crowdkit.aggregation.classification.kos.KOS
| Source code
KOS( self, n_iter: int = 100, random_state: int = 0)
The KOS (Karger, Oh, and Shah 2011) aggregation model is an iterative algorithm that calculates the log-likelihood of the task being positive while modeling the worker reliability.
Let be a matrix of the responses of a worker on a task .
If the worker does not respond to the task , then . Otherwise, .
The algorithm operates on real-valued task messages and worker messages . A task message represents the log-likelihood of task being a positive task, and a worker message represents how reliable worker is.
At -th iteration, the values are updated as follows:
David R. Karger, Sewoong Oh, and Devavrat Shah. Budget-Optimal Task Allocation for Reliable Crowdsourcing Systems.
Operations Research 62.1 (2014), 1-38.
https://arxiv.org/abs/1110.3564
Parameters | Type | Description |
---|---|---|
n_iter | int | The maximum number of iterations. |
random_state | int | The state of the random number generator. |
labels_ | Optional[Series] | The task labels. The |
Examples:
from crowdkit.aggregation import KOSfrom crowdkit.datasets import load_datasetdf, gt = load_dataset('relevance-2')ds = KOS(10)result = ds.fit_predict(df)
Method | Description |
---|---|
fit | Fits the model to the training data. |
fit_predict | Fits the model to the training data and returns the aggregated results. |
Last updated: March 31, 2023