The invention relates to a parking space prediction recommendation algorithm based on big data. The method comprises the steps of: constructing a prediction model by utilizing statistical data of an existing parking lot parking space number, performing standardization processing on parking space original data by adopting a normalization method, and standardizing the data to be within the range; after the network is trained by using a training sample, checking the generalization ability of the network by using a test sample; making clear the main consideration factors of a parking driver when selecting a parking space; constructing a parking space comprehensive index by combining the walking index of each parking space and the walking index, and selecting an optimal parking space accordingto the parking space comprehensive index; and solving the model by adopting a genetic algorithm so as to select an optimal parking space for a vehicle owner. The recommendation method has the advantages that: one-stop intelligent parking closed-loop management of parking space query/prediction, parking space reservation, parking space recommendation, reverse vehicle searching and dynamic charging/quick payment is provided for a C-end vehicle owner, and parking space prediction and recommendation play a decisive role in user experience.
本发明涉及一种基于大数据的车位预测推荐算法,利用已有的停车场车位数的统计数据来构造预测模型,采用归一化的方法,对车位原始数据进行标准化处理,将数据范围规范至范围内;使用训练样本对网络训练后,用测试样本检验网络的泛化能力;明确泊车驾驶者在进行泊车位选择时的主要考虑因素;结合每个泊车位的步行指标与步行指标构造泊车车位综合指标,根据车位综合指标选择最优泊车位;采用遗传算法对模型进行求解,进而为车主选出了最优停车位。本发明的优点在于:本发明中的推荐方法,为C端车主提供打造“车位查询/预测、车位预定、车位推荐、反向寻车、动态计费/快捷支付”的一站式智能停车闭环管理,车位预测与推荐对用户体验起着决定性的作用。
Parking space prediction recommendation algorithm based on big data
一种基于大数据的车位预测推荐算法
2020-05-22
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Parking strategy recommendation based on parking space availability data
European Patent Office | 2019
|