Optimal worker selection method and system based on reverse knapsack in abnormal data crowdsourcing detection

Abstract

The invention belongs to the technical field of crowdsourcing networks and particularly relates to an optimal worker selection method and system based on a reverse backpack in abnormal data crowdsourcing detection. The method comprises steps of determining a bidding worker set according to an abnormal data detection task and crowdsourcing worker information in a crowdsourcing network; obtaining a plurality of worker combinations according to abnormal data detection task requirements, and adding the worker combinations into a candidate crowdsourcing worker set; obtaining the prediction reliability of an abnormal data detection result by subjecting a crowdsourcing worker trust value to weighted calculation, by taking the total bidding price of the crowdsourcing workers as a constraint condition, the candidate crowdsourcing worker set being as a backpack article, the crowdsourcing worker trust value as the weight of the backpack article, and an expected reliability threshold value as the backpack capacity, modeling a worker selection problem in abnormal data detection as a reverse backpack model; and selecting an optimal worker combination by solving the reverse backpack model. According to the method, detection task expected reliability and the bidding price are comprehensively considered to select the corresponding optimal worker combination so that the labor and time cost is saved.

Description

Technical field [0001] The present invention belongs to the field of prefectural network technology, and in particular, the present invention relates to an exception data preparation method and a system based on reverse backpack selection method and system. Background technique [0002] Abnormal data detection is widely used in financial, aerospace, medical and other fields, which guarantees authenticity and availability of data. Although there are currently many research results on large data, there have been many abnormal data detection methods based on machine learning or mathematical statistics, but social computing, social public safety, etc., are highly demanding on large data computing performance. . When processing data abnormality detection problems in the data scale, the single-node power is not sufficient to meet the timeliness of user needs, resulting in low detection efficiency, and the reliability of the detection cannot be guaranteed. Therefore, how to efficiently ...

AI Technical Summary

Problems solved by technology

[0003] However, most of the existing crowdsourcing worker selection methods for reliability and cost only consider the correlation between crowdsourcing worker skills and task types to ensure the reliability of crowdsourcing task results

In abnormal data detection, the reliability of crowdsourcing task results is not predicted by the combination of worker trust value or historical behavior and worker selection scheme

Moreover, in most of the current research, the crowdsourcing tasks are only completed by one crowdsourcing worker

Complex and massive abnormal data detection tasks usually require multiple crowdsourcing workers to complete a single detection task. There is no worker selection method that comprehensively considers the reliability of task results when multiple crowdsourcing workers are selected to complete a single detection task.

In cost-based crowdsourcing research, the common research direction is mostly to consider the moving distance of crowdsourcing workers or the consumption of battery and traffic in mobile crowdsourcing. There are few studies on the cost of crowdsourcing task issuers, which leads to task issuers No control over the cost of crowdsourced worker compensation and the reliability of crowdsourced task results

Images