关键词: 关键词: 群智感知, 信誉评估, 逻辑回归, 参与者选择, 数据质量

Abstract: Abstract: Due to the large groups of participants and almost no limitation on acquiring and submitting tasks in crowd sensing network, it causes problems of high data redundancy and no guarantee of data quality. Based on that, this paper presents a method called Participant Reputation Evaluation Method to evaluate the quality of data submission and participant’s reputation. This method takes timeliness, distance, historical reputation, relativity and data quality into account. After evaluation, participant’s reputation ranges from zero to one, furthermore, it clustered in the 0-0.2 and 0.8-1.0 intervals. This feature makes it easier for crowd sensing network to select suitable participants, thus it can ensure the quality of sensing data. Evaluation results show that the accuracy is over 90%.

Key words: Keywords: Crowd Sensing, Reputation Evaluation, Logistic Regression, Participant selection, Data Quality