Mining Shopping Patterns for Divergent Urban Regions by - - PowerPoint PPT Presentation

mining shopping patterns for divergent urban regions by
SMART_READER_LITE
LIVE PREVIEW

Mining Shopping Patterns for Divergent Urban Regions by - - PowerPoint PPT Presentation

Mining Shopping Patterns for Divergent Urban Regions by Incorporating Mobility Data Author : Haishan Liu,Tianran Hu, Ruihua Song, Yingzi Wang, Xing Xie, Jiebo Luo Source : CIKM 16 Advisor : Jia-Ling Koh Speaker : Chia-Yi Huang


slide-1
SLIDE 1

Mining Shopping Patterns for Divergent Urban Regions by Incorporating Mobility Data

Author : Haishan Liu,Tianran Hu, Ruihua Song, Yingzi Wang, Xing Xie, Jiebo Luo Source : CIKM’ 16 Advisor : Jia-Ling Koh Speaker : Chia-Yi Huang Date : 2017/08/29

slide-2
SLIDE 2

Outline

▸ Introduction

▸ Method ▸ Experiment ▸ Conclusion

2

slide-3
SLIDE 3

Introduction

3

▸ Motivation

slide-4
SLIDE 4

Introduction

4

▸ Shopping Pattern : <Tables, Photography,

Digital accessories,…>

▸ Mobility Pattern : <School Dormitories,

School Libraries,…>

▸ Region

slide-5
SLIDE 5

Introduction

5

▸ Market Basket Analysis

▸ Consumers usually have demands for a group

  • f products

▸ People’s demands are highly related to their

lives

slide-6
SLIDE 6

Outline

▸ Introduction

▸ Method

▸ Experiment ▸ Conclusion

6

slide-7
SLIDE 7

7

Method

slide-8
SLIDE 8

8

Method

slide-9
SLIDE 9

▸ Shopping Patterns Extraction

▸ Browsing log of 


shopping website

▸ NMF ▸ Ps ▸ Coefficient Matrix ▸ Rs : sum up the weight of location 


in the same region.

9

Method

slide-10
SLIDE 10

▸ Mobility Patterns Extraction

▸ User ID, POI category,


latitude, longitude

▸ NMF ▸ Pm ▸ Coefficient Matrix ▸ Rm : sum up the weight 


  • f user in the same region.

10

Method

slide-11
SLIDE 11

▸ Collective Matrix Factorization

11

Method

slide-12
SLIDE 12

▸ Collective Matrix Factorization

▸ d

12

Method

slide-13
SLIDE 13

▸ City-wide Interaction Regularization

▸ Gravity Model ▸ City-wide Interaction Regularization

13

Method

slide-14
SLIDE 14

▸ Gravity Model

▸ Oi , the number of individuals leaving region i ▸ Dj , the number of individuals arriving at region j ▸ The distance between two regions,

14

Method

slide-15
SLIDE 15

▸ City-wide Interaction Regularization

▸ The more interactions between two regions, the

more alike their lifestyles are.

15

Method

slide-16
SLIDE 16

▸ Hybrid Model

▸ Combine the collective matrix factorization

and interaction regularization

16

Method

slide-17
SLIDE 17

Outline

▸ Introduction ▸ Method

▸ Experiment

▸ Conclusion

17

slide-18
SLIDE 18

18

▸ Data Set

▸ Online browsing dataset : 250 product

categories

▸ Check-in dataset : 1.5 million check-in data, 200

POI categories

▸ Bus dataset : 3 million bus-trip records ▸ Taxi dataset : 1.9 million taxi-trip records

Experiment

slide-19
SLIDE 19

19

▸ Baseline

▸ Matrix Factorization (MF) ▸ Collective Matrix Factorization (CMF) ▸ CMF with neighboring information

Experiment

slide-20
SLIDE 20

▸ Evaluation

▸ 


20

Experiment

slide-21
SLIDE 21

21

Experiment

slide-22
SLIDE 22

22

Experiment

slide-23
SLIDE 23

23

Experiment

slide-24
SLIDE 24

24

Experiment

slide-25
SLIDE 25

25

Experiment

slide-26
SLIDE 26

Outline

▸ Introduction ▸ Method ▸ Experiment

▸ Conclusion

26

slide-27
SLIDE 27

Conclusion

▸ Connecting the shopping patterns with the

mobility patterns in a region.

▸ Modeling the interactions between regions, and

leverage the information of known regions to infer the shopping patterns in unknown regions.

27

slide-28
SLIDE 28

THANK YOU