2023-08-26 01:06:03
A group of researchers from the University of West Virginia (WVU) tested the new official plugin for ChatGPT called Code Interpreter.
While they found some limitations to using the plugin in scientific work with biological data, they were impressed with the additional features that can enhance the classroom learning experience.
Read more:
About the ChatGPT plugin
Assistant Professor Gangqing Michael Hu, Director of the Department of Microbiology, Immunology and Cell Biology at the WVU School of Medicine, highlighted that Code Interpreter makes programming more accessible for students in STEM fields; however, researchers found that the plugin does not have the necessary resources to be used in the area of bioinformatics, where scientists work with biological data to prioritize treatments directed at cancer and genetic diseases; Although this may be overcome in the future, the current functionality of Code Interpreter is focused on education, but does not meet the needs of biomedical research professionals; WVU researchers tested Code Interpreter in various tasks and concluded that the plugin facilitates access to programming for people without a scientific background, in addition to awakening students’ interest in data analysis; However, users need to understand how to interpret the results and know how to interact with the chatbot.
Code Interpreter also helps minimize fictional responses, known as hallucinations, by providing codes as a source or citation when requested. Although the plugin has some limitations, such as supporting only the Python programming language and lack of internet access and ability to handle large files, researchers have found positive results in Code Interpreter’s ability to convert data into graphs.
The researchers suggest some improvements for Code Interpreter, such as internet access to download genomic data, installation of specific software for bioinformatics, increased storage capacity and support for other programming languages.
Additionally, they pointed out the need for privacy and security apps to comply with regulations such as HIPAA. Despite the limitations, Professor Hu plans to introduce Code Interpreter into his course to help students learn about data visualization and hopes that future updates will expand the plugin’s use in the area of bioinformatics.
The study by WVU researchers is part of an ongoing effort to explore the possibilities of artificial intelligence (AI) and improve communication with chatbots such as ChatGPT. They are also on the lookout for new developments and features that may emerge in the field of AI programming.
Overall, Code Interpreter is seen as a promising tool for education, but it still needs to be improved to meet the specific needs of bioinformatics and biomedical research. The researchers look forward to following the plugin’s progress and exploring its innovative applications.
Have you watched the new videos on YouTube of the Digital Look? Subscribe to the channel!
#Researchers #point #pros #cons #plugin #ChatGPT