Objectives of Information Retrieval

Information retrieval is the activity of obtaining information resources relevant to an information need from a collection of information resources. Searches can
be based on metadata or on full-text indexing.

Automated information retrieval systems are used to reduce what has been called "information overload". Many universities and public libraries use IR systems
to provide access to books, journals and other documents. Web search engines are the most visible IR applications.

To understand the actual concept of Information Retrieval, it’s essential to step back and review the hierarchy of information. Figure 2. 1 below depicts the

hierarchy of Information with its refinement.

As seen the base of the pyramid encompasses data at a broader spectrum and as we move to the top the data goes through refinement and filtering to yield more

usable and accessible information. As seen Data is raw material of information. It is broader and more abstract. It cannot be used as it is and will surely need a

great deal of restructuring. As we move in the upward direction, Information gets refined as data is organised and presented in a particular manner. This can be

more usable than raw data. The data in knowledge form is Information that can be acted upon. Further up the pyramid is Wisdom which is distilled and integrated

knowledge. It demonstrates high-level “understanding”.

A (Facetious) Example of all four rungs of the pyramid are as follows

● Data - 98.6º F, 99.5º F, 100.3º F, 101º F

● Information -Hourly body temperature: 98.6º F, 99.5º F, 100.3º F, 101º F

● Knowledge-If you have a temperature above 100º F, you most likely have a fever

● Wisdom-If you don’t feel well, go see a doctor

Given the basic perception of information, take one to the concept of Information Retrieval. The need of Information retrieval is the fact that information has

been growing multifold, the Internet being the main reason for the explosion of information. There is a plethora of information available in all shapes ,sizes and

forms. This is why a proper retrieval system is imperative for management and accurate retrieval of desired data in desired form. Thus retrieval is an activity of

prime importance. Figure 2.2 below shows a diagrammatic representation of data retrieval.
As seen above,

● “Fetch something” that’s been stored

● Recover a stored state of knowledge

● Search through stored messages to find some messages relevant to the task at hand

Thus an Information retrieval can be structurally defined as “Information retrieval is a problem-oriented discipline, concerned with the problem of the

effective and efficient transfer of desired information between human generator and human user”

Given the definition of Information Retrieval, perception of the functionality of retrieval system is as follows

1. Sourcing and information Gathering methods.

2. Information retrieval procedures.

3. Updation of information .

4. Provides “QUALITY” information services .

With availability of information , the next big question is how useful is all the information available. For the seeker of information, availability of information is

one parameter. However, quality of information is far more important . Thus the last function of a retrieval system i.e. Quality of information services has

become not just primary functionality but also the most important functionality.

Quality of information is therefore judged on four parameters

1. Accuracy of retrieved information

2. Up-to-dateness of Information

3. Timely retrieval of information

4. Relevance of retrieved information with respect to required information

Thus , to make the above mentioned parameters of retrieval system achievable , it is essential to define objectives for a retrieval System which are listed as

follows

� Efficiency {Crucial and main objective}

� Minimise search Overheads:- Overhead deals with “Time spent in all steps leading to the reading of items containing the needed

information”
● Sufficient information in the system to complete a task.

● All information in the system is relevant to the user's needs.

● Example –shopping:

o Looking for an item to purchase.

o Looking for an item to purchase at minimal cost.

● Example –researching:

o Looking for a bibliographic citation that explains a particular term.

o Building a comprehensive bibliography on a particular subject.

� Measurement of success:-

● Two dual measures:

o Precision: Proportion of items retrieved that are relevant.

▪ Precision = relevant retrieved / total retrieved

▪ = |Answer Ç Relevant | / |Answer |

o Recall: Proportion of relevant items that are retrieved.

▪ Recall = relevant retrieved / relevant exist

▪ = |Answer Ç Relevant | / | Relevant |

Figure 2. 3 below shows the set relation between relevant and retrieved documents

� Facilitate good search Tools

● Ambiguities inherent in languages.

● Homographs: Words with identical spelling but with multiple meanings.

● Example: Chinon—Japanese electronics, French chateau.

● Limits to the user's ability to express needs.

● Lack of system experience or aptitude.

● Lack of expertise in the area being searched.

● Initially only vague concepts of information were sought.

● Differences between user's vocabulary and authors' vocabulary: different words with similar meanings.

� Comprehendible result presentation

Present search results in format that helps user determine relevant items:

● Arbitrary (physical) order

● Relevance order

● Clustered (e.g., conceptual similarity)

● Graphical (visual) representation

The above listed objectives are achieved by information retrieval system. Various tasks are performed to achieve these tasks.

Therefore the information retrieval system has to deal with the following tasks:

•Generating structured representations of information items: this process is called feature extraction and can include simple tasks, such as extracting words from a

text as well as complex methods, e.g. for image or video analysis.

•Generating structured representations of information needs: often this task is solved by providing users with a query language and leaving the formulation of

structured queries to them. This is the case for example for simple keyword based query languages, as used in Web search engines. Some information retrieval

systems also support the user in the query formulation, e.g. through visual interfaces.

•Matching of information needs with information items: this is the algorithmic task of computing similarity of information items and information need and

constitutes the heart of the information retrieval model. Similarity of the structured representations is used to model relevance of information for users. As a result

a selection of relevant information items or a ranked result can be presented to the user.

Since information retrieval systems deal usually with large information collections and/or large user communities, the efficiency of an information retrieval

system is crucial. This imposes fundamental constraints on the retrieval model. Retrieval models that would capture relevance very well, but are computationally

prohibitively expensive are not suitable for an information retrieval system.