Home -> Research -> Knowledge Acquisition -> Re-FrameNet

    Re-FrameNet is a project to build an electronic resource of common verbs for Human-Robot Interaction and automated task planning of intelligent robots/agents. The knowledge of common verbs in Re-FrameNet is extracted from FrameNet (Baker, Fillmore, and Lowe 1998) and other semantic dictionaries, and rewritten into formal expressions in a meta-language which can be employed by automated planners. Therefore, whenever a robot/agent needs knowledge of common verbs, it can search Re-FrameNet and may get the rewritten knowledge in the formal expression. More and more rewritten knowledge of common verbs will be added into Re-FrameNet in this project.

    There are many works on utilizing open-source resources for robots to understand and accomplish user tasks (e.g., Fong, Thorpe, and Baur 2003; Tenorth, Nyga, and Beetz 2010; Tenorth et al. 2011; Cantrell et al. 2012; Chen et al. 2012; Nyga and Beetz 2012; Chen et al. 2013). However, as observed in our previous efforts on enabling open-source knowledge resources for robot task planning (Chen et al. 2012), common verbs are normally not explained in these knowledge resources, so that a very large proportion of user tasks cannot be planned with the knowledge in these resources. This class of knowledge gap constitutes a bottleneck of utilizing existing open knowledge from online resources for task planning. To attack the bottleneck, definitions of common verbs are extracted from semantic dictionaries, especially FrameNet, and rewritten into a meta-language representation. We have also developed robot planners (Chen et al. 2012; Chen et al. 2013) that can use the rewritten knowledge in Re-FrameNet to generate plans for user tasks.

    You can access the linked site. The left part of that websites is Frame Index from FrameNet. When you select a frame and click it, you will see the right part showing all the meta-language expressions of the frame for a verb.
    The Re-FrameNet Project :click here

    Baker, C. F.; Fillmore, C. J.; and Lowe, J. B. The berkeley framenet project. In Proceedings of the 17th inter-national conference on Computational linguistics-Volume 1,86–90.Association for Computational Linguistics. 1998. .
    Cantrell, R.; Talamadupula, K.; Schermerhorn, P.; Benton,J.; Kambhampati, S.; and Scheutz, M. Tell me when and why to do it!: run-time planner model updates via natural language instruction. In Proceedings of the 7th annual ACM/IEEE international conference on Human-Robot Interaction (HRI-12), 471–478. 2012. .
    Chen, X.; Ji, J.; Sui, Z.; and Xie, J. Handling open knowledge for service robots. In Proceedings of the 23rd International Joint Conference on Artificial Intelligence (IJCAI-13). 2013. .
    Chen, X.; Xie, J.; Ji, J.; and Sui, Z. Toward open knowledge enabling for human-robot interaction. Journal of Human-Robot Interaction 1(2):100–117. 2012. .
    Fong, T.; Thorpe, C.; and Baur, C. Robot, asker of questions. Robotics and Autonomous systems 42(3):235–243. 2003. .
    Nyga, D., and Beetz, M. Everything robots always wanted to know about housework (but were afraid to ask). In Proceedings of the 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS-12), 243–250.IEEE. 2012. .
    Tenorth, M.; Nyga, D.; and Beetz, M. Understanding and executing instructions for everyday manipulation tasks from the world wide web. In Proceedings of the 2010 IEEE International Conference on Robotics and Automation (ICRA-10), 1486–1491. IEEE. 2010. .
    Tenorth, M.; Klank, U.; Pangercic, D.; and Beetz, M. Web-enabled robots. Robotics & Automation Magazine, IEEE 18(2):58–68. 2011. .