RAPID: Text Message-Based Infrastructure for Emergency Response

海地7.0级地震动员了全世界来支持救援工作，特别是通过网络空间的新用途。 尽管地震破坏了海地的大部分通信基础设施，但海地的互联网连接是强大的，因为大多数海地互联网服务提供商使用卫星而不是受损的海底光纤电缆连接到互联网。因此，救援人员、普通公民和非政府组织（NGO）广泛使用Twitter来传播和分享有关需求、事件和救援行动的信息。 然而，这些推文并不容易聚合成有意义的主题，以便于向迫切需要信息的人提供。该研究项目的目标是与非政府组织协调机构（NetHope）合作开发和部署一个可重复使用的基于文本消息的基础设施，该基础设施按主题和地区分类和汇总有关海地救援行动的多语言推文和文本消息，以便非政府组织，海地幸存者及其朋友和家人可以轻松订阅。该研究项目利用为实体提取和主题分类开发的现有软件，以便该系统可以快速开发和部署，以满足海地时间紧迫的需求。 系统的评估包括来自NetHope的反馈以及关于系统使用和性能的定量度量。 该研究的学术价值在于开发了利用上下文信息从短消息中自动识别主题和地理位置的集成工程解决方案，为设计下一代应急响应和灾害管理信息基础设施奠定了重要基础。欲了解更多信息，请参阅项目网页：URL：http://www.emerse.ist.psu.edu

The 7.0 Earthquake in Haiti has mobilized the entire world to supportthe relief effort, especially through novel uses of the cyberspace. Even though the earthquake damaged much of the communication infrastructure in Haiti, Haiti's Internet connectivity is robust because most Haitian Internet Service Providers use satellite, ratherthan the damaged undersea fiber optic cable link, to connect to the Internet. Consequently, relief workers, regular citizens, and non-governmental organizations (NGO's) have used Tweets extensively to spread and share information about the needs, events, and relief operations. However, these tweets are not easily aggregated into meaningful topics for ease of delivery to people who critically need the information. The goal of this research project is to develop and deploy, in collaboration with an NGO coordination body (NetHope), a reusable text message-based infrastructure that classifies and aggregates multi-lingual tweets and text messages about Haiti relief operations by topics and regions so that they can be easily subscribed by NGO's, survivors in Haiti, and their friends and families. The research project leverages existing software developed for entity extraction and topic classification so that the system can be developed and deployed quickly to respond to the time critical needs at Haiti. The evaluation of the system includesfeedbacks from NetHope as well as quantitative metrics about theusage and performance of the system. The intellectual merit of the research is developing integrated engineering solution to automatic identification of topics and geo-location from short-messages utilizing contextual information, which lays important foundation for designing the next-generation information infrastructure for emergency response and disaster management. For further informationsee the project web page: URL: http://www.emerse.ist.psu.edu