Ordinary people who directly consume medicines do not know or know the route through which they can obtain information about medicines, but they are getting information on medicines through unverified, unprofessional channels. Currently, the database market that provides drug information provides long drug information, and it is inevitably difficult for the general public who lacks specialized knowledge to understand.
Therefore, Professor Myung-wan Koo’s research team in the Department of Computer Science and Engineering at Sogang University developed a drug information acquisition site and chatbot that is user-friendly and provides verified information. The research team provides a method for medical consumers to more easily access and understand drug information by using a highly accessible chatbot and Once Global’s processed drug information.Natural language generation technology (NLG) refers to a technology that understands organized information and converts it into a human-readable language. If the amount of information is complex and unfamiliar, such as pharmaceuticals, it may be difficult to understand even if the general public can access the information itself. Natural language generation technology provides information in the form of a language used in daily life according to the type of information desired by the user. This includes inquiring drug-related information and consulting on medical-related information with a chatbot.
Deep neural networks can learn relationships between information that appear relatively uncorrelated. It is being used as a core model for deep learning. The deep neural network includes a deep trust neural network (DBN) based on unsupervised learning, a deep autoencoder, etc., depending on the algorithm, and is synthesized for processing two-dimensional data such as images. There are a convolutional neural network (CNN) and a recurrent neural network (RNN) for processing time series data.
It refers to the drug database and generates natural language sentences appropriate to the situation in the conversation situation like the picture above, and it shows strength in general and natural sentence generation.
Once Global Co., Ltd. hoped to transfer the technology, and in October 2019, signed a contract with the Sogang University Industry-Academic Cooperation Foundation to transfer ‘ConnetDI, Know-How Technology for Natural Language Generation of Medicine Information with Patients’.
Once Global was able to create a drug information chatbot using SLU, DM, and NLG, which are conversational systems developed by the research team that identify users’ queries and provide appropriate answers.