EventGO System Tasks

The automatic extraction of meetup event information from the Internet can substantially enhance the ease of discovering activities. However, when event organizers share event details across various platforms, including their respective websites, social media, and specialized event-listing sites, different types of events may be distributed across multiple channels. Consequently, users are often compelled to search through various avenues to find events in which they are interested.In this study, we propose a cost-effective framework for mining events from event organizers' websites instead of crawling the entire web. We begin by utilizing Facebook Events and event-based social networks to identify potential event-organizer websites and extract webpages containing the latest news listings from these sites. Subsequently, we apply unsupervised data-extraction techniques to segment the input pages into individual messages. Finally, we conduct structured event extraction to extract event information from the individual message pages. Our experimental results, obtained between July 13, 2023, and November 22, 2023, indicate that our event-mining framework successfully collected 7,012 web scraping APIs by automatically identifying URLs on the Web. Among 429,266 published links on message source pages, 134,902 messages are acquired following boilerplate removal. Subsequently, 33,988 event messages are identified by the event-detection module. Finally, the event-extraction module extracts 33,577 events.