Dunamu Machine Learning Team Selected for AAAI 2026 Demo Track - Demonstration of Financial News Analysis System Using AI Technology

Source: DigitalToday Original Title: Dunamu Machine Learning Team Selected for ‘AAAI 2026’ Demo Track Original Link:

Dunamu announced that its machine learning team’s paper was selected for the demo track(Demo Track) of the international artificial intelligence conference ‘Association for the Advancement of Artificial Intelligence(AAAI)’ and was successfully demonstrated.

AAAI is a prestigious conference where AI researchers from around the world present the latest technologies and research achievements. It is considered one of the top three AI conferences worldwide.

Dunamu’s machine learning team paper was chosen for the highly competitive demo track. The demo track evaluates the practical effectiveness of systems by demonstrating them in real-time, beyond theoretical presentations, and requires proving research results through application cases.

Researcher Park Hee-soo from Dunamu’s Machine Learning Team presented a paper titled ‘Market-Aware Event Timeline Summarization: Integrating Price Signals to Improve Financial News Understanding(Original: Market-Aware Event Timeline Summarization: Integrating Price Signals to Improve Financial News Understanding)’ and demonstrated their self-developed system.

This research combines news data with digital asset price chart fluctuation data to selectively provide only the core news that caused price changes.

Previously, it was difficult to identify the key news affecting actual price movements and to immediately understand the causes of price fluctuations. To address this, Dunamu’s machine learning team proposed a new model that combines large language models(LLM) and volatility(Volatility) indicators.

▲Automatically extracts digital asset-related events from news feeds ▲Selects only the events causing volatility during periods of high price fluctuation ▲The LLM summarizes relevant events and background knowledge needed to understand these events, visualizing them in a timeline with charts to help investors intuitively understand the background of chart movements.

Kim Dae-hyun, Chief Data Officer(CDO) of Dunamu, said, “This AAAI presentation not only recognizes Dunamu’s AI technology globally but also demonstrates its practical utility in resolving information asymmetry,” and added, “We will continue to leverage AI technology to provide more valuable information to investors and contribute to market transparency.”