A Comparison Study on Similarity and Dissimilarity Measures in Clustering Continuous Data

Shirkhorshidi, Ali Seyed; Aghabozorgi, Saeed; Wah, Teh Ying

doi:10.1371/journal.pone.0144059

Cited by 323 publications

(152 citation statements)

References 34 publications

Supporting

Mentioning

148

Contrasting

Unclassified

Order By: Relevance

“…The Manhattan distance observes the absolute distance between two data items and dimensions. This distance measure is known as a city block distance, absolute value distance, rectilinear distance, taxicab distance or L1 distance family and gives cluster shape as hyper-rectangular [36]. Manhattan distance formulation shows as Eq.2 based upon the Eq.1.…”

Section: Distance Measures Taxonomymentioning

confidence: 99%

An Empirical Perusal of Dis tance Measures for Clustering with Big Data Mining

Pandey¹,

Shukla²

2019

IJEAT

View full text Add to dashboard Cite

The distance measure is the core idea of data mining techniques such as classification, clustering, and statistical analysis and so on. All clustering taxonomies such as partition, hierarchical, density, grid, model, fuzzy and graphs used to distance measures for the data point’s categorization under difference cluster, cluster construction and validation. Big data mining is the advanced concept of data mining respect to the big data dimensions. When traditional clustering algorithm is used under the big data mining the distance measure is needed for scalable under big data mining and support to a huge size dataset, heterogeneous data and sources, and velocity characteristics of the big data. From a theoretically, practically and the existing research perspective, the paper focuses on volume, variety, and velocity big data criterion for identifying a distance measure for the big data mining and recognize how to distance measure works under clustering taxonomy. This study also analyzed all distance measures accuracy with the help of a confusion matrix through clustering.

show abstract

Section: Distance Measures Taxonomymentioning

confidence: 99%

An Empirical Perusal of Dis tance Measures for Clustering with Big Data Mining

Pandey¹,

Shukla²

2019

IJEAT

View full text Add to dashboard Cite

show abstract

“…If the value is <50 it is dissimilar or difference (F 1 ). [7] Stability studies and report Stability studies were performed using stability chamber (Eye Instruments Pvt. Ltd., Ahmedabad) on D-3 batches tablets for 6 months.…”

Section: Similarity and Dissimilarity Studymentioning

confidence: 99%

Untitled

Bhattacharya

2017

AJP

View full text Add to dashboard Cite

Aim: Celecoxib has a tremendous role in the treatment of colon polyps and Crohn's disease. However, gastric resistant has always been an issue for drug delivery to the colon. Our main intention of this experiment was to prepare and evaluate a formula which could resist gastrointestinal fluid and releases drug content not more than 10% within simulated gastric fluid for 2 h from the time of administration. Materials and Methods: In this experiment at first, we prepared Celecoxib nanoparticles using Dyno Mill taking Acconon MC8-2EP as a surfactant and Capmul MCM L-8 as cosurfactant. Freshly prepared Celecoxib nanoparticles were then admixed with Lactopress ® anhydrous and using dry granulation technique 10 batches of Celecoxib tablets were formulated by altering various ratios of resistant starch, dextran, and gellan gum. Tablets were prepared using dry granulation technique, where ALTRIN ® was considered as a principal binder. Results and Discussion: All the pre and post compression parameters were evaluated, and it was found that D-3 batch has legitimate cumulative percentage dissolution profile up to a 24 th h (98.12%). Furthermore, similarity and dissimilarity studies were performed against Ortho bed tablet (marketed) and the test optimized formula D-3. The similarity factor (F 2 ) and difference or dissimilarity factors (F 1 ) were found to be 60.90 and 10.16, respectively, which is within the specified limits. Finally, as per ICH guideline Q1A (R2) at 40°C ± 2°C/75% RH ± 5% RH accelerated stability studies was performed in the D-3 formulation for 6 months. Stability results were quite satisfactory. Conclusion: Hence, it can be concluded that the optimized D-3 batch can be conceded for the pilot scalp.

show abstract

“…If these conditions are provided, clustering can be successful, otherwise, it is unsuccessful. Clustering, which is an important method in information inference with all these features, has several application fields such as pattern recognition, bio-information, text mining and machine learning [7,8].…”

Section: Introductionmentioning

confidence: 99%

Design and Implementation of an Intelligent Mobile Game

Ekinci¹,

Gülağız²,

Omurca³

2017

IJMECS

View full text Add to dashboard Cite

Abstract-While the mobile game industry is growing with each passing day with the popularization of 3G smart devices, the creation of successful games, which may interest users, become quite important in terms of the survival of the designed game. Clustering, which has many application fields, is a successful method and its implementation in the field of mobile games is inevitable. In this study, classical ball blasting game was carried out based on clustering. In the game, clustering the color codes with K-means, Iterative K-means, Iterative Multi K-means and K-medoids methods and blasting the balls of colors located in the same cluster by bringing them together were proposed. As a result of the experiments, the suitability of clustering methods for mobile based ball blasting game was shown. At the same time, the clustering methods were compared to produce the more successful clusters and because of obtaining more accurate results and stability, the use of K-medoids method has been chosen for this game.

show abstract

A Comparison Study on Similarity and Dissimilarity Measures in Clustering Continuous Data

Cited by 323 publications

References 34 publications

An Empirical Perusal of Dis tance Measures for Clustering with Big Data Mining

An Empirical Perusal of Dis tance Measures for Clustering with Big Data Mining

Untitled

Design and Implementation of an Intelligent Mobile Game

Contact Info

Product

Resources

About