Andreas Kontogiannis

Work Experience:
Research Intern in National Centre for Scientific Research “Demokritos” (2020)

MEng in Electrical and Computer Engineering, Electrical and Computer Engineering, NTUA, 2015-2021

Research Interests:
Reinforcement Learning, Machine/Deep Learning, Natural Language Processing, Data Science.

The speaker's profile picture


A Personalized Artificial-Intelligence-enabled Method for Efficient Research in Ethnopharmacology

Traditional medicine (ethnopharmacology), is now widely considered as a promising alternative medicine for complementary treatment of the well-known western medicine. Thousands of plant species, many of which have medicinal values, are still widely used as alternative or complementary therapy (mainly in developed countries). The documentation of indigenous knowledge on the use of plants in Eastern Europe (in our case southern Balkans, Greece and coastal zone of Asia Minor), is a very challenging task.
Historically significant cultural influences that this region has received from the West and the East and at the same time, the complex linguistic differentiation of the name of compounds / drugs in different regions of the same country requires a linguistic differentiation to confirm their action. The result is that today we have undefined and unused database filtered by time.
A significant volume of traditional knowledge regarding ethnopharmacology is shared through online means: blogs, social media, dedicated websites. However, the volume of sources that needs to be monitored, the variety of formats used, the different quality of language use across sources, present some of what we call “big data” challenges in the analysis of this data.
To face these challenges, we propose the use of Information Technology (IT) and Artificial Intelligence (AI) to: allow efficient discovery, collection and digitization of existing knowledge, through information retrieval and natural language processing methods; support the experts for the validation of this knowledge, utilizing machine learning as decision support helpers; interlink resources in a web of knowledge related to ethnopharmacology, through intelligent focused crawling agents empowered by deep reinforcement learning.
In the presentation, we will:
• Discuss the need for an open eco-system and software infrastructure for the empowerment of ethnopharmacology research
• Overview our vision for the architecture of the infrastructure
• Describe the first steps towards implement it, with the power of AI methods.