A novel clustering technique for efficient clustering of big data in Hadoop Ecosystem

Kumar, Sunil; Singh, Maninder

doi:10.26599/bdma.2018.9020037

Cited by 35 publications

(19 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…only the patient will know that he/she is in the stroke by experiencing the outward symptoms. [3] Hybrid clustering technique is used in this existing work. Provide accurate results.…”

Section: Related Workmentioning

confidence: 99%

Early Strokes Detection of Patient and Health Monitoring System Based On Data Analytics Using K-Means Algorithm

Menaka¹,

Bharathiraja²,

Alekhya³

et al. 2021

Proceedings of the First International Conference on Computing, Communication and Control System, I3CAC 2021, 7-8 June 2021, Bh

View full text Add to dashboard Cite

The processing chain of scientific information in particular consists of information series, information storage, data sharing, and records evaluation. In the prevailing device, there is lots of improvisation wished for our health care device. The existing technique our affected person monitoring is a guide and time-consuming process. To conquer the task the proposed work proposed actual data series from sensors, IoT-primarily based totally sharing, and information analytics. This proposed device gives the gain for the respective medical doctor to display the affected person's health 24*7 regardless of geographical location. Example: The medical doctor can display the affected person's health even after the affected person receives discharged. This proposed work implies a health sensor named heartbeat sensor to display affected person fitness. The affected person information is monitoring through the sensorsand transmitted to the Arduino. The actual-time statistics from the COM port have acquired the usage of Net beans and stored in an SQL database. The actual-time information may be monitored with the aid of using each affected person and medical doctor. The real-time statistics are processed from Net beans as datasheet to R programming for statistical evaluation. For Clustering, we use the K-Means algorithm and for Classification, we use the Support Vector Machine. Also relying on the affected person's health situations emergency pills or injections are counseled routinely with the aid of using our device. Also, the affected person statistics have encrypted the usage of an ABE (Attribute-based Encryption) set of rules and saved in the public cloud particularly Dropbox. Thus invoking the statistics evaluation approach facilitates in identifying early stroke in the sufferers and offer medicinal diagnosis immediately.

show abstract

“…only the patient will know that he/she is in the stroke by experiencing the outward symptoms. [3] Hybrid clustering technique is used in this existing work. Provide accurate results.…”

Section: Related Workmentioning

confidence: 99%

Early Strokes Detection of Patient and Health Monitoring System Based On Data Analytics Using K-Means Algorithm

Menaka¹,

Bharathiraja²,

Alekhya³

et al. 2021

Proceedings of the First International Conference on Computing, Communication and Control System, I3CAC 2021, 7-8 June 2021, Bh

View full text Add to dashboard Cite

show abstract

“…From several articles studied mention that class grouping is done through questionnaires and interviews with teachers in providing learning over a period of 5 to 11 years [3], this is of course the results are less precise because it should be questionnaire and interview about grouping learning classes given to students who will start the process learning is not to the instructor [4], and the determination of class groups takes a long time because they have to wait for evaluation for several years [5]. In another article also mentioned the process of class grouping is done with the questionnaire choice "a" and "b" [6] and only to measure the student's personality value [7], this is less precise if what the institution wants is the potential value or ability of students to determine the concentration of certain lessons [8].…”

Section: Introductionmentioning

confidence: 99%

K-means method for clustering learning classes

Indriyanti

Prehanto

Vitadiar³

2021

IJEECS

View full text Add to dashboard Cite

<span>Learning class is a collection of several students in an educational institution. Every beginning of the school year the educational institution conducts a grouping class test. However, sometimes class grouping is not in accordance with the ability of students. For this reason, a system is needed to be able to see the ability of students according to the desired parameters. Determination of the weight of test scores is done using the K-Means method as a grouping method. Iteration or repetition process in the K-Means method is very important because the weight value is still very possible to change. Therefore, the repetition process is carried out to produce a value that does not change and is used to determine the ability level of students. The results of the class grouping test scores affect the ability of students. Application of K-Means method is used in building an information system grouping student admissions in an educational institution. Acceptance of students will be grouped into 3 groups of learning classes. The results of testing the system that applies K-Means method and based on data on the admission of prospective students from educational institutions have very high accuracy with an error rate of 0.074. </span>

show abstract

“…In addition to deep learning models, there have been efforts to analyze large datasets using clustering techniques on cloud architectures [18,19] . In this section, we briefly discuss the existing approaches to address vanishing gradient problem and study the impact of learning rates on the training time of deep learning models.…”

Section: Introductionmentioning

confidence: 99%

Gradient amplification: An efficient way to train deep neural networks

Basodi

Zhang

et al. 2020

Big Data Min. Anal.

View full text Add to dashboard Cite

Improving performance of deep learning models and reducing their training times are ongoing challenges in deep neural networks. There are several approaches proposed to address these challenges, one of which is to increase the depth of the neural networks. Such deeper networks not only increase training times, but also suffer from vanishing gradients problem while training. In this work, we propose gradient amplification approach for training deep learning models to prevent vanishing gradients and also develop a training strategy to enable or disable gradient amplification method across several epochs with different learning rates. We perform experiments on VGG-19 and Resnet models , and study the impact of amplification parameters on these models in detail. Our proposed approach improves performance of these deep learning models even at higher learning rates, thereby allowing these models to achieve higher performance with reduced training time.

show abstract

A novel clustering technique for efficient clustering of big data in Hadoop Ecosystem

Cited by 35 publications

References 20 publications

Early Strokes Detection of Patient and Health Monitoring System Based On Data Analytics Using K-Means Algorithm

Early Strokes Detection of Patient and Health Monitoring System Based On Data Analytics Using K-Means Algorithm

K-means method for clustering learning classes

Gradient amplification: An efficient way to train deep neural networks

Contact Info

Product

Resources

About