The overall goal of the intent classification task is to develop a machine learning model that accurately identifies the intent behind natural language statements within the enteral tube feeding domain. The objective is to improve the efficiency and effectiveness of customer service interactions by automatically routing input sentences to corresponding flow’s dialogue initiator. To achieve this goal, the model must be trained on a large dataset of labeled examples that represent the various intents within the domain. The model should also be able to handle variations in language, syntax, and context that are common in natural language queries as well as spelling errors and typos. Ultimately, the success of the model will be measured by its ability to accurately classify statements with a good accuracy and f-score results which have been calculated with a batch testing procedure.
Technically, we can rephrase the problem as “Single label text classification task using transformers model for the purpose of intent classification”
Our classifier is a nutrition type and method agnostic.classifier model doesn’t focus on those 2 inputs. Attention is focused on the essential intention of the user independent from the nutrition type and method. But after the classification step we are using nutrition type and method information to map the intention to more specific user flows.
