Applications - PowerPoint PPT Presentation

1 / 7
About This Presentation
Title:

Applications

Description:

Analysis of Stock Market. Medicine. Nuclear Medicine. Scientific Data Analysis. Astronomy ... Graphics, Images and Video. Eg. LightSurf : Exchanges image over ... – PowerPoint PPT presentation

Number of Views:14
Avg rating:3.0/5.0
Slides: 8
Provided by: scha58
Category:

less

Transcript and Presenter's Notes

Title: Applications


1
Applications
  • Network Analysis
  • Resource Management
  • Data Processing
  • Traffic
  • Fraud Detection
  • Security
  • Financial data
  • Analysis of Stock Market
  • Medicine
  • Nuclear Medicine

2
  • Scientific Data Analysis
  • Astronomy
  • Biology ( Biotechnology)
  • Eg. GeneMine Project at Univ.of California, LA
  • Integrates heterogeneous databases across the web
    and correlates subsequences with known patterns.
  • Nuclear Physics
  • Eg. STAR ( Solenoidal Tracker) at relativistic
    Heavy Ion Collider at Brookhaven National
    Laboratory records and analyses complex physics
    events, resulting in data streams of 1000s of MBs
    per sec.

3
  • Media Telecommunications
  • Graphics, Images and Video
  • Eg. LightSurf Exchanges image over call
    phones, archives 10 terabytes on company servers,
    and the data are chunks of images atht require
    special algorithms.
  • Space Exploration and Defense
  • Collection, analysis of large data streams like
    data from radar imaging
  • Sensor Data Processing Network
  • Ecommerce Applications

4
Defence Application (illustration)(Analysis of
information obtained through radars signals)
5
Summary of the discussion
  • What are data streams ?
  • What are the problems faced in mining and
    retrieving data from data streams ?
  • What are the general techniques used in handling
    such data streams ?
  • How are the data streams queried for useful
    applications ?
  • Where are these techniques with data streams used
    in real life ?

6
  • General Flow of processing data streams
  • Block Diagram of the process

Original Data
Stratified (Meta) Data
Mined Data
Parallel Incremental Algorithms
Data Stratification Algorithms
System Support for Mining Dynamic Datasets
7
Conclusion
  • Main Obstacle
  • Truly vast data volumes that occur continuously
  • The data being time varying
  • No way to augment data stream with structure
    information
  • Requirement
  • Urgent need to comprehend higher level
    information inherent in the data stream.
  • Method
  • Using techniques for querying and mining data
    streams
Write a Comment
User Comments (0)
About PowerShow.com