Explore chapters and articles related to this topic
Artificial intelligence for sense making in survival supply chains
Published in International Journal of Production Research, 2023
Ajit Sharma
In Figure 13, the different components, technologies and process flow for the implementation of a COVID BOT for Twitter messages, are presented. The process starts with the acquisition of tweets that have COVID related tags into a raw text file. This Tweet data was acquired by using the Twitter API V2. Due to the societal nature of the project, Twitter allowed free access to its tweets without any rate limits. This raw tweet file was then parsed using OpenAI to extract information content from the tweet and place it into a Json file format. The Json file was then trained using OpenAI to classify each tweet into either a demand or a supply signal. These parsed and classified Tweets are next stored in an open source, cloud based database called SupaBase. In Appendix 3 a sample of the tweets parsed by OpenAI is presented with personalised data redacted for privacy reasons. This parsed and stored demand and supply information can now be presented to users in a variety of formats including online websites and mobile apps.