Enhancement of CURE algorithm using Map-Reduce Technique with Parallelism

The extraction of useful information from huge databases is one of the key areas of data mining and is open for research. Clustering integrates the data with higher similarity into the same group, enhancing the extraction process. Clustering Using Representatives (CURE) is an efficient clustering al...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2023 International Conference on Data Science, Agents & Artificial Intelligence (ICDSAAI) s. 1 - 4
Hlavní autoři: Chavva, Subba Reddy, Vijayaraj, A., Mageshkumar, N., Senthilvel, P. Gururama
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 21.12.2023
Témata:
Artificial intelligence
Clustering
Clustering algorithms
Data mining
Map-Reduce
Outliers
Parallel processing
Parallelism
Scalability
Time complexity
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The extraction of useful information from huge databases is one of the key areas of data mining and is open for research. Clustering integrates the data with higher similarity into the same group, enhancing the extraction process. Clustering Using Representatives (CURE) is an efficient clustering algorithm that handles voluminous data. However, CURE uses sampling, which encounters scalability and accuracy issues while processing huge databases. This limitation can be repressed by using the Map-reduce technique in CURE instead of sampling. The clustering performance can be further enhanced by using parallelism integrated into the CURE algorithm to reduce processing time and enhance efficiency.
DOI:10.1109/ICDSAAI59313.2023.10452533