Improving data quality with an accumulated reputation model in participatory sensing systems

Area of Science:

• Computer Science
• Ubiquitous Computing
• Data Science

Background:

• Mobile devices enable participatory sensing for environmental data collection.
• Corrupted data from participants can compromise the reliability of sensing results.
• Existing methods struggle with inaccurate data in large-scale sensing networks.

Purpose of the Study:

• To introduce a novel model for improving the accuracy of sensing results in participatory sensing.
• To address the challenge of unreliable data caused by inexperienced or malicious participants.

Main Methods:

• Proposed the Accumulated Reputation Model (ARM).
• ARM computes and accumulates participant reputation based on sensing data.
• Reputable participants' data contribute more significantly to the final sensing result.

Main Results:

• ARM effectively improves the accuracy of sensing results.
• The model demonstrates robust performance even with a high proportion of unreliable participants.
• Accurate sensing results are achievable even in challenging scenarios.

Conclusions:

• The Accumulated Reputation Model is a viable solution for enhancing data quality in participatory sensing.
• ARM offers a scalable and effective approach to mitigate the impact of corrupted data.
• This model ensures more dependable environmental insights from mobile sensing networks.