A novel time series clustering method with fine-tuned support vector regression for customer behavior analysis

Abbasimehr, Hossein; Baghery, Farzam Sheikh

doi:10.1016/j.eswa.2022.117584

Cited by 14 publications

(4 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(1) A novel clustering method is developed and demonstrated for clustering complex crowdfunding projects. The quality of clustering methods (Abbasimehr & Sheikh, 2022) affects the clustering effect of complex crowdfunding projects. Our research applies the Apriori algorithm to improve the quality of clustering and product recommendations in complex crowdfunding ecosystems.…”

Section: Theoretical Insightsmentioning

confidence: 99%

Matchmaking in reward-based crowdfunding platforms: a hybrid machine learning approach

Mangla

et al. 2022

International Journal of Production Research

View full text Add to dashboard Cite

Traditional clustering methods fail to accurately cluster the feature vectors of backers and recommend compatible crowdfunding projects mainly because of their sensitivity to the setting of the initial values. Following a hybrid machine learning approach, this study aims to provide a practical solution that can accurately match listed products/projects with registered backers through a crowdfunding platform. For this purpose, we examine how the Apriori algorithm can be used in conjunction with other tools to optimise traditional clustering and provide more accurate recommendations for crowdfunding projects. Focusing on potential projects listed in a major crowdfunding website, we first train the data obtained from the available list of backers. Through the association rules of the Apriori algorithm, we obtain the degree of association between the different backers of each project and evaluate the backers' final correlation strength according to their association degree obtained from various crowdfunding projects. The degree of association is used as a key index to cluster backers according to their similar characteristics. Next, we test the model and determine whether it can correctly classify the data in the test set based on the Apriori algorithm.Experimental results show that the applied model 90% accuracy, precision, and recall of the applied model. The proposed solution outperforms the other five benchmark methods in connecting the right crowdfunding projects to the relevant backers.

show abstract

Section: Theoretical Insightsmentioning

confidence: 99%

Matchmaking in reward-based crowdfunding platforms: a hybrid machine learning approach

Mangla

et al. 2022

International Journal of Production Research

View full text Add to dashboard Cite

show abstract

“…Using angular features extracted from short time series genes, a clustering algorithm was employed in [8]. Using the Laplacian feature ranking method, Abbasimehr and Baghery propose a clustering algorithm for customer behavior analysis [7].…”

Section: Related Workmentioning

confidence: 99%

“…Due to its effectiveness in numerous applications, such as anomaly detection, selective marketing, decision support, privacy-preserving data publishing [1,2], and many others, data mining is now a hot research topic and is one of the most important tools in today's data-intensive world [3][4][5][6]. One of the most active research areas is data clustering, among others [7][8][9]. Data clustering techniques can be used to perform similarity searches, pattern recognition, trend analysis, grouping, and classification, among other tasks [3,[10][11][12][13].…”

Section: Introductionmentioning

confidence: 99%

Accelerated univariate sequential clustering

Enayati

Mortazavi

Basiri

2022

Preprint

View full text Add to dashboard Cite

Data clustering is an important task in the field of data mining. In many real applications, clustering algorithms must consider the order of data, resulting in the sequential clustering problem. For instance, analyzing the moving pattern of an object and detecting community structure in a complex network are related to sequential clustering. The constraint of the continuous region prevents previous clustering algorithms from being directly applied to the problem. A dynamic programming algorithm was proposed to address the issue, which returns the optimal sequential clustering. However, it is not scalable. This paper addresses the issue via a greedy stopping condition that prevents the algorithm from continuing to search when it's likely that the best solution has been found. Experimental results on multiple datasets show that the algorithm is much faster than its original solution while the optimality gap is negligible.

show abstract

“…Today, a vast amount of data in various fields in science, such as economics [43], business [1,26], healthcare [8,59] and social sciences [38], is captured in the form of time series [29,34]. Each time series, while consisting of a sequence of temporal data, can also be seen as a single object, often showing similar properties to other time series [63].…”

Section: Introductionmentioning

confidence: 99%

Time Series Overlapping Clustering Based on Link Community Detection

Ghahremani,

Amiri

2024

IEEE Access

View full text Add to dashboard Cite

Given the nature of time series and their vast applications, it is essential to find clustering algorithms that depict their real-life properties. Among the features that can hugely effect the options available for time series are overlapping and hierarchical properties. In this paper a novel approach to analyze time series with such features is introduced. Using the two concepts of network construction and link community detection, we have attempted to analyze and identify the mentioned properties of time series using data that is often gathered first hand. The proposed algorithm has been applied using both recent and common similarity measures on ten synthetic time series with hierarchal and overlapping features, alongside various distance measures. When testing the proposed approach, the element-centric measure of similarity indicated a clear increased accuracy for this algorithm, showing the highest accuracy when used alongside the Dynamic Time Warping distance measure. Moreover, the proposed algorithm has been very successful in identifying and forming communities for both large and small time series, thus solving another one of the main issues previous algorithms tended to have.

show abstract

A novel time series clustering method with fine-tuned support vector regression for customer behavior analysis

Cited by 14 publications

References 41 publications

Matchmaking in reward-based crowdfunding platforms: a hybrid machine learning approach

Matchmaking in reward-based crowdfunding platforms: a hybrid machine learning approach

Accelerated univariate sequential clustering

Time Series Overlapping Clustering Based on Link Community Detection

Contact Info

Product

Resources

About