Skip to content
This repository has been archived by the owner on Oct 4, 2021. It is now read-only.

Approximating Latent Semantic indexing on streaming data using Adaptive learning methods

License

Notifications You must be signed in to change notification settings

rkmalaiya/dynamic_topic_modeling

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Twitter Multi Variate Analysis

Information about Data

Twitter feeds for 3 different streams, Ethereum, Bitcoin, Facebook for March and April 2018

Research Question

Does, otherwise uncorrelated twitter channels, tend to discuss on similar topics as a result of world level event?

Methods

Correspondence Analysis

  • Contingency table: Count of words (Top 10) / Log of Count of Words (Top 50)
    • Week 9 – 18
    • Week 9, 16, 18
    • Week 13, 14, 17
    • Week 10, 11, 12, 15

Dataset

image1

image2

image3

Heatmap

Top 10

image13

Top 50

image13

Correspondence Analysis

Scree Plot

For Count of words

image4

For log of count of words

image5

Plotting word occurances vs week - Top 10

image32

Plotting word occurances vs weeks - Top 10

image34

Plotting word occurances vs week - Top 50

image42

Plotting word occurances vs weeks - Top 50

image44

Mapping word occurances vs week (Only means of each dataset)

For Count of words (Top 10)

image6

For log of count of words

image7

Dividing weeks into groups

For Count of words (Top 10)

Week 9, 16, 18

image15

Week 10, 11, 12, 15

image14

Week 13, 14, 17

image16

For log of count of words (Top 50)

Week 9, 16, 18

image21

Week 10, 11, 12 ,15

image20

Week 13, 14, 17

image22

About

Approximating Latent Semantic indexing on streaming data using Adaptive learning methods

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published