Semanalytics is a fun way to visualize and analyize what hashtags, words, and phrases (i.e., query) are trending (or not) in social media. Besides being a lot of fun, Semanaltyics' bubble visualizations helps you to understand the context in which your query is being used in social media by counting and drawing a 'Bubble Graph' of the number of times your query appears in a large sample (>= 500) of social media posts.
Our first implementation leverages the Twitterverse, with future versions incorporating other social media sites. Simply type in a search query, together with the minimum (default: 5) and maximum (default: 100) times the word should appear in your sample. Then start exploring and refining your searches. Note that you can used advanced query techniques to narrow down your search. For example, warriors - 'golden state' will search on the word warriors while excluding golden state from your search.
To save your search queries, create an account and login before searching. You'll find your last 10 searches on your profile page. You can drill down further on your search by clicking on any node in the bubble
A bubble cloud of individual words/hashtags that are found in a twitter api query. The larger the word, the more often it occurs. The query spans a set of time, and has some basic filtering/processing to remove utility words (e.g., the, and, is) and twitter related noise (e.g., hash tags). We use two models. User, Query.
Relationships look like this.
- A user has many queries, a query belongs to a user
email - String
password_digest - String
time - Timestamp
query - String
min word count - Integer
max word count - Integer
user_id - Integer
Bubble Cloud Visualization - d3.js Twitter Queries - twitter REST