S-MALL Future Work

From Visual Analytics and Applications
Revision as of 15:44, 6 August 2017 by Zpchiam.2016 (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Turning Concrete Malls into Smart Malls (S-MALL):
A web-based analytics application for visualizing and mapping in-mall customer journeys and shopping behaviours

Team S-MALL: Chen Yun-Chen | Chiam Zhan Peng | Zheng Bijun

Overview

Application

Poster

User Guide

 

APPLICATION

Movement Analysis Member Analysis Association Analysis
T3 Appoverview.png

USER GUIDE

Part1: Movement Analysis

Screenshot Guidance
T3 Part1 control.PNG
  • "Plot by" and "Segment by" dropdown are used for line chart. Select "Segment by" will show cycle plot accordingly.
  • "View" and "View hour" dropdown are for selecting date and hour respectively and display the inter-flow movement and hexbin map for that selected date and hour.
T3 Trellis.JPG
  • Hover on any points on the line chart to display the tooltip for detailed information.
T3 Part1 chord.PNG
  • Hover on the chorddiagram to display the tooltip for detailed information.
  • Example: There are 61 customers moving from 3F to 2F and 57 customers moving from 2F to 3F at selected time.
T3 HexbinMap.jpg
  • Zoom in by selecting area with left click hold and release as required and click on any hexbin for the selected hexbin to be captured.
  • Click on "Show Detail" button to show the distribution of the selected hexbin. Members and Non-members count will be displayed.
Error creating thumbnail: File with dimensions greater than 12.5 MP
  • Click on the bar chart to show detailed information such as their Device ID and number of counts.
T3 Part1 table.jpg
  • Count means the customer stays in selected hexbin for how long (min) in selected time interval.


Part2: Member Analysis

Screenshot Guidance
T3 Part2 control.PNG
  • Selection All in Operating Hours: All members
  • Selection Normal in Operating Hours: Members who only appear in operating hours.
  • Selection Outside in Operating Hours: Members who appear out of operating hours.
  • Exclude dwell time less then: This control panel is for Shopper's Journey. User can exclude those records which only stay in a particular floor for a few minutes.
  • Profile Count by: This control panel is for member distribution. Non-Unique Customers: Count by records. Unique Customers: Count by member ID.
T3 Part2 sunburst.PNG
  • Hover to see the path and relevant statistics.
  • Example: There are 23 of 2.2k journeys which start from FLOOR M and go to FLOOR G then go to FLOOR 2.
T3 Part2 treemap.PNG
  • Click on floor level to drill down to store level.
  • Click on any store to show the relevant boxplot for selected floor as following.
T3 Part2 boxplot.PNG
  • Select records you are interested in to show the relevant distribution(bar chart) as following.
T3 demobar.JPG
  • Get the distribution of records in box selection.

Part3: Association Analysis

Screenshot Guidance
T3 Part3 control.PNG
  • Set date range using radio button to see the plot of different months.
  • Set parameters (support/confidence/min items) to generate valid associations rules and render plot forquadrant and network.
T3 Part3 bar.PNG
  • Investigate popular shopping path of members based on floors.
  • Hover over the bar to fade out non-relevant rules in the quadrant as following.
T3 Part3 node.PNG
  • Visualize the associations among departments.
  • Hover over the network to fade out non-relevant rules in the quadrant as following.
T3 Part3 qua.PNG
  • Analyze the competitive position of different rules based on set parameters.
Part3 table.PNG
  • Detail information of different rules based on set parameters.

Please be patient.