Crowdsourcing quality improvement method based on integrated TSK fuzzy classifier
A technology of fuzzy classification and sub-classifier, applied in the field of crowdsourcing system, can solve problems such as increasing the economic cost of crowdsourcing, and achieve the effect of reducing cost
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0066] 1. TSK fuzzy subclassifier TSK-noise-FC with label anti-noise ability
[0067] According to the TSK classifier and the label noise problem in the crowdsourcing environment, we improved the TSK fuzzy classifier and proposed a TSK fuzzy classifier TSK-noise-FC with label anti-noise ability. The output of the classic TSK fuzzy classifier can be transformed into the following linear regression problem
[0068]
[0069] In traditional regression problems, the goal is to minimize a quadratic loss function. According to the principle of structural risk minimization and considering the actual situation of crowdsourcing, the answers given by crowdsourcing workers include definite and uncertain. Here, we mark the uncertain label as 0, so y∈{-1 ,+1,0}, the parameter learning of formula (6) can be expressed as the following form:
[0070]
[0071] Among them, the parameter a g Denotes the consequent parameters to be learned, ξ i is the prediction error, and ρ represents a...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 



