Method for predicting target area based on social media sign-in, terminal and storage medium

Abstract

The invention discloses a method for predicting a target area based on social media sign-in, a terminal and a storage medium, and the method comprises the steps: obtaining a geographic position labelin a social media sign-in record, and generating an area feature vector according to the geographic position label; generating a multi-dimensional region feature vector according to the area feature vectors of all the areas, training a machine learning model according to the multi-dimensional area feature vector and a target area correlation vector generated based on the multi-dimensional area feature vector, and taking the trained machine learning model as a prediction model; and obtaining a to-be-predicted area feature vector, sorting to-be-predicted areas through the prediction model and the to-be-predicted area feature vector, and determining a target area in the to-be-predicted areas according to a sorting result. According to the method, the task for determining the resident region of the user is abstracted into a sorting problem, the regions accessed by the user are sorted by using the machine learning model, and finally, the resident region of the user is successfully predicted. 本发明公开了基于社交媒体签到预测目标区域的方法、终端及存储介质，通过获取所述社交媒体签到记录中的地理位置标签，根据所述地理位置标签生成区域特征向量；根据所有区域的区域特征向量生成多维区域特征向量，根据所述多维区域特征向量以及基于所述多维区域特征向量生成的目标区域相关性向量对机器学习模型进行训练，将训练完毕的机器学习模型作为预测模型；获取待预测的区域特征向量，通过所述预测模型以及所述待预测的区域特征向量对待预测的区域进行排序，根据所述排序结果在所述待预测的区域中确定目标区域。本发明将确定用户常驻区域的任务抽象为一个排序问题，并使用机器学习模型对用户访问过的各区域进行排序，最终成功预测用户的常驻区域。