Mining Location-Based Service Data for Feature Construction in Retail Store Recommendation

Chen, Tsung Yi; Chen, Lyu Cian; Chen, Yuh Min

doi:10.1007/978-3-319-62701-4_6

Cited by 7 publications

(10 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to the work of Chen et al (2017), this study uses location-based social network data and develops a retail store recommendation method by analyzing the relationship between the user footprint and POI. According to the correlation analysis of the target area and the extraction of crowd mobility patterns, the features of retail store recommendation are constructed.…”

Section: Discussionmentioning

confidence: 99%

“…To measure the business density in an area, kernel density estimation (KDE) is used in this study. The density estimate method adopted by Chen et al (2017) is employed, which is defined as follows (Equation 2):…”

Section: Regional Relevance Feature Extractionmentioning

confidence: 99%

“…In the study of Chen et al (2017), where the Euclidean distance was used, the feature category similarity did not convey a clear meaning and yielded low performance. Therefore, we simplified the usage of category weights and multiplied the number of industrial distributions in the region by wc, as follows:where wvc represents the categories corresponding to all the locations within the analytical region.…”

Section: Regional Relevance Feature Extractionmentioning

confidence: 99%

“…3. Architecture design of retail store recommendation system According to the recommendation factors proposed by Chen et al (2017), a more integrated system is used to extract and train features for retail store recommendation. Three data layers from Chen et al are adopted as inputs of this system: the industry and preference filter provides the method for separating categories; the location and region selection describe the method for selecting candidates from the LBS dataset; two feature extraction mechanismsregional relations (RR) and HM-are proposed to explore the hidden information in LBS data for training the recommendation model.…”

Section: Related Workmentioning

confidence: 99%

See 3 more Smart Citations

On a method for location and mobility analytics using location-based services: a case study of retail store recommendation

Chen

2020

OIR

Self Cite

View full text Add to dashboard Cite

PurposeLocation-based services (LBS) have become an effective commercial marketing tool. However, regarding retail store location selection, it is challenging to collect analytical data. In this study, location-based social network data are employed to develop a retail store recommendation method by analyzing the relationship between user footprint and point-of-interest (POI). According to the correlation analysis of the target area and the extraction of crowd mobility patterns, the features of retail store recommendation are constructed.Design/methodology/approachThe industrial density, area category, clustering and area saturation calculations between POIs are designed. Methods such as Kernel Density Estimation and K-means are used to calculate the influence of the area relevance on the retail store selection.FindingsThe coffee retail industry is used as an example to analyze the retail location recommendation method and assess the accuracy of the method.Research limitations/implicationsThis study is mainly limited by the size and density of the datasets. Owing to the limitations imposed by the location-based privacy policy, it is challenging to perform experimental verification using the latest data.Originality/valueAn industrial relevance questionnaire is designed, and the responses are arranged using a simple checklist to conveniently establish a method for filtering the industrial nature of the adjacent areas. The New York and Tokyo datasets from Foursquare and the Tainan city dataset from Facebook are employed for feature extraction and validation. A higher evaluation score is obtained compared with relevant studies with regard to the normalized discounted cumulative gain index.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Regional Relevance Feature Extractionmentioning

confidence: 99%

Section: Regional Relevance Feature Extractionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

On a method for location and mobility analytics using location-based services: a case study of retail store recommendation

Chen

2020

OIR

Self Cite

View full text Add to dashboard Cite

show abstract

“…Ref. [27] used regional relevance analysis and human mobility construction to establish the feature values of retail store recommendation. Ref.…”

Section: Urban Computingmentioning

confidence: 99%

Structured Knowledge Base as Prior Knowledge to Improve Urban Data Analysis

Zhang

Chen

et al. 2018

IJGI

View full text Add to dashboard Cite

Urban computing at present often relies on a large number of manually extracted features. This may require a considerable amount of feature engineering, and the procedure may miss certain hidden features and relationships among data items. In this paper, we propose a method to use structured prior knowledge in the form of knowledge graphs to improve the precision and interpretability in applications such as optimal store placement and traffic accident inference. Specifically, we integrate sub-graph feature extraction, sub-knowledge graph gated neural networks, and kernel-based knowledge graph convolutional neural networks as ways of incorporating large urban knowledge graphs into a fully end-to-end learning system. Experiments using data from several large cities showed that our method outperforms the baseline methods.

show abstract