Interactive Network Graph

Graph Help

Explore how information spreads across Twitter with an interactive network graph using the OSoMe decahose archive.

Try out one of these great queries:
Network Type
Hashtag
Co-occurrence
User
Retweet
Mention
{{graphctrl.graph.stats.filtered_node_count}}
Nodes
{{graphctrl.graph.edge_count}}
Edges
K Core Filtered for Performance
{{graphctrl.graph.stats.k}}
K
{{graphctrl.graph.stats.original_node_count}}
Orig Nodes
{{graphctrl.graph.stats.original_edge_count}}
Orig Edges

{{spinnerCtrl.globals.spinner_message}}

User {{modalctrl.user.username}}

Is mentioned by:
Is retweeted by:
Is quoted by:
Has mentioned:
Has retweeted:
Has quoted:

To share, copy and paste the following URL:


{{modalctrl.globals.error}}

:D

{{modalctrl.globals.success}}

If you find these tools useful, please cite the following article:

Davis CA, Ciampaglia GL, Aiello LM, Chung K, Conover MD, Ferrara E, Flammini A, Fox GC, Gao X, Gonçalves B, Grabowicz PA, Hong K, Hui P, McCaulay S, McKelvey K, Meiss MR, Patil S, Peli Kankanamalage C, Pentchev V, Qiu J, Ratkiewicz J, Rudnick A, Serrette B, Shiralkar P, Varol O, Weng L, Wu T, Younge AJ, Menczer F. (2016) OSoMe: the IUNI observatory on social media. PeerJ Computer Science 2:e87 https://doi.org/10.7717/peerj-cs.87

Query Tips


You may search for hashtags (e.g. #cats, #dogs) and URLs (e.g. https://osome.iu.edu).

Must be at least 3 characters in length and may have a trailing wildcard (*) at the end: #cat* will return results with #cat and #cats

URLs must begin with http:// or https://: https://osome.iu.edu*

Hashtags must begin with #, can only contain letters, numbers, and underscores: #cats2020 and #cats_2020 are valid but #cats-2020 is not.

Use a comma-separated list to search for multiple terms (i.e. an OR query): #cats, #dogs will return results that include only #cats, only #dogs, and both #cats and #dogs together.

Network Graph FAQ


The Interactive Network Graph allows you to explore how information spreads across Twitter. The data used to construct the graph comes from the OSoMe Decahose Archive which includes 10% of all tweets. Using this data and the interactive graph, you can discover interesting stories surrounding specified search terms.


What are Network Types?

The networks tool can generate one of two different kinds of networks: A Retweets and Mentions network and a Co-Occurrence network.

In a retweet and mentions network, the nodes represent Twitter accounts and edges represent tweets. The direction of the arrow represents the direction in which the information is flowing (from account A to account B).

In a co-occurrence network, the nodes represent hashtags that occur along side the hashtag that you search for. An edge represents a tweet in which the two connected nodes occur together with the query hashtag.


Why am I unable to search for tweets occuring before 2016?

Our archive of Twitter data contains literally billions of tweets and searching through it requires a significant amount of compute resources. While we would love to provide this pre-2016 data, we are unfortunately unable to provide this capability due to hardware and human resource limitations.


What does K Core Filtered for Performance mean?

When visualizing a network, you may see the message "K Core Filtered for Performance" in the Nodes and Edges count box.

Some queries will produce a network with many hundreds or even thousands of nodes and edges. Unfortunately, if we visualize the entire network, the visualization tool will become exceedingly slow and could even crash your browser. To solve this problem, we must trim the network down to a manageable size that can be visualized while also presenting the overall story of how the queried meme circulates around Twitter. We do this using k-core filtering, in which we focus on the dense core of the network where each node is connected to at least K other nodes. Of course, we only need to do this if the graph is too large to display, so with smaller graphs, we don’t bother to do any filtering.

If the network has been filtered, the "nodes" and "edges" in the legend are the count of nodes and edges presented in the visualization. The "original nodes" and "original edges" are the count of nodes and edges of the original graph before performing any filtering on it.