Research Focus

Primary research of iDSSP LAB

Our laboratory's primary research pursuits are focused on two distinct directions: "Biomedical Data Analysis" and "Financial Technology Data Analysis." Each group is dedicated to exploring application scenarios and leveraging artificial intelligence (AI) to address a wide range of challenges. The following sections provide comprehensive details about each research group.

Biomedical Data Analysis

Our primary objective is to explore practical implementations of deep learning models and associated techniques, overcoming obstacles presented by diverse biomedical data in real-world scenarios. 

We aim to uncover the intricate relationships between inputs (like gene expression, biochemical tests, and protein sequences) and their corresponding targets (such as cancer prognosis, chronic disease risk, and drug affinity). This enables precise predictions of these targets. Additionally, we focus on scrutinizing model uncertainties and enhancing interpretability. Our efforts are categorized into three subgroups.

Financial Technology Data Analysis

The importance of deep learning in financial technology cannot be underestimated. Its ability to enhance predictive capabilities, strengthen risk management, and improve decision-making efficiency for financial institutions brings added value and a competitive advantage to the industry. 

As deep learning techniques keep advancing and evolving, we can anticipate their profound impact and transformative potential within the financial sector. Our research is categorized into two directions.

Supporting Project

• 肝癌個人化風險預測模型建立及臨床驗證 

Ministry of Health and Welfare

• 基於新穎圖神經網路、半監督與多模態學習架構之端對端癌症預後分析

National Science and Technology Council

• 前瞻技術產學合作計畫-人工智慧與大數據在金融科技的應用研究

National Science and Technology Council

• 肝癌檢測聯邦式學習

Qisda

• 產學合作計畫－基於遷移學習的多模態跨域電子商務推薦系統

National Science and Technology Council & Avivid

• 高等教育深耕計畫─特色領域研究中心【子計畫一：尖端多體學和計算生物學技術】

National Taiwan University