Upcoming SlideShare
Loading in …5
×

# Arcomem training twitter-dynamics_advanced

291 views

Published on

This presentation on Twitter Dynamics is part of the ARCOMEM training curriculum. Feel free to roam around or contact us on Twitter via @arcomem to learn more about ARCOMEM training on archiving Social Media.

Published in: Education, Technology
0 Comments
0 Likes
Statistics
Notes
• Full Name
Comment goes here.

Are you sure you want to Yes No
Your message goes here
• Be the first to comment

• Be the first to like this

No Downloads
Views
Total views
291
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
9
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

### Arcomem training twitter-dynamics_advanced

1. 1. Twitter Dynamics ATHENA – Research and Innovation Center in Information, Communication and Knowledge Technologies
2. 2. Objective • Based on data extracted from Twitter, we construct a network of time-stamped term associations • Queries that can be answered – “What are the hashtags associated with #obama at time instance t?” – “Give me the tweets that mention #cnn during the periods that #obama is associated with #romney” – “How the hashtags associated with #obamawins have evolved over time?” 2Twitter Dynamics
3. 3. Temporal Term Associations Model • Model is a set of quintuples • n and c are target and context nodes, respectively, corresponding to terms in tweets • T is the set of time instances for which a tuple is valid • g is the time granularity • w is the association weight – based on co-occurrence of hashtags – Higher weight means more likely to see n and c in the same tweet 3 M = n,c,w,T,g n,c ∈ V,w ∈ [0,1],T ∈ 2Ζ+ ,g ∈ Z +{ } Twitter Dynamics
4. 4. Query operators – M’ = filter(M, cond); • Select the tuples of model M satisfying cond – M’ = fold(M, g); • Return model M’ with base time unit multiplied by g, • from model with weights computed hourly, we get with g=24 a model with weights computed daily – M’ = merge(M); • Merge all tuples from M having the same target n and context c – M’’ = join(M, M’, cond); • Select the tuples of model M satisfying cond with respect to model M’ 4Twitter Dynamics
5. 5. Example Query • What are the tweets that mention #cnn during the periods that #obama is associated with #romney m1=model(""); m2=filter(m1, m1.n="obama" AND m1.c="romney”); m3=filter(m1, m1.n="cnn"); m4=join(m3, m2, m3.T EQ m2.T); 5Twitter Dynamics
6. 6. Implementation • Developed in JAVA • Wide range of storage options for models – Relational databases (MySQL, Postgresql) – Graph database (Neo4J) – RDF triplestore (H2RDF) • Tested with up to 13 millions of Tweets 6Twitter Dynamics
7. 7. Implementation • Developed in JAVA • Wide range of storage options for models – Relational databases (MySQL, Postgresql) – Graph database (Neo4J) – RDF triplestore (H2RDF) • Tested with up to 13 millions of Tweets 6Twitter Dynamics