China
Africa
Explore chapters and articles related to this topic
The effectiveness of social media in flood risk communication
Published in Edmund C. Penning-Rowsell, Matilda Becker, Flood Risk Management, 2019
There are some simple but unique features of SM that enables effective communication, one of which is the hashtag on Twitter. A hashtag is composed of a hash symbol (#) followed by any words that categorise the tweet (i.e. message), allowing possible the almost instant search for tweets that share a common topic (Twitter, n.d.). Such a feature is particularly useful in filtering relevant information in flood risk communication (Palen and Hughes, 2018). The hashtag #qldfloods was quickly adopted in Queensland flooding in 2011 by both citizens and government agencies to mark and share flood-related information: more than 35,000 tweets containing the #qldfloods hashtag were generated during the flash floods there, with over a third containing links to further information such as official websites or first-hand photos of the flood (Bruns et al., 2012). During Typhoon Maring in the Philippines in 2014, unified hashtags were suggested, for example #RescuePH, #FloodPH, #ReliefPH, so that people could include these labels in their tweets for more effective searches for information (UNOCHA, 2014).
Text as data
Published in Benjamin S. Baumer, Daniel T. Kaplan, Nicholas J. Horton, Modern Data Science with R, 2021
Benjamin S. Baumer, Daniel T. Kaplan, Nicholas J. Horton
Problem 5 (Easy): Wikipedia defines a hashtag as “a type of metadata tag used on social networks such as Twitter and other microblogging services, allowing users to apply dynamic, user-generated tagging which makes it possible for others to easily find messages with a specific theme or content.” A hashtag must begin with a hash character followed by other characters, and is terminated by a space or end of message. It is always safe to precede the # with a space, and to include letters without diacritics (e.g., accents), digits, and underscores.” Provide a regular expression that matches whether a string contains a valid hashtag.
Hashtag Recommendation Approach Based on Content and User Characteristics
Published in Cybernetics and Systems, 2018
Van Cuong Tran, Dosam Hwang, Ngoc Thanh Nguyen
Users can utilize the hashtag symbol (#) before a relevant keyword or phrase without whitespace in the message. The hashtags are used for different purposes such as categorizing tweets, tagging content related to special events, etc. They make tweets more easily in searching about a specific topic and facilitate conversations among the users (Kywe et al. 2012). Furthermore, hashtags can be used to collect public opinions on events and their ideas at the individual, community, or even the world level (Li and Xu 2016). The hashtag is the simplest way for users to categorize their tweets and to join a tweet to a specific topic on Twitter. However, the wide range of the topics discussed on Twitter is difficult for users to choose the relevant hashtags for their tweets. Especially, the same tweets in terms of content could converge the same topics, and the hashtags are suitable to the usage trends of the users and reflect their interest. The practical approaches can be roughly divided into two categories: content-based methods and collaborative filtering methods (Wang et al. 2014). Content-based methods take different techniques to measure the content of the tweets to find the relevant hashtags from historical tweets (Zangerle et al. 2011b; Mancilla-Amaya et al. 2010; Godin et al. 2013). However, a tweet is too short to capture the accurate semantics. Therefore, the results of measuring similar contents are of low accuracy and it is very hard to recommend right hashtags. Collaborative filtering is a popular technique for recommendation systems. It considers user preferences by building a model from a user’s past behaviors as well as similar decisions made by other users (Kywe et al. 2012; Wang et al. 2014; Nguyen and Merayo 2017). Collaborative filtering method overcomes the weakness of the content-based method; nevertheless, recommending hashtags only based on similar users leads to local decisions.