Discovering the Coopetition Relationship Between Agents Using Clustering

Both competition and cooperation are prevalent among entities in different fields. An approach for discovering the coopetition relationship between agents based on time series correlation is proposed in this study. First, in a minimum observation period, the correlation degree between observed values of agents is determined. Then, correlation sequences are calculated to measure the strength of the coopetition between time series of observed values. To discover the coopetition relationship between agents in a prescribed observation period, a clustering algorithm is used to resolve the positive and negative correlation sequence sets that represent the competition and cooperation relationships, respectively. The approach is applied to public transport systems in Xiamen, China to optimize transit networks and achieves higher accuracy and precision than existing algorithms.