2017
DOI: 10.1007/s10586-017-1436-9
|View full text |Cite
|
Sign up to set email alerts
|

A new cluster computing technique for social media data analysis

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
4
0

Year Published

2018
2018
2023
2023

Publication Types

Select...
9

Relationship

0
9

Authors

Journals

citations
Cited by 18 publications
(4 citation statements)
references
References 20 publications
0
4
0
Order By: Relevance
“…The real number of connections e i was the degree of interconnection between nodes in Facebook network of i ; this was calculated by dividing the number of mutual friends between transmitters and each of their friends, and then dividing the result by 2 to avoid recalculating the mutual relationship between i and n 1 or i and n 2 in the data collected by the application (Xu and Li, 2019). Thus, e i was formally defined using the adjacency matrix J , as shown in Eq.…”
Section: Methodsmentioning
confidence: 99%
“…The real number of connections e i was the degree of interconnection between nodes in Facebook network of i ; this was calculated by dividing the number of mutual friends between transmitters and each of their friends, and then dividing the result by 2 to avoid recalculating the mutual relationship between i and n 1 or i and n 2 in the data collected by the application (Xu and Li, 2019). Thus, e i was formally defined using the adjacency matrix J , as shown in Eq.…”
Section: Methodsmentioning
confidence: 99%
“…Working with conventional, locally based computer systems can make this challenging, let alone cloud datacenters, which may be thousands of miles away, run 40,000 virtual machines (VMs) across 512 servers, and are accessed by 1000 tenants, of which the accessed is one. This may just be too much information for the ordinary juror to process given that they only have a basic grasp of how to utilise a home computer [73][74][75].…”
Section: Digital Cloud Forensicsmentioning
confidence: 99%
“…To solve this problem, literature [19] introduces the concept of "generation" in the operation of the single-pass algorithm, inputting the document set in batches, clustering each batch of documents first, and then clustering the initial clustering results with the existing topic clusters, which effectively alleviates the order-sensitive problem of the single-pass algorithm's order-sensitive problem but makes the clustering results affected by the preliminary clustering process. In literature [20], based on the K-means clustering algorithm, the canopy algorithm is introduced to initialize the data, and the results of the algorithm are continuously updated by combining the hood center in the canopy algorithm and the class cluster center in the K-means algorithm, while the parallelized operation of the canopy-k-means algorithm is realized based on the Hadoop platform. e topic clustering results obtained in literature [21] based on this scheme are less affected by the input order of news data but still need to set the number of topics in advance, which is difficult to predict accurately in the complex Internet environment.…”
Section: Complexitymentioning
confidence: 99%