In this study, we focus on an itemset-frequency (IF) sketch of a given transactional data stream. Let I be the universe of items and S = ?e1, . . . , en? a stream where each ei ⊆ I is a transaction appearing at timestamp i. The proposed sketch can answer an approximation of the frequency f(q) for any query q ⊆ I within a bounded error ratio ?. This paper provides a deterministic algorithm that constructs an ?-deficient IF sketch in the online manner.

RoboCupRescue Simulation serves as a benchmark for evaluating disaster rescue agents' strategies. Rescue strategies for disaster rescue agents must adapt to various disaster environments, including situations where agents cannot communicate with each other. In this study, we focus on the response threshold method for selecting rescue tasks in such environments. We evaluate its effectiveness by comparing the benchmark scores of the response threshold method and the alternative. The results indicate that the response threshold method outperforms the alternative in certain environments.

In recent years, AI-powered services have been proposed in the tourism sector. This paper proposes an improved decentralized federated learning method that utilizes digital signage placed at various tourist spots for improving model accuracy. In the proposed method, digital signage stores multiple models received from nearby mobile devices and integrates them to construct a more accurate model. Evaluation results show that the best models built on each mobile device using the proposed method achieves an average accuracy of 0.712, improving by 0.078 compared to conventional methods using only opportunistic communications between mobile devices. Additionally, the number of communications during training in the proposed method was 6.79 per mobile device, reducing by 15.9% compared to the method that utilizes both P2P communications and communications with digital signage.

RoboCupSoccer simulation is a simulation of soccer played by robots. As in real soccer, it is important to understand the strategy of the opponents and select an appropriate strategy. However, it is difficult to understand the strategy in real time, and it is common to select a strategy from pre-defined strategies based on the team name, but it is not possible to deal with unknown teams. Therefore, we propose a method for analyzing and classifying strategies based on the opponent's behavior using a mixture model.