Master in Data Science (SMS, TU) Statistical Computing with R Question Paper 2078 (Set First Assessment 2078, p13-14) Nepal

Explain how can you import following types of data into the R software with simple examples/codes:

a) a text file saved in the local computer b) a table embedded in any webpage c) json file with web API

Explain the logic behind extraction of the following subsets from a 5x5 data frame in R software:

a) First two rows b) Third and fifth row with second and fourth column c) Add 5 new rows in this data frame

Explain data mining in data science with focus and examples on:

a) Tasks b) Analytics c) Learning's

Explain how to work efficiently with "big data" in R software in relation to the:

a) Subsetting with base R and dplyr packages b) ff, ffbase and ffbase2 packages c) data.table package

Explain social network analysis and describe its use in a real-life situation with:

a) Nodes b) Links c) Attributes

Open the R or R studio software and do the followings with R script:

a) Define integers from 1 to 15 using three different coding approaches in R b) Define these five numbers: 1.1, 2.2, 3.3, 4.4 and 5.5 and save it as column vector N c) Add, subtract, multiply and divide vector R from vector N and interpret the results carefully d) Define a list using "This" "is" "my" "first" "programming" "in" "R" and save it as L e) Transform these list elements as characters of UL object.

Import the "pollution.csv" file into R studio and do as follows with R script:

a) Check the structure of the data and explain class of each variable b) Change the attributes of "particulate matter", "date time" and "value" variables c) Get the summary of all the variables and replace the outliers as missing value d) Get summary statistics of "value" variables by "particulate matter" variable categories e) Write a summary of the results obtained in the earlier steps with interpretation and conclusion

Use the "pollution.csv" file imported and cleaned in R studio and do as follows with R script:

a) Create bar plot of "particulate matter" variable b) Create histogram of "value" variable c) Create line plot of "date time" and "value" variables d) Create histogram of "value" variable by particulate matter categories e) Write a summary of the results obtained in the earlier steps with interpretation and conclusion

Load the "term Doc Matrix.R data" file into R studio and do as follows with R script:

a) Define the term document matrix data object as matrix and store it as "m" object b) Define the frequencies of the terms using "row Sums" function and get the term frequencies c) Create a histogram of the term frequencies using ggplot2 package d) Create a histogram of the terms with 10 or more frequencies using ggplot2 package e) Create word cloud of term frequencies using word cloud package and interpret it carefully

OR

Load the "rdm Tweets.rdata" file in R studio and do as follows with "tm" and "tweetR" packages:

a) Convert twitter list as data frame and assign it as "df" object b) Create corpus using the "text" column of the data frame c) Perform pre-processing to clean the corpus for text mining d) Create term document matrix using the cleaned corpus e) Find the most frequent terms using the term document matrix f) Find the co-occurrence of the term "r" with filter of 0.1 and above.

Load the "igraph" package in R studio and do the basic SNA as follows with R scripts to:

a) Define g as graph object with (1,2) as its elements b) Plot the g and interpret it carefully c) Define g1 as graph object with ("S", "R", "R", "G", "G", "S", "S", "G", "A", "R") as its elements d) Plot g1 with node color as green, node size as 30, link color as red and link size as 5 e) Get degree, closeness and betweenness of g1 and interpret them carefully.

OR

Load the "term Doc Matrix.R data" file into R Studio and do as follows with R script:

a) Define term Doc Matrix as matrix m b) Transform it into adjacency matrix c) Build an undirected SNA graph with the adjacency matrix data d) Remove loops and plot the SNA graph again e) Interpret all the results carefully