Professor of IPB University Discusses the Importance of the Bioinformatics Approach in the Big Data Era

In today’s big data era, data processing is essential. Both are applied in processing data on the internet and the field. Smart agriculture also has implications for analysis needed in agriculture through a bioinformatics approach.

Prof. Imas S Sitanggang, Professor of IPB University from the Department of Computer Science, Faculty of Mathematics and Natural Sciences (FMIPA), said that data mining processes are essential in the big data era. Data mining is the process of extracting interesting information and patterns from extensive data. Output data obtained which has the potential to be analyzed and used further.

According to him, data mining is closely related to various fields. Such as machine learning, statistics, and databases. However, it is possible to apply it in other related areas.

“Therefore, statistics courses need to be offered to students. This knowledge needs to be mastered if you want to become a data miner so that you can get an interesting pattern,” he explained in the Public Lecture of Master of Informatics Faculty of Science and Technology, State Islamic University (UIN) Sunan Kalijaga Yogyakarta (26/10) which was held online.

In general, he continued, the basic techniques of data mining are categorized into three. Namely cluster, classification, and association analysis. The goal is to create a predictive model and provide a description of the data held.

“An example of its application is in analyzing market segmentation or marketing targets according to customer characteristics. This approach can also be applied to expenditure analysis. Likewise with research in the environmental field, such as analyzing environmental characteristics based on the incidence of forest fires,” he added.

He added that the pre-processing or data cleaning process is an essential part of data mining. Generally, the real data obtained is not good and has ‘impurities,’ so there must be data cleaning.

He has applied the data mining approach to several studies. Among them are aerial photo research and the identification of garlic fields using deep learning. When compared to conventional machine learning, deep learning is rated better. However, the use of deep learning remains tailored to the needs.

“The research on garlic is carried out because it is an agricultural commodity whose demand continues to increase. So there must be a mapping of areas that are suitable for garlic cultivation. This study uses a data mining algorithm for the Magetan and Solok areas. Then apply the garlic suitability class referring to the FAO (Food and Agriculture Organization),” he explained.

According to him, garlic spatial data is obtained by spatial operations and entered into data mining algorithms. The output is in the form of a garlic suitability profile. The conformity results are then conveyed to the user through a specially made information system, such as a geographic information system.

“The issue of how to present mining results to users is also an interesting research area,” he added.
He also discussed the data warehouse, a collection of data integrated from various sources in the form of a repository. The output becomes a source for data mining processes, and its application can also be used for multiple analytical purposes. (MW/Zul) (IAAS/ERN) 

Published Date : 27-Oct-2021

Resource Person : Prof Imas S Sitanggang

Keyword : Professor of IPB University, bioinformatics, big data, data mining, data warehouse