Journal of Big Data Science

Literature Review: Data Mining For Student Data Classification

Billi Mahardika — Sat, 30 Sep 2023 13:20:47 +0000

The abundance of student data and data on the number of student graduations, hidden information can be found by processing student data so that it is useful for the university. Processing of student data needs to be done to find out important information in the form of new knowledge (knowledge discovery). Method: Literature review is carried out based on issues, methodology, equations.result : Decision Trees) Decision tree analysis (or decision tree analysis is a technique that belongs to the machine-learning family) is arguably the most popular classification technique in the data mining area.

Scientometrics Study and Authorship Network Analysis in Universitas Bina Darma Lecturer Publications

Ferdi Aditya, Deni Erlansyah, Dendi Triadi, Rezki Syaputra — Wed, 18 Oct 2023 00:00:00 +0000

Scientometrics is a study of the measurement and analysis of science, innovation and technology through scientific publications. One form of measurement that can be done is through the measurement of the network of authors. This study uses authorsip network analysis as a measuring tool implemented in scientomentrics studies. The purpose of this study is to observe the Authorsip network formed between Bina Darma university lecturers, so as to determine which lecturers and faculties are the most productive in producing articles published in proceedings or journal. The method used in this research is Graph Degree Centrality. The software used for visualization of Gephi 0.9.2. The data used in this study is publication data for 2015 – 2020. Based on the results of this study, it can be concluded that the actor with the highest degree centrality value is the EU with a value of 28, where the EU is the actor with the highest number of publications. Meanwhile, the actor who has influence or relationship and often collaborates in publications with the highest score on Betweenness Centrality is AM with a score of 61500.94.

Hadoop Perfomance Improvement For Big Data Processing On Trans-Regional Networks Using ViNe

Rahmat Novrianda, Alek Wijaya, Timur Dali Purwanto, Rezki Syaputra — Wed, 18 Oct 2023 12:45:00 +0000

The exponential growth of data continues to increase in human life, the data is sourced from sensor data, social media data and social networking service (SNS) data which is referred to as the Big Data era. As data grows bigger and faster, the cost of storing and analyzing data becomes more and more expensive. This article proposes an approach to improve Hadoop performance in a trans-regional network environment through a virtual networking tool called ViNe. This approach enables IoT-sourced Big Data to store data in geographically distributed storage. Additionally, the approach allows IoT Big Data analyzers to deploy MapReduce applications on top of a trans-regional Hadoop Distributed File System.