Novri Suhermi SSi MSi as award winner of The Best Poster

ITS Campus, ITS News – Covid-19 pandemic does not stop the civitas academica of Institut Teknologi Sepuluh Nopember (ITS) from achieving an award. One of the lecturers of Department of Statistics ITS, Novri Suhermi SSi MSi, proves his proficiency by mapping the energy consumption in Lancaster University, UK and successfully brought home the Best Poster Award in Machine Learning Summer School (MLSS) held by Telkom University for seven days, starting Monday (3/8).

Novri design the posters relating to the energy consumption of the universities where he currently took his doctoral program. Energy consumption is a number of energy exploited directly from natural resources without the conversion process. Based on Novri’s description, the presented poster is the research from his doctoral program. The research aims to model the pattern of energy consumption every time.

Furthermore, this 28-year-old man also explained the usefulness of his research displayed in the form of posters, namely identifying the factors that affect the magnitude of energy consumption and predicting the magnitude of energy consumption in the future. In this regard, energy consumption at Lancaster University consists of several types, including lighting, heating, water use, and general electricity consumption. “The factors can include occupancy rate, building design, maintenance arrangement, and weather,” said lecturer who is also graduates of ITS.

This data analysis experts also pointed out that to make the poster he needs enough data to process because at Lancaster University there are more than 100 campus buildings and there must be thousands of sensors that record energy consumption scattered throughout the building. Recorded data are also collected in real-time. Real-time computing is a condition of operation of a hardware and software system that is limited by time.

In processing its data, Novri uses a method called Functional Data Analysis (FDA). That is a set of methods for analyzing data through curves, statistical surfaces, and continuum. Novri immuted that the FDA method allows its users to model flexible data of multidimensional, high-frequency, and on a large scale basis.

Poster of energy consumption modelling presented in MLSS

With the number of building on campus and various types of energy consumption, the FDA method helps its users to perform automation at scale so that researchers can know the outcome. For example, in what building, what floor, what space, and the time when its energy consumption is inefficient. “This information will later be followed up to the stakeholders so that the policy for energy savings, the other example is when we make a prediction of the magnitude of energy consumption in the future,” explained the lecturer from Riau.

From Novri’s description, this energy consumption modeling will be related to the amount of energy needed and financing planning. Thus, data processing can be a platform to make regulations for energy usage policies. The final goal of this research is the adoption of a data-driven energy policy system.

Novri presents his poster at MLSS. That is a workshop or a kind of short-course that talks about machine learning and its application, such as image classification and Natural Language Processing (NLP). NLP itself is one of the branches of Artificial Intelligence (AI) Science that processes the use of natural language. “Machine learning is currently trending and its development is very rapid,” Novri said.

The poster presentation is one of the subactivities of MLSS and Novri is one of the participants. Participants who join MLSS are students of S1, S2, S3, lecturers, researchers, and practitioners from Indonesia and abroad, including Asia, Europe, America, Africa. “People, in general, may be more familiar with the terms Big Data, Data Science, Artificial Intelligence, and so on,” he said.

In the end, Novri explained that MLSS activities are great especially for research and machine learning practitioners in Indonesia. Also, this activity is capable of adding skills and improving the quality of research to a higher level especially for Indonesia, where opportunities to utilize technology, data, and AI are still extensive. “Therefore, the talents in the field of data science, machine learning, and AI will certainly be needed,” said the alumnus of Denis Diderot University, France. (jev/rev/ITS Public Relations)