an area prone to the crime of motor vehicle theft.

Download Full-text