Big Data Analytics - Introduction
3 likes2,783 views
The document discusses big data analytics, defining big data as large, semi-structured or unstructured data sets beyond the capacity of traditional relational databases. It highlights the exponential growth of internet traffic and the rise of the Internet of Things, emphasizing the implications for marketing, medical applications, and government tracking. Various tools and solutions for managing and analyzing big data, including Hadoop, Spark, and NoSQL databases, are also mentioned, along with resources for getting started with data analytics.
![JSON
{
"employees": [
{ "firstName":"John" , "lastName":"Doe" },
{ "firstName":"Anna" , "lastName":"Smith" },
{ "firstName":"Peter" , "lastName":"Jones" }
]
}
Source: http://www.w3schools.com/json/json_syntax.asp](https://image.slidesharecdn.com/bigdataanalytics-140328102200-phpapp02/85/Big-Data-Analytics-Introduction-21-320.jpg)
Big Data Analytics - Introduction
- 2. What Is Big Data Analytics? ● Big Data – Buzz word – Two definitions: ● Data sets too large for modern relational databases ● Semi-structured/Unstructured data sets ● Analytics – The science of measuring and discovering patterns and trends with data
- 5. Data, Data, Everywhere... ● In 2004: – Internet traffic: 1 Exabyte (that's 134,217,728 8GB flash drives) – A lot of other media: ● Newspapers/books/magazines ● DVDs
- 6. Data, Data, Everywhere... ● Today: – Internet traffic: 1.3 Zettabytes (that's 178,670,639,360 8 GB sticks) ● 110.3 exabytes per month – Even more media: ● Mobile devices (phones/tablets/mp3 players/etc) ● The Internet of Things ● Streaming Media
- 7. The Internet of Things ● How many of you have... – Fitness trackers? – E-readers? – Ipods? ● Tie them to social sites (i.e. Facebook)?
- 8. The Internet of Things ● You're being tracked! ● So what? – Marketing – Medical – Government ● Building fuller picture of what's tracked.
- 10. Six Degrees of Separation Source: http://www.83toinfinity.com
- 12. Data Storage
- 13. Data Storage ● Relational Databases – Structured data – Can scale to huge volumes of data ● Hadoop – Semi-structured/unstructured data – Massively parallel storage and processing
- 17. What Solution to Pick? ● Data Volume and Speed – Relational Databases Will Cap out – ”Big Data” Stores Scale (For Now) ● Hadoop ● Spark ● Lucene – Alternative Modeling Techniques ● Hyper Normalized (6-8NF) – Inmon's Textual Disambiguation – Anchor Modeling – Data Vault
- 19. Hadoop ● Version 1 – Giant data store – File distribution – File parsing tools – Generic security ● Version 2 – Giant data store – Replaced foundation work – Unified security -LDAP/Kerberos support
- 20. Tools ● Oozie ● Hive ● NoSQL Databases – Hbase – MongoDB
- 21. JSON { "employees": [ { "firstName":"John" , "lastName":"Doe" }, { "firstName":"Anna" , "lastName":"Smith" }, { "firstName":"Peter" , "lastName":"Jones" } ] } Source: http://www.w3schools.com/json/json_syntax.asp
- 22. How to Analyze? ● Performance ● Timeliness ● Accuracy ● Feedback
- 23. “Big Data” Solutions ● Search the entire data set ● Great performance ● Highly accurate ● Integrates into Analytics tools – Only some of the tools are able to support Hadoop, etc.
- 24. Statistics ● Designed for all sizes of data sets ● Decreases time to results ● As accurate as needed ● Analytics tools fully support ● Most “Big Data” tools support
- 25. Analytics Tools ● Can access data of most sizes – Most can handle Hadoop and some NoSQL databases ● Built for Predictive Modeling ● Starting to handle social/network modeling
- 26. How to Get Started ● Grab some tools! – RapidMiner (http://rapidminer.com/) – R (http://www.r-project.org/) – Weka (http://www.cs.waikato.ac.nz/ml/weka/) ● Grab some data! – http://www.kdnuggets.com/datasets/index.html – http://aws.amazon.com/publicdatasets/ – http://www.reddit.com/r/datasets
- 27. Prizes/Challenges ● Kaggle - https://www.kaggle.com/ ● MIT - http://bigdata.csail.mit.edu/challenge ● Heritage Health Prize - http://www.heritagehealthprize.com/c/hhp
- 28. ● Twitter - @OpenDataAlex ● LinkedIn – alexmeadows ● Github - dbaAlex Questions? Comments?