This task addresses two specific interaction issues relevant to AIs. Firstly, multimedia contents commonly used in various applications (e.g., web applications and educational software) increase the interactivity. On the other hand, accessing multimedia contents on a mobile device increase the cost of downloading mobile data and require a high bandwidth for smooth playback. Many AIs in the Northern Plains live in a rural and isolated areas, which may not have a fast Internet connection for downloading multimedia contents and they are more concerned about the cost for data usage rather than the rich multimedia information. Therefore, it is desirable to adjust the form of information presentation based on the interaction context and user’s personal preferences. Secondly, a successful self-management system depends on the daily continuous data input for tracking the person’s activities and health conditions. However, users can easily get tired of daily data entry. The mobile application especially makes long-term data input challenging due to the lack of keyboard. By forming a virtual community through social networking, the interaction among users can encourage them to persist on the data entry. In addition to the persistency, we also consider the efficiency of data entry for AIs, who in general may have a low computer skill and may not be familiar with touch-based technologies. Therefore, it is desirable to provide multimodal interface, which provides multiple redundant input modes for data entry. Such a multimodal interface not only allows the AIs to input the data with a modality that they are familiar with, but also potentially benefits the long-term data entry (e.g., users can switch between different modalities to increase the interest).
Details of this task are still under development