Social Analysis 2.0

The most distinctive feature of social analysis 1.0 is to understand the public context
by social media,in other words it is to understand public opinion through the social media. However the key of social networking is a connection between indivuduals
thus understanding of social anaysis 2.0 is to understand
the context of the individual social networking.

Social Analysis 1.0

Understanding context of the public through
analyzing keywords in social media

Social Analysis 2.0

Understanding context of the individuals through
analyzing their social actions with mathematical algorithms

Big Data Modeling

Social Big data is time series data of user’s social actions. Modeling formula is needed to analyze unstructured social data. First, we define 5 axis for categorizing social action patterns and second, we define attributes such as ‘larger or small’ and ‘strong or weak’ for each axis.

  • Unstructured Social Data

  • 1st Stage of Data Modeling

  • 2nd Stage of Data Modeling

    1. Categorizing user’s social action patterns
    2. Predicting user’s future behavior

1. Categorizing user’s social action patterns

We analyze user’s social action patterns and categorize them into 32 types of social style

  • Model student

    Model student

  • Edison


  • Announcer


  • Producer


  • Chatting Bot

    Chatting Bot

  • Auntie


  • Writer


  • Brainy Smurf

    Brainy Smurf

  • Happy-Go-Lucky


  • Bar exam taker

    Bar exam taker

  • Neighborhood bully

    Neighborhood bully

  • Baby Bear

    Baby Bear

  • King of reading

    King of reading

  • Friend on the block

    Friend on the block

  • Sergeant in his last year

    Sergeant in his last year

  • Messenger


  • Broadcastings station

    Broadcastings station

  • Hasty Spring Breaker

    Hasty Spring Breaker

  • Coy


  • Teacher


  • Shopaholic


  • Newspaper


  • The Charmer

    The Charmer

  • Traveler


  • Entertainer


  • Explorer


  • Connoisseur


  • Courtship


  • Courtship


  • Crammer


  • Secret crush

    Secret crush

2. Predicting user’s future behavior

We analyze interaction patterns between users and categorize them into over 400 types. Based on that, we predict their future relationship with about 200 types

  • Good Have you noticed how popular you have become recently? More and more people seem to feel empty if they don’t get to see what you have been up to. How great is it to get all this ...
  • Bad You may soon face troubles. Even though you are tired, you are the only person who can solve them. Fix the problems as soon as you can and then come back to your regular...
  • SoSo It’s time to leave for somewhere new. There are lots of things to do and think about, but you don’t have any room in your head. And you are just becoming...

Patented Algorithms

If we give the same value for every social action, it cannot reflect its social meaning. Sometimes, same actions of “Like” have different meaning in the real world. We give different value for each social action according to its communication context based on calculating probability distribution of social interaction occurrence between users It enables us to evaluate individual’s social influence power.

Concept of Valuating Social Action

Patented Core Algorithms

Holding patented algorithms for analyzing social big data

  • Social media contents analysis
  • Social influence power analysis
  • Social influence marketing business model
  • Method and apparatus for setting data value
    KOR. 10-1346278, PCT. KR2013-001764
  • Method and apparatus for setting user influence value in network service
    KOR. 10-1346288, PCT. KR2013-001750
  • Method and apparatus for providing adversement
    information KOR. 10-2012-0112677, patent-pending


Credited Analysis System

We developed scientific and reasonable algorithms and robust analysis system optimized for analyzing social big data. Amazon Web Service credited RANKWAVE for one of excellent cloud examples in big social data platform.

Big data Analysis system

  • Big Data Analysis
    Methodology such as clustering analysis, reputation analysis, and text mining
  • Big Data Storage
    HBase, Membase,
    Cassandra, Redis,
    MongoDB, CouchDB
  • Distributed processing management
    Hive, Pig, Sqoop, ZooKeeper
  • Distributed batch processing
    Distributed file management
    Hive, Pig, Sqoop, ZooKeeper

Analysis System credited by AWS

Achieving Profitability on AWS

AWS Cloud Taekwon for Start-Ups and Developers(slideshare, Korean)