Instruction
- Total number of questions : 4
- Time allotted : 06:00 Minutes:Seconds.
- Each question carry 1 mark, no negative marks.
- DO NOT refresh the page.
- All the best
Data Mining [ Question No: 1 To 4 ] :
Which of the following is the characteristic which is not exhibited by social network ?
The information gain, one of the attribute selection measures, used in partition of data (D), is expressed as
Info (D) = .................
where \(p_i\) is the probability that an arbitrary tuple in D belongs to class \(c_i\) :
A statistical method used to make numerical predictions is ................. .
For constructing Data warehouse from operational databases, which of the following operations are necessary ?
Total number of questions : 4
Number of answered questions : 0
Number of unanswered questions : 4
Data Mining [ Question No: 1 To 4 ] :
Which of the following is the characteristic which is not exhibited by social network ?
Your Answer : (Not Answered)
Correct Answer : C
The characteristic not exhibited is Widening diameter.
Studies of real social and information networks, notably by Leskovec, Kleinberg and Faloutsos, found three regularities. Degree distributions are heavy-tailed, with a few nodes having very many connections. Networks densify over time, the number of edges growing faster than the number of nodes according to a power law. And the effective diameter shrinks as the network grows.
So social networks show shrinking, not widening, diameters, contrary to what earlier models predicted. Since one listed characteristic is not exhibited, 'None of the above' is wrong.
The information gain, one of the attribute selection measures, used in partition of data (D), is expressed as
Info (D) = .................
where \(p_i\) is the probability that an arbitrary tuple in D belongs to class \(c_i\) :
Your Answer : (Not Answered)
Correct Answer : C
The expected information is \(\text{Info}(D) = -\sum_{i=1}^{m} p_i \log_2(p_i)\).
Info(D), the entropy of D, is the average number of bits needed to identify the class of a tuple. Each class \(c_i\) with probability \(p_i\) contributes \(-p_i \log_2 p_i\); the logarithm is to base 2 because information is measured in bits, and the minus sign makes the result non-negative, since \(\log_2 p_i \le 0\) for \(0 \lt p_i \le 1\).
The options without the minus sign would give a negative value, and those with an unspecified base do not give the measure in bits used in ID3's information gain.
A statistical method used to make numerical predictions is ................. .
Your Answer : (Not Answered)
Correct Answer : A
The method is Regression.
Regression analysis models the relationship between a dependent variable and one or more independent variables, for example by fitting a line in linear regression, and uses it to predict a continuous numerical value, such as a house price from its size and location. It is the standard data mining method for numeric prediction.
A support vector machine is mainly used for classification into categories. A hidden Markov model is a probabilistic model of sequences of hidden states, used in speech and sequence labelling. Link analysis studies the connections between nodes in a network, as in PageRank.
For constructing Data warehouse from operational databases, which of the following operations are necessary ?
Your Answer : (Not Answered)
Correct Answer : A
The necessary operations are Extract and Load.
A data warehouse is built from operational databases by the ETL process: data is extracted from the source systems, cleaned and transformed into a consistent form, and loaded into the warehouse, usually in periodic batches.
Create and Edit and Query and Update are operations on operational (OLTP) data, not steps in building a warehouse. Slice and dice are OLAP operations used to analyse the warehouse once it exists.
Marks
Total number of questions : 4
Number of answered questions : 0
Number of unanswered questions : 4