Once you feel your search string properly captures the tweets you want to analyze, it helps to make a copy of the string for future reference. Revise and retest your search result, until you you feel that it captures the data you need. Include keywords, pairs of words, relevant hashtags, and handles as appropriate. You can learn more about search options by clicking on the operators link in the Twitter search page. Once you construct a Boolean search to collect your topic network you can use the Twitter search page to test your search string ( ). Check out the operators and advanced search links. You can learn more about Twitter search operators on the Twitter search page ( ). This post will not be included: “I love cats and especially kittens.” These post will be included: “I love all kittens and puppies I love dogs and cats.” Results would include any post that contains a combination of the two nested boolean queries. Consider the following Boolean search: “(Cats OR Kittens) AND (Dogs OR Puppies).” The search will first be performed within the parentheses, and only then between them. Sometimes called nesting, parentheses add a level of organization for your Boolean search, allowing you to formulate complex search strings. Parentheses require the terms and operations that occur inside them to be searched first.

This post will not be included: “Actually, falling in love is not as simple as you may think.” In order to communicate your topic to Twitter, you will use a Boolean search query that combines these terms into a specific machine-readable format. Next, explore related keywords and hashtags using the built-in Twitter Search and third party apps such as to identify trends, popularity, and related terms. State your topic first (e.g., Soft drinks and obesity, or the name of a political party, celebrity, or product). In order to collect topic-specific Twitter network data, you must first define your topic. Twitter conversations span a wide range of topic and issues. You can use NodeXL to collect, analyze and visualize this type of social media network data from Twitter. For example, the relationships of all users who mentioned “HPV” (short for human papilloma virus) during a particular period of time creates a dataset containing a slice of the HPV topic-network. Subgroups of users are selected based on their use of a particular keyword.

As a result, discussion “communities” are much more dynamic and emergent. On Twitter, the only indication of a tweet's topic is found within its content, which includes hashtags or keywords relevant to the topic. Instead, individual users can tweet about a wide range of personal, political and social topics. Twitter is not explicitly organized into topic specific discussions, like discussion forums ( Chapter 10). Itai Himelboim, in Analyzing Social Media Networks with NodeXL (Second Edition), 2020 11.2 Defining your topic-networks: Formulating a social media monitoring query

Boolean search creator