| Project/Area Number |
17500164
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
情報図書館学・人文社会情報学
|
|---|
| Research Institution | Kanazawa Institute of Technology |
|---|
Principal Investigator |
ABE Takehiko Kanazawa Institute of Technology, College of Informatics and Human Communication, Associate Professor, 情報フロンティア学部, 助教授 (60298320)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NAMBO Hidetaka Kanazawa University, Graduate School of Natural Sci. and Tech., Assist.Pro., 自然科学研究科, 講師 (30322118)
KIMURA Haruhiko Kanazawa University, Graduate School of Natural Sci. and Tech., Professor, 自然科学研究科, 教授 (60141371)
|
|---|
| Project Period (FY) |
2005 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥3,200,000 (Direct Cost: ¥3,200,000)
Fiscal Year 2006: ¥1,300,000 (Direct Cost: ¥1,300,000)
Fiscal Year 2005: ¥1,900,000 (Direct Cost: ¥1,900,000)
|
|---|
| Keywords | Marketing / Decision Support System / People Counting / Gender Determination / Artificial Intelligence / Agent / Infrared Sensor / Management Information System / 意思決定支援 / エリアスキャナ |
|---|
| Research Abstract |
In this research, we executed the foundation of research of intelligent marketing decision support system for an increase in the productivity of selling area. The research results are as follows.
1.Intelligent people counting system for selling area
We use infrared sensors and artificial intelligence technology to develop people counting system for marketing. The main functions of the people counting system are as follows.
(1)People counting
We develop an automatic people counting system using infrared sensors. The advantage of infrared sensors is that they can be used when the light is not good. Moreover, it is hardly possible that a problem about invasions of privacy will occur compared with the existing counting system using video technology.
(2)Gender identification
This research proposes a method for determining pedestrian's gender using area scanners and Bayesian network. This research treats three parameters that are height, walking speed and step, because they have sexual distinction. Bayesian network is made by these three parameters. Experimental results of identification ratio are 85.5% (male), 81.7% (female) and 84.3% (average). Unlike conventional identification systems, proposed system has the advantage of being able to protect each individual's right to privacy and being insensitive to changes in lighting conditions compared with video based systems.
2.Simulator of consumer's behavior in a store using multi-agent model
This research proposes a simulator of purchaser's flow in a supermarket using a multi-agent model. The simulator can be used as a decision support system for determining the optimum arrangement of goods shelves, which is able to lengthen each purchaser's flow in a store, since the simulator enables us to analyze purchaser's flow under various conditions.
|
