A Parallel Clustering Algorithm for Power Big Data Analysis

This study sought to propose a big data analysis and prediction model for transmission line tower outliers to assess when something is wrong with transmission line tower big data based on deep reinforcement learning. The model enables choosing automatic cluster K values based on non-labeled sensor big data. It also allows measuring the distance of action between data inside a cluster with the Q-value representing network output in the altered transmission line tower big data clustering algorithm containing transmission line tower outliers and old Deep Q Network. Specifically, this study performed principal component analysis to categorize transmission line tower data and proposed an automatic initial central point approach through standard normal distribution. It also proposed the A-Deep Q-Learning algorithm altered from the deep Q-Learning algorithm to explore policies based on the experiences of clustered data learning. It can be used to perform transmission line tower outlier data learning based on the distance of data within a cluster. The performance evaluation results show that the proposed model recorded an approximately 2.29%~4.19% higher prediction rate and around 0.8% ~ 4.3% higher accuracy rate compared to the old transmission line tower big data analysis model.

Download Full-text

THE MODELING OF "MUSTAHIQ" DATA USING K-MEANS CLUSTERING ALGORITHM AND BIG DATA ANALYSIS (CASE STUDY: LAZ)

JURNAL TEKNIK INFORMATIKA ◽

10.15408/jti.v13i2.19610 ◽

2021 ◽

Vol 13 (2) ◽

pp. 213-230

Author(s):

Nurhayati Buslim ◽

Rayi Pradono Iswara ◽

Fajar Agustian

Keyword(s):

Big Data ◽

Data Analysis ◽

Clustering Algorithm ◽

System Modeling ◽

Big Data Analysis ◽

Graphical Form ◽

Data Types ◽

Online Data ◽

Combining Data ◽

Data Survey

There are a lot of Mustahiq data in LAZ (Lembaga Amil Zakat) which is spread in many locations today. Each LAZ has Mustahiq data that is different in type from other LAZ. There are differences in Mustahiq data types so that data that is so large cannot be used together even though the purpose of the data is the same to determine Mustahiq data. And to find out whether the Mustahiq data is still up to date (renewable), of course it will be very difficult due to the types of data types that are not uniform or different, long time span, and the large amount of data. To give zakat to Mustahiq certainly requires speed of information. So, in giving zakat to Mustahiq, LAZ will find it difficult to monitor the progress of the Mustahiq. It is possible that a Mustahiq will change his condition to become a Muzaki. This is the reason for the researcher to take this theme in order to help the existing LAZ to make it easier to cluster Mustahiq data. Furthermore, the data already in the cluster can be used by LAZ managers to develop the organization. This can also be a reference for determining the zakat recipient cluster to those who are entitled later. The research is "Modeling using K-Means Algorithm and Big Data analysis in determine Mustahiq data ". We got data Mustahiq with random sample from online and offline survey. Online data survey with Google form and Offline Data survey we got from BAZNAS (National Amil Zakat Agency) in Indonesia and another zakat agency (LAZ) in Jakarta. We conducted by combining data to analyzed using Big Data and K-Means Algorithm. K-Means algorithm is an algorithm for cluster n objects based on attributes into k partitions according to criteria that will be determined from large and diverse Mustahiq data. This research focuses on modeling that applies K-Means Algorithms and Big Data Analysis. The first we made tools for grouping simulation test data. We do several experimental and simulation scenarios to find a model in mapping Mustahiq data to developed best model for processing the data. The results of this study are displayed in tabular and graphical form, namely the proposed Mustahiq data processing model at Zakat Agency (LAZ). The simulation result from a total of 1109 correspondents, 300 correspondents are included in the Mustahiq cluster and 809 correspondents are included in the Non-Mustahiq cluster and have an accuracy rate of 83.40%. That means accuracy of the system modeling able to determine data Mustahiq. Result filtering based on Gender is “Male” accuracy 83.93%, based on Age is ”30-39” accuracy 71,03%, based on Job is “PNS” accuracy 83.39%, based on Education is “S1” accuracy 83.79%. The advantaged of research expected to be able to determine quickly whether the person meets the criteria as a mustahik or Muzaki for LAZ (Amil Zakat Agency). The result of modeling is K-Means clustering algorithm application program can be used if UIN Syarif Hidayatullah Jakarta want to develop LAZ (Amil Zakat Agency) too.

Download Full-text

Research and Application on Spark Clustering Algorithm in Campus Big Data Analysis

Journal of Computer Science Research ◽

10.30564/jcsr.v2i1.1808 ◽

2020 ◽

Vol 2 (1) ◽

Author(s):

Qing Hou ◽

Guangjian Wang ◽

Xiaozheng Wang ◽

Jiaxi Xu ◽

Yang Xin

Keyword(s):

Big Data ◽

Data Analysis ◽

Clustering Algorithm ◽

Big Data Analysis ◽

Psychological Status ◽

Student Management ◽

Multiple Dimensions ◽

Daily Behavior ◽

Teachers And Students ◽

Management Work

Big data analysis has penetrated into all fields of society and has brought about profound changes. However, there is relatively little research on big data supporting student management regarding college and university’s big data. Taking the student card information as the research sample, using spark big data mining technology and K-Means clustering algorithm, taking scholarship evaluation as an example, the big data is analyzed. Data includes analysis of students' daily behavior from multiple dimensions, and it can prevent the unreasonable scholarship evaluation caused by unfair factors such as plagiarism, votes of teachers and students, etc. At the same time, students' absenteeism, physical health and psychological status in advance can be predicted, which makes student management work more active, accurate and effective.

Download Full-text