Question

*Decribe Your Problem* relative to Analytics: Big Data to Information*

Characterize your issue – as an inquiry.

What are your objectives? How would you measure them?

Characterize particular achievement criteria

What imperatives do you hope to experience?

Take a theory at what you believe is expected to take care of the issue.

Answer 1

Characterization of Big data analytics issue: Big data is the enormous volume of data in which it can store more than a computer data. Hence it leads to the issues such as searching through several volume of data can make the system slow Another issue is that it is more complicated if any error occurs at the time IF we want to analyze the whole data set by accessing several time it can lead to trouble The cost is much higher to maintain the large volume of data hence there is not convenient tools to access the process Objectives: It is critical for the use of effective data analysis. The data can be accessed from end to end connection between the users and data set. There are various algorithm proposed to overcome the existing challenges such as parallelize existing algorithm. For efficient usage of data the fast implementation comes from the deep research and development strategies for example the matrix multiplication operation works faster in MATLAB code than in C coding this is due to the deep research in the process. This can be consumed by using optimized basic linear algebra subroutine which makes the high level process in the MATLAB Similarly the big data analytics require mathematical algorithm to run in a complicated architecture. Achievement criteria: Many companies now use the distributed classification because of CTR problem. The CTR problem is such that click through rate. But there is notachievement yet because of ongoing research process Open source development is also important for the big data analytics. Imperative to experience: The most necessary thing to experience is that we should use both development algorithm and tools and also include systems into consideration

Regarding issues: The big data can be stored using Hadoop distributed file system, HBase, Hive etc. it will store the graph, document and key value The redundancy can be eliminated by choosing the correct data store based on the data characteristics. Hadoopl give the flexibility to store the data before we are going to process it. It will store upto petabytes of data hence the user can make the better decisions