Prof. DrsEc Ir Riyanarto Sarno, MSc PhD, is the only professor from ITS who has returned to the Top 2% of Scientists in the World 2023

ITS Campus, ITS News — Professor of Institut Teknologi Sepuluh Nopember (ITS) Prof. Drs Ec Ir Riyanarto MSc PhD has once again kicked off the world of research by presenting a stock prediction method using public sentiment or opinion. Applying the Transformer Encoder Gated Recurrent Unit (TEGRU), his work achieved his position as one of the Top 2% of Scientists in the World in 2023.

This study, which relies on everyday life phenomena, is considered by Riyan to play an essential role in individual decision-making. However, to study these sentiments, the transformation of qualitative data into quantitative data is necessary so that it can be analyzed more effectively. “Therefore, I applied the TEGRU method to analyze sentiment data,” he explained.

Sentiment data analysis scheme using the Transformer Encoder Gated Recurrent Unit (TEGRU) method

In simple terms, this research combines two essential processes in data interpretation. This approach includes using the Transformer Encoder method, which is tasked with extracting information from the dataset, as well as implementing a Gated Recurrent Unit (GRU), which functions to process and predict data continuously. With the results of this approach, this method is very suitable for situations that require long-term analysis.

In its implementation, Riyan explained that TEGRU utilized data from various trusted news sources. In the next stage, this information will be organized into multiple categories, including but not limited to industrial sector news, investment, finance, national policies and international events.

The process of sorting news headlines in the sentiment analysis process

Through a meticulous selection process, TEGRU will sort various headlines and categorize them into two main classifications, namely positive or negative sentiment reflected in the news. This allows TEGRU to provide insight into how society responds to events and issues through rigorous sentiment analysis.

After successfully obtaining sentiment data, this graduate of the University of New Brunswick said the next step was to collect transaction data on the stock market. With two existing data sources, TEGRU will carry out a synchronization process and design scenarios for stock price movements in the future period using various careful calculation methods.

Presentation of the latest innovations by Prof Drs Ec Ir Riyanarto Sarno MSc PhD at the 2023 Education Fund Management Institute (LPDP) Festival

Looking at its use, Riyan revealed that the market will benefit from more accurate and reliable predictions. This is caused by an approach that integrates two critical aspects, namely, share price transaction data and public sentiment. “Through this method, we can estimate fluctuations in public opinion that have the potential to influence share sales,” said Riyan.

Furthermore, TEGRU is not just a stock sales prediction tool. The Head of the Intelligent Management Laboratory of the ITS Informatics Engineering Department is confident that this method can solve various problems that require predictive analysis in multiple areas of human life.

The latest innovation is one of the brilliant works that brought him back to win the title of Top 2% Scientists in the world in 2023. Riyan said that the achievement of these four was a gift and a big responsibility that he carried in providing benefits to the nation. “I am delighted to be able to continue to contribute to the glory of the nation through thought and research,” he concluded, feeling proud. (ITS Public Relations)

Reporter: Hibar Buana Puspa