Abstract
A general framework allowing prediction of consumer preferences from a knowledge network was described. The method was based on reconstruction of individual tastes of customers from a sparsely connected network of their opinions on products, services or each other. A Gaussian model was used to study the formation of customer opinions. The prediction power was shown to depend on the ratio between the known and unknown opinions.
| Original language | English (US) |
|---|---|
| Article number | 248701 |
| Pages (from-to) | 2487011-2487014 |
| Number of pages | 4 |
| Journal | Physical review letters |
| Volume | 87 |
| Issue number | 24 |
| State | Published - Dec 10 2001 |
| Externally published | Yes |
ASJC Scopus subject areas
- General Physics and Astronomy