Blog 7

1.    What are some of the reasons that might warrant the need to use a search system on a website?

The implementation of the search system is recommended for the website that has huge amount of content in it.  This not only improves the information retrieval easier, but also improves the user experience as well. Search systems are very useful in the following way: -

1)   Easier to retrieve information.

2)   Searching systems helps to display the content, which exhibits similar information in them.

3)   Search results are filterable.

4)   Searching system reduces time require to extract information from a website.

5)   Helps user in completing their tasks.

2.    Why is an Information Architect interested in search systems?

Information Architect is directly influenced by the need of searching system, because the implementation of search system requires a lot of invisible work that can includes: -

1)   How a search engine might benefit a user by leveraging metadata

2)   How a search engine’s interface could be improved

3)   How search engine should be integrated with browsing/navigation

3.    Describe the core components of a search engine.

The core components of a search engine are given below: -

1)   The WebCrawler  - the part of the search engine which combs through the pages of the website, and gather the information for the search engine.

2)   The database – The search engine’s database is what you are actually searching.

3)   The search algorithm – the algorithm used to retrieve the results from the database.

4)   The ranking algorithm – to determine the frequency and location of the content, along with the link analysis.

4.    What is a search zone? What are the approaches for creating search zones?

A search zone is a subset of a website that has been separately indexed from the rest of its content.  Create search zones by segregating documents or logically tagging them. For example – distinguishing documents on the basis of their content type, audience, role, subject/topic, geography, chronology, author, department/business unit

5.    Explain the difference between recall and precision in terms of search results.

Recall – It is a number based on the # of relevant record retrieved from a database compared to the total # of relevant records.

Precision – It is the number on the # of relevant records retrieved from a database compared to the total # of records.

Recall provides more accurate and returns % of ACTUAL relevant documents available, whereas precisions looks at the total number of records available.

6.    Consider the following search engines:

a. Search engine A retrieves 600 documents out of a total of 8,200 documents. Out of the 600 documents retrieved, only 500 are relevant out of a total of 923 relevant documents. Calculate the recall and precision rates for the query.

A = 500, B = 423, C = 100

Recall = 500/(500+423)* 100 = 54.17%

Precision = 500/(500+100)*100 = 83.33%

b. Search engine B retrieves 131 documents out of a total of 8,200 documents. Out of the 131 documents retrieved, all 131 are relevant out of a total of 923 relevant documents. Calculate the recall and precision rates for the query.

A = 131, B = 792, C = 0

             Recall = 131 / (131 + 792) * 100 = 14.19%

             Precision = 131 / (131 + 0) * 100= 100%

c. Search engine C retrieves 700 documents out of a total of 8,200 documents. Out of the 700 documents retrieved, 0 are relevant out of a total of 923 relevant documents. Calculate the recall and precision rates for the query.

A = 0, B = 923, C = 700

             Recall = 0 / (0 + 923) * 100= 0%

             Precision = 0 / (0 + 700) * 100= 0%

d. Search engine D retrieves 5,000 documents out of a total of 8,200 documents. Out of the 5,000 documents retrieved, 923 are relevant out of a total of 923 relevant documents. Calculate the recall and precision rates for the query.

A = 923, B = 0, C = 4077

             Recall = 923 / (923 + 0) * 100= 100%

             Precision = 923 / (923 + 4077 * 100= 18.46%

7.    What is the purpose of a stemming tool? Explain the difference between strong and weak stemming. Provide examples of strong and weak stemming.

Stemming returns different words that come from a common stem.  An easily understandable example of this would be –

Computer, computers, computing

Strong Stemming Vs. Weak Stemming – The strong stemming expands a user’s query and returns all related documents whereas weak stemming expands the query to only return plurals.

Examples: -

Strong Stemming – bat, bats, bating

Weak Stemming – bat, bats

  

8.    What are two main issues to consider when displaying the results of a search?

1)   What are the content to be displayed

2)   How to list or group those results

9.    How many documents should you display in a search result?

The number of documents to be displayed depends upon the various factors –

1)   User’s internet connectivity speed

2)   User’s display settings  (May require pagination)

3)   User’s monitor resolution

4)   User’s browser settings

10. Describe some approaches for sorting and ranking search results for display

Sort by alphabetical – result sorting based on their alphabetical order (e.g. A comes before B)

Sort by chronological – results sorting based on their date of publication.

Rank by relevance – ranked by the relevance of the document using algorithms

Rank by popularity – ranked by the number of links there are to a document

Rank by user or expert ratings – ranked based on ratings made by users/experts

Rank by pay-for-placement – websites pay to be ranked higher than others

11. When sorting search results alphabetically, why is it a good idea to omit articles such as “a” and “the”?

It is always a good practice to remove the article such as “a” and “the” because terms such as A and THE can affect an alphabetical listing as they would force a sort “A New World” by A, rather than the keyword or title letter “N”, which is what likely would be searched.

12. How does “best bets” ranking operate?

Best bet ranking is derived from human indexing that is done through expertise or search-log analysis. Best bets give a human, educated, researched and expert opinion of the best hits for a keywords search to give the user a more educated and relevant hit at the top of the list.

13. What are four key factors to consider when designing a search system interface?

1)   Level of searching expertise and motivation

2)   Type of information needed

3)   Type of information being searched for

4)   Amount of information being searched

14. What are some of the ways search system designers can help a user when no results are returned for a query?

1)   Give user tips to improve their query

2)   If user is typing too many words, and getting no results, than ask user to try fewer keywords

3)   Allow user to try advanced search options.

4)   Provide human support if any of the above doesn’t work.