In every research conducted for the healthcare sector, opinions, concerns, questions, and complaints about a disease and its treatment methods, including medicines, vaccines, and operations, are of great importance. Therefore, many companies focus on subject-based research. However, the source of the data is as important as its content. For example, the reliability of research depends on the identification of the source of a conversation about breast cancer (whether it is coming from the patients, the caregivers, or someone who has no idea and experience about the disease), and classification of the data accordingly.
"Patient" model is provided to the companies in the healthcare industry that want to understand patients and their families better, respond to their requests, and reduce complaints. The model can classify contents about a disease and its treatment methods, including medicines, vaccines, and operations by the source as patient and caregiver in the fastest and effortless way.
About Training Set
The "Patient" model was trained with a dataset from Kimola's past eight pieces of research for the companies in the healthcare industry. Dataset labeled as "patient" and "caregiver" consists of chosen 3,280 contents. Before the last version of this dataset was reached, company names were cleared from the data to prevent proper names from being associated with the categories and to be mislabeled. Unrelated contents from the specified disease or treatment were removed from the data.
The training dataset was tested with seven different data consisting of comments, requests, questions, and complaints about a disease and its treatment methods. Each wrongly labeled row was exemplified until correct labeling was achieved.