Medication prediction method and system based on medical big data

Abstract

The invention provides a medication prediction method and system based on medical big data, and the method comprises the steps: predicting a target user who does not buy a new drug after the drug is used up based on a trained binary classification model which is obtained through the training according to drug buying data of a historical user, classifying and screening the data of which the purchase time does not reach the export time and the data of which the purchase time exceeds the data export time; and associating the target user with the drug data, and pushing an association result to thetarget user. According to the medication prediction method and system based on the medical big data provided by the embodiment of the invention, the next purchasing time of a user purchasing a chronic disease medicine can be predicted more accurately by analyzing the purchasing rule of the chronic disease medicine in a targeted manner, and then the user who has used up the medicine but has not purchased again is found out; and through the proportion of each feature obtained in model training, the user who is most likely to return to a pharmacy for purchase can be calculated more accurately.

Description

technical field [0001] The present invention relates to the field of big data technology, and more specifically, to a method and system for medication prediction based on medical big data. Background technique [0002] At present, there are a lot of patients with chronic diseases in my country. There are more than 300 million people with chronic diseases in the country. The number of deaths from chronic diseases has accounted for 80% of the deaths due to diseases in my country. Most chronic diseases require long-term medication, and the average age of people suffering from chronic diseases is relatively old. It is often the case that the drugs have been used up and they forget to buy again. Interruption of medication will seriously threaten the health of patients. As a result, users will be lost and drug sales will be reduced. Therefore, if the pharmacy can predict the medication time of users who have purchased chronic disease medicines in the pharmacy and remind them at t...

AI Technical Summary

Problems solved by technology

[0003] At present, the general commodity purchase prediction is made by analyzing all the commodities purchased by the user and the user's browsing and clicking behavior. However, this prediction method is not suitable for the prediction of chronic disease drugs. It has the following shortcomings: 1. There is a big difference between them. Chronic disease medicine is a special commodity, and its purchase population, purchase frequency and purchase time interval are different from other commodities; 2. The general commodity purchase prediction can predict the purchase of other commodities through the purchased commodities , while chronic disease drugs generally only buy one type all the time; 3. General-purpose product purchase predictions can have multiple data sources, while chronic disease drug purchase prediction data only comes from purchase behavior; 4. User privacy protection, general-purpose products The purchase prediction may use the user's private data to make better predictions, while the chronic disease drug purchase prediction only uses the user's consumption information, which pays more attention to privacy protection

Images