Literature Review

Write a literature review part for the paper " Tips for Handling Big Data": find 1 source. 200 words.



I added an attachment of specific requirements of literature review and the project planning I made. Can you write literature review based on my project planning? I will make another order(choose you as the writer too) when I need to write the report after finishing the literature review. Thanks!

Literature Review (Include all the requirements in the writing) 

· Purpose: To place your project within the larger whole of what is currently being explored, while demonstrating to your readers that your work is original and innovative

o   Think about the questions other researchers have asked

o   What methods they have used

o   What is your understanding of their findings and their recommendations

o   Do not be afraid to challenge the conclusions of prior research

· Assess what you believe is missing and state how the previous research has failed to adequately examine the issue that your study addresses

· Provide a description of the research work that you intend to include and how you intend to include it. (based on the project planning points)

· Write about one legitimate sources for this assignment, giving roughly 150 words. Some useful ways to approach this requirement:

o   Cite literature

o   Compare

o   Contrast

o   Critique

o   Connect

Project(Report) Planning

My project title: Tips for Handling Big Data

1. Storing raw data.

•  
  • Never manipulate the raw data itself. Make changes to the copy.
  • Store the raw file somewhere that creates automatic backups.
  • Also choose the place of storage such that other workers can access.
  • Keep the data protected.
  • Metadata

1. Data Cleaning 

a.    Structural data 

• Converting data types so they are uniform 
• Structuring unstructured data

1.  
   1. Dealing with missing values 

• When to use mean, median, interpolate, extrapolate + other statistical summaries depending on context and data 

1.  
   1. Outlier analysis 

• Remove unwanted (duplicate or irrelevant) observations.

1. Conducting EDA 

• Visualizing the data 
  • Plot a lot of graphs to catch changes in every stage.
• Summary statistics to understand your data 

1. Dimensionality Reduction (use statistical analysis here) 

• E.g. pros and cons of using PCA when you want to interpret the data 

1. Show your workflow 

• record your entire data workflow: which version you used, the clean up, quality checkups, any processing code you ran,...
• Use version control (descriptions from articles….)

• https://businessoverbroadway.com/2018/03/18/top-10-challenges-to-practicing-data-science-at-work/
• https://towardsdatascience.com/the-ultimate-guide-to-data-cleaning-3969843991d4
• https://towardsdatascience.com/the-art-of-cleaning-your-data-b713dbd49726
• https://elitedatascience.com/data-cleaning
• https://research.aimultiple.com/data-cleaning/
• https://www.dataquest.io/blog/data-science-problems-fix/
• https://elitedatascience.com/data-cleaning
• https://www.upgrad.com/blog/data-cleaning-techniques/