IntelSurv uses Large Language Models (LLMs) to achieve the ‘chat’-like functionality where users ask questions about disease surveillance. IntelSurv looks within its ‘trained’ knowledge to provide a suitable answer. A large language model (LLM) is a ‘deep learning’ algorithm that can perform a variety of Natural Language (NLP) tasks, such as question and answering, summarising information, extracting data from text and so on. ‘Deep learning’ algorithms are those using complex structures for encoding and representing information which mimic the layered structures of the human brain see as a network of neuron, or in the language of computer science, processing units. LLMs use massive datasets of text which is stored in their memory and processed in a way that can be then utilised in a variety of NLP tasks. IntelSurv uses LLMs to achieve the ‘chat’-like functionality where users ask questions about disease surveillance and IntelSurv looks within its ‘trained’ knowledge to provide a suitable answer.
There are four types of questions that IntelSurv will be able to handle:
Questions on form and fields encompass questions such as: