A New Approach to Determine Eps Parameter of DBSCAN Algorithm

Ozkok, Fatma Ozge; Çelik, Mete

doi:10.18201/ijisae.2017533899

Cited by 26 publications

(15 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our results show that the DBSCAN clustering method can be used to determine instars of insect. The advantages of the DBSCAN clustering algorithm compared with the other two clustering methods are that it does not require a known number of clusters prior to clustering, it can find clusters of arbitrary shapes, and it works well with large spatial databases and handles noise effectively (Ozkok & Celik, ; Rodriguez et al, ).…”

Section: Discussionmentioning

confidence: 99%

Determining the number of instars in potato tuber moth Phthorimaea operculella (Zeller) using density‐based DBSCAN clustering

Zheng

Wang

Chen

et al. 2019

J Applied Entomology

View full text Add to dashboard Cite

The potato tuber moth (PTM), Phthorimaea operculella (Zeller), is an important pest of Solanaceae crops and especially devastating to potatoes. There is no significant difference in morphological characteristics of PTM from the first to third instar larvae; therefore, it is difficult to directly determine the number of instars of this pest based on morphology. In the present study, head capsule width and length and mandible width of 340 PTM individuals were measured. Density‐based spatial clustering of applications with noise (DBSCAN) clustering was used for instar grouping. The results of DBSCAN clustering were compared with those obtained using Gaussian mixture models and k‐means clustering; the results of the three clustering methods were verified using Brooks–Dyar rule, Crosby rule and linear regression model. The clusters obtained using the three methods were the same and comprised four PTM instars with three morphological characteristics. Moreover, the results of the three methods fit the Brooks–Dyar rule, Crosby rule, frequency analysis and logarithmic regression model well. Head capsule width was the best morphological characteristic for determining the number of instars of PTM, and this characteristic may be used for determining PTM instars in the field. These results show that the DBSCAN clustering method is a promising tool for the identification of insect instars.

show abstract

Section: Discussionmentioning

confidence: 99%

Determining the number of instars in potato tuber moth Phthorimaea operculella (Zeller) using density‐based DBSCAN clustering

Zheng

Wang

Chen

et al. 2019

J Applied Entomology

View full text Add to dashboard Cite

show abstract

“…DBSCAN algorithm requires two parameters : Eps, which is used to determine the neighbouring area of an object (or point) and MinPts, which is the minimum number of points in the Eps radius [28]. In this study, it is proposed to determining the value of Eps and MinPts with trial and error parameter, meaning that it determines the value of a parameter must be doing several times to get the expected number of clusters.…”

Section: Density-based Spatial Clustering Of Application With Noisementioning

confidence: 99%

Fraud Detection Call Detail Record Using Machine Learning in Telecommunications Company

Jabbar¹,

Suharjito²

2020

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

Fraud calls have a serious impact on telecommunications operator revenues. Fraud detection is very important because service providers can feel a significant loss of income. We conducted a fraud research case study on one of the operators that experienced fraud in 2009 and 2018. Call Detail Record (CDR) containing records of customer conversations such as source and destination number, call start time, duration of calls at the operator can be a source of information to use in fraud detection. The method used in this study uses machine learning with unsupervised learning techniques which are quite popular methods used in fraud detection. The purpose of this study is to propose an effective method that can be applied to detect fraud on the CDR. Variables used include caller number, number dialled, duration, fee and destination city of the dataset totalling 11,418 rows from record periods 01 to 31 May 2018. In analyzing our CDR using the K-Means and DBSCAN algorithms, we then evaluate the results to calculate accuracy by comparing to actual fraud data. Based on evaluations using confusion matrix on actual CDR fraud, we obtained the K-Means algorithm to show a better accuracy value to model fraud on telecommunications CDR compared to DBSCAN.

show abstract

“…To address the scalability issue of high-dimensional data clustering, Müller et al (2009a) proposed a subspace clustering algorithm based on density of objects. Ozkok and Celik (2017) propose AE-DBSCAN algorithm which includes a new method to determine the value of neighborhood radius Epsautomatically. Another attempt to estimate appropriate parameter values for subspace clustering is done by Lee and Shim (2015).…”

Section: Recent Related Workmentioning

confidence: 99%

A Novel Parameter-Light Subspace Clustering Technique Based on Single Linkage Method

Kelkar

Rodd²,

Kulkarni

2019

J. Info. Know. Mgmt.

View full text Add to dashboard Cite

Subspace clustering is a challenging high-dimensional data mining task. There have been several approaches proposed in the literature to identify clusters in subspaces, however their performance and quality is highly affected by input parameters. A little research is done so far on identifying proper parameter values automatically. Other observed drawbacks are requirement of multiple database scans resulting into increased demand for computing resources and generation of many redundant clusters. Here, we propose a parameter light subspace clustering method for numerical data hereafter referred to as CLUSLINK. The algorithm is based on single linkage clustering method and works in bottom up, greedy fashion. The only input user has to provide is how coarse or fine the resulting clusters should be, and if not given, the algorithm operates with default values. The empirical results obtained over synthetic and real benchmark datasets show significant improvement in terms of accuracy and execution time.

show abstract

A New Approach to Determine Eps Parameter of DBSCAN Algorithm

Cited by 26 publications

References 14 publications

Determining the number of instars in potato tuber moth Phthorimaea operculella (Zeller) using density‐based DBSCAN clustering

Determining the number of instars in potato tuber moth Phthorimaea operculella (Zeller) using density‐based DBSCAN clustering

Fraud Detection Call Detail Record Using Machine Learning in Telecommunications Company

A Novel Parameter-Light Subspace Clustering Technique Based on Single Linkage Method

Contact Info

Product

Resources

About