Information Systems

Question 1

What is an Algorithm?

Answer 1

A finite set of rules that gives a sequence of operations for solving a specific type of problem.

Question 2

What is a Computer Program?

Answer 2

An instance or concrete representation, fro an algorithm in some programming language.

Question 3

Name 4 features of algorithms

Answer 3

they are finite, definite, have 0 or more inputs, 1 or more outputs and are effective

Question 4

What is Data

Answer 4

Raw facts eg alphanumeric data, image data etc

Question 5

What is information

Answer 5

Data with some meaning

Question 6

Name 7 aspects of data and information

Answer 6

1. Storing and Processing data
2. Encrypting and security of data
3. Information theory and communication theory.
4. Value of Information
5. Frequency
6. Linguistic theories
7. Human cognition

Question 7

What is a web search engine?

Answer 7

An online web information retrieval system that, given a query, which represents a users information need, returns a list of web pages that match that query.

Question 8

Name the 3 types of data

Answer 8

Structured, unstructured and semi-structured

Question 9

Define Structured Data

Answer 9

Data that resides in a fixed field within a record or file eg often relational (or other) database approach.

Question 10

Define Unstructured Data

Answer 10

Data that isn’t organised in any obviously meaningful way.

Question 11

Define semi-structured Data

Answer 11

Data that doesn’t have a formal structure but does have tags or other information that convey meaning of data, eg XML or RDF documents with headings/sections, emails etc

Question 12

What is the most used data type today

Answer 12

unstructured data

Question 13

What is Organic Content

Answer 13

Unpaid marketing content that potential and existing customers can find naturally.

Question 14

What is sponsored content

Answer 14

Ads with words matching the query words that are ranked above the web documents returned.

Question 15

What is ranking

Answer 15

It involves ordering results returned in response to a user query.

Question 16

Discuss the web link distribution

Answer 16

Web page links are not randomly distributed. Distribution is widely reported to be a power law, in which the total number of web pages with in-degree i is proportional to 1/i^c (c a constant).
ie only a small portion of web pages have a huge number of linksW

Question 17

What does an index do?

Answer 17

Associates a web page with one or more terms

Question 18

Explain pre-processing

Answer 18

1. Case folding (words are changed to lowercase)
2. Punctuation is removed
3. “stop words” are removed
4. “Stemming” is performed

Question 19

What are stop words and why are they removed?

Answer 19

Words that do not provide any extra information about the meaning of the document. They are removed in order to save storage space and speed up searches.

Question 20

What is “stemming”?

Answer 20

Tries to find the “stem” of each word. A stem represents variant forms of a word which share a common meaning. eg consist, consisted and consisting have the same stem “consist”.

Question 21

Describe Lemmatisation.

Answer 21

A lemma is a base form of a word and it is what we look up in a dictionary. i.e. walking -> walk. Lemmatisation is the conversion of a word to its lemma. It is harder than finding its stem

Question 22

What is tf and idf

Answer 22

tf is the term frequency i.e. how often a term occurs in a document
idf is the inverse document frequency which shows is the term occurs often across all document which are being searched

Question 23

How do you calculate the tf-idf and what is it

Answer 23

It is a representation of a real numbe that represents the weights such that the higher the weight the more important the term is in describing the meaning of the document.
The tf is calculated as follows:
no. times term t occurs/ no. terms in a document.
the tf-idf is then calculated:
tfx Logˇ10(N/c + 1)
Where N is the no. documents and c is the no. documents the term occurs in