Optimal chain store location method based on extreme learning machines

Abstract

The invention discloses an optimal chain store location method based on extreme learning machines. The method comprises steps that firstly, region division is carried out according to city road networks, secondly, characteristics are constructed through utilizing social medias and sensor data of each region, and characteristic fusion is carried out through utilizing an automatic extreme code machine; and lastly, according to data distribution difference factors among cities, a stack extreme learning machine and an adaptive field extreme learning machine are flexibly employed to train the optimal chain store location method. The optimal chain store location method is advantaged in that the automatic extreme code machine is utilized to carry out fusion of different view data, data of other sources can be effectively expanded, moreover, based on the adaptive field extreme learning machine technology, the relatively good effect on small cities having relatively small samples is realized.

Description

technical field [0001] The invention belongs to the field of data mining and urban computing, and in particular relates to an optimal chain store site selection method based on an extreme learning machine. Background technique [0002] The optimal location of shops can bring strong economic benefits. An optimized location usually attracts more customers than a random location. For example, a new coffee shop can be opened near the intersection of roads. Usually, the intersection has convenient traffic and better passenger flow. However, factors such as traffic congestion may also have a negative impact on this location. With the continuous development of urban computing, it is possible to use the massive data in the city to optimize the location of shops. Traditional optimal location selection usually uses the characteristics of an area itself, such as the flow of people, purchasing power, traffic, etc., to construct a model. However, people do not necessarily stay in cert...

AI Technical Summary

Problems solved by technology

However, in some areas of the city or for some small cities, some data such as social media are quite sparse, and models built solely using data from these areas or the city itself cannot achieve good results

In addition, the traditional learning model needs to optimize a large number of parameters, and the training efficiency is low.

Images