Getting valuable insight through data mining

Every day 2.5 quintillion bytes of data are created. In order to get valuable insight from those data, you need to uncover patterns and relationships among them. That’s what data mining does. 

Dian Puteri Ramadhani, who is from the Social Computing & Big Data Laboratory and a lecturer at the Telkom University, affirmed that data mining would discover patterns in a data set and be useful for conducting research.

She explained it during the DMSN Talk Webinar, episode 2, Friday (19/2/2021). The organizer was the Decision Making and Strategic Negotiation (DMSN), an interest group of SBM ITB, which invited all academicians to discuss the usefulness of data mining for research.

The rapid development of data mining is driven by the advancement of information and communication systems. Mining means getting valuable things within a large quantity of material. In daily life, the application of data mining can be found in convenience stores that record the transactions of consumers. By analyzing the transaction (data) they can find a particular pattern and lead them to insight. For instance, they may know the most favourable product favoured by customers. 

In another application, Dian explained that data mining is really helpful to discover the information spread mechanism about a product and find out the top influencers or communities. It can help companies uncover potential partners to collaborate with.

“Based on research by a friend of mine, he discovered that K-Pop communities are the most information spread of Samsung products by analyzing graph-based data mining,” she said. So, it gives us an answer to why Samsung collaborated with Korean artists to promote its product.

Furthermore, Muhammad Apriandito revealed that in business, data mining can also be applied to predict employee turnover by using a machine learning approach. He argued that Machine learning algorithms could automate the process of learning a model that captured the relationship between the descriptive features and the target feature dataset.

There are also many applications of machine learning in our entire activities. As he explained, a simple application of machine learning is when we select a video on youtube, then the next playlist will lead us to the several related videos we may want.

“Nowadays, we are lucky because there are many tools and journals that we can use to learn about machine learning. It was quite different in 2018 when I started this research, at the time this study was still undeveloped,” revealed Muhammad who is also an alumnus of SBM ITB.

Written by Student Reporter (Deo Fernando, Entrepreneurship 2021)