Audience Profiles: Dynamic Footfall and Brand Affinity (Points)

Locomizer’s Audience Profiles – Dynamic Footfall and Brand Affinity Products consist of two datasets built using human mobility data collected from mobile devices where users have provided explicit consent for data to be used in this way. The product contains aggregated (Footfall) and abstract (Brand Affinities) data derivatives and adheres to local privacy regulations such as GDPR. The point data for location selection and user interest (affinity) profiling is supplied by Precisely, Inc. as a Points of Interest (POI) dataset extracted from World Points of Interest – Premium product. The POI dataset is updated monthly. Footfall Dataset This dataset provides the key footfall metrics at a point level (latitude and longitude of POI + 69-meter radius) where people spend time and/or money. The metrics are built using the mobile trace data generated by a representative panel of anonymous users. This includes the observed and extrapolated number of users who spent time at the point during a specific period, the number of their signals (observations) at the point, and the footfall data derivatives like reach, dwell-time, score. These metrics are provided for different movement (pedestrian and non-pedestrian) and visitation (residents, workers, transient) modalities. This dataset covers all needs for anyone to measure the popularity of a specific area and to perform comprehensive data analysis, build attribution and predictive models, and develop applications for retail analytics, city planning, marketing/advertising, financial forecasting and risk assessment etc. Brand Affinity Dataset This dataset complements the Footfall dataset and provides the quantitative measure of the real-world interest (affinity) of the users observed at a point level (latitude and longitude of POI + 69-meter radius). This includes hundreds of potential real-world interests including (but not limited to) shopping, leisure, sport, transport, entertainment, education. eating/drinking etc grouped at the brand level like KFC, Superdrug, VUE cinema, Virgin Active Gyms etc. The user interest (affinity) profiling methodology uses the supervised machine learning algorithm described in the Locomizer’s patent “Interest Profile of a User of a Mobile Application”.

搜集汇总

背景与挑战

背景概述

该数据集由Locomizer提供，包含人流统计和品牌偏好两个部分，均基于匿名移动设备数据构建并符合GDPR等隐私法规。人流数据集提供特定兴趣点（POI）的访客指标，用于分析区域热度和支持零售、城市规划等应用；品牌偏好数据集则通过机器学习方法量化用户在POI处的真实兴趣，涵盖购物、休闲等多个类别。数据来源于Precisely的月度更新POI数据集。