Assignment Code: MLII-102/AST/SEM/ Jul.2022- Jan.2023

Course Code: MLII-102

Assignment Name: Information Processing and Retrieval

Year: 2022-2023

Verification Status: Verified by Professor

1.1 Define indexing language. Discuss its need, purpose and characteristics. (10)

Ans) Indexing is defined by the Online Dictionary for Library and Information Science as a made-up language made up of subject headings or content descriptors chosen to make it easier to find information by acting as entry points in a catalogue or index. It also includes any lead-in words and rules for the form of entry, syntax, etc. In essence, it is a set of terms and tools for figuring out how the terms relate to each other so that an index description can be made. This kind of language is also sometimes called "retrieval language." An indexing language is not a natural language because its meaning and structure may be different. It has a list of words and rules about which expressions are allowed.

Need of indexing language

1. Different authors may write about the same ideas in different ways in different documents.

2. The same words can sometimes mean different things.

3. These ideas might be interesting to users. They need to know that there are specific documents that talk about the ideas.

4. Users might also want to look at documents about related ideas, so similar ideas need to be grouped together to show how they are related.

Purpose of an indexing language

1. Explain the ideas that are talked about in the documents.

2. Show how different ideas are related to each other.

3. Help show how all of the related ideas fit together.

Characteristics of indexing language

Semantic Structure: Semantics is the study of how meaning works. There are two types of relationships between concepts in an indexing language: hierarchical and non-hierarchical. There may be a Genus-Species Relationship or a Whole-Part Relationship. Relationships that are not in a hierarchy can be either Equivalent or Associative.

Syntactic Structure: The syntax means grammar. In indexing, the order of words in a subject heading is determined by the syntax of the language. For example, the title "Export of Iron" could be written as Iron, Export or Export, Iron.

Syndetic Structure: Indexing language should be used to build a syndetic structure that shows the relationships that are described in the semantic structure. Syndetic structure in the indexing language is meant to connect ideas that would otherwise be spread out, and it also helps to group ideas that are related. It tells the indexer and the person looking for information how to write index entries and where to look.

2.1 Explain Cutter’s contribution to Subject Indexing giving examples. (10)

Ans) In his Rules for a Dictionary Catalogue (RDC), which came out in 1876, Charles Ammi Cutter was the first person to give a general set of rules for subject indexing. Cutter never said the word "indexing." Instead, he said "cataloguing."

Cutter found two problems with developing subject headings for organising library materials: a problem with specific subject headings and a problem with the order of terms in compound subject headings. Most of the time, it seemed hard to describe the topic of a document in a single word or phrase. So, the whole subject of the document is looked at, and this is what made it possible for subject cataloguing to lead to subject indexing. In his RDC, Cutter gave rules for both specific and compound subject headings. He did this to make it easier to write subject headings.

Specific Subject Heading

Cutter didn't say what the exact subject heading was, but Rule 161 of RDC made the idea clear. "Enter a work under its subject heading, not under the heading of a class that includes that subject," Cutter said about specific subject headings [Rule 161]. According to this rule, "Roses" should be put under "Roses" and not under "Flower," which is a more general category. But as an example of a specific subject heading, Cutter gave the following:

Subject of the Work Specific Subject Heading

Movement of fluids in plant Botany, Physiological

Here, the work is put under the heading of the class that includes that subject, i.e. under a broader class. So, we can see from Cutter's own example that he couldn't keep things the same. This lack of consistency could be caused by:

1. Cutter thought of a list of "stock subjects" that every book would have to fit into.

2. Cutter couldn't picture the thing that had no name or couldn't be named.

Compound Subject Heading

Rule 175 of RDC says, "Enter a compound subject name by its first word, inverting the phrases only when another word is much more important or is often used alone with the same meaning as the whole name." This rule says that the order of the terms in a compound subject heading should be based on which one is the most important. But Cutter couldn't say how one would show up to decide which was more important. Different people have different ideas about what is important. For example, Sociologists and Psychologists may both look at a document about "social psychology." If this rule is followed, the same document would have two subject headings: Social Psychology and Psychology, Social. So, we can see that the decision about what is "significant" is up to the indexer, which is a subjective choice.

3.1 What is keyword indexing? Enumerate its different variants. Discuss any two of them. (10)

Ans) "Natural Language Indexing" or "Free Text Indexing" may be used to describe an indexing system that doesn't control the words used. Keyword indexing is also called Free Text Indexing or Natural Language Indexing. "Keyword" refers to a "catch word," a "significant word," or a "subject-denoting word" that is usually taken from the titles and sometimes from the abstract or text of a document for indexing purposes. So, keyword indexing is based on the natural language of the documents to make index entries, and this indexing system doesn't need a controlled vocabulary. Indexing by keywords is not new. It was around in the 1800s and was called "catchword indexing" at that time. In the 1950s, computers were used to help people find information. Since 1952, the machine-made keywords index from Title has been used by the Central Intelligence Agency (CIA) of the United States. H. P. Luhn and his colleagues made copies of machine-made permuted title indexes and gave them out at the International Conference of Scientific Information, which was held in Washington in 1958. He called it the Keyword-In-Context (KWIC) index and wrote a paper about how it was made. After adopting KWIC in 1961 for its book "Chemical Titles," the American Chemical Society set its value:

Variants of keyword indexing

KWOC (key-word out-of-context) Index: The KWIC index is similar to the KWOC. Here, each keyword is taken out and printed on its own in the left margin. The full title is printed on the right, in its usual order. Sometimes, the keyword is used as the heading, and the title is put on the next line instead of the same one.

KWAC (key-word Augmented-in-context) Index: Key-word-and-context is another name for KWAC. A document's title doesn't always tells everything need to know about what it says. KWIC and KWOC were not able to solve the problem of retrieving documents that were not important. To stop false drops, KWAC adds keywords from the abstract or the original text of the document to the title or adds them at the end to give more index entries. People also call KWAC "enriched KWIC" or "KWOC." CBAC (Chemical Biological Activities) of BIOSIS uses the KWAC index, which adds another title-like phrase to the title.

4.1 What do you understand by compatibility in ISAR systems? Discuss the areas and principles of compatibility in ISAR systems. (10)

Ans) Compatibility of Information Storage and Retrieval Systems (lSARS) is the quality that makes it easier for two or more organisations to talk to each other and share information. ISAR systems are judged to be compatible with each other based on how they are built and what they do.

Areas of Compatibilities in ISAR Systems

Systems and Components: Today's business, science, research, and academic worlds are all hungry for knowledge, and information systems often have several terabytes of information. Most of the time, these systems can't work without an internet connection. Instead, they have to make the terabytes of information available online with very efficient retrieval engines and easy-to-use interfaces.

Record Format: In the last 30 years, there has been a lot of talk about how bibliographic record formats should be compatible. ISO 2709, the UNISIST Reference manual, UNIMARC, MARC21, and the Common Communication Format are all examples of efforts to address important compatibility issues for international standards and representation in this field.

Media Formats: Changes in technology and techniques have made it so that the way information is handled is always changing. Microfilms, negative films, cine films, videotapes, discs, sound tapes, microcomputers, and optical storage systems were all used to store information.

Retrieval Formats: Most of the time, the formats for storage and retrieval are different. It is best for an IR system to let users choose how they want to retrieve information. Most bibliographic database management systems get information in a standard format, like AACR2.

Principles of compatibility in ISAR systems

ISAR systems' main goals, structure, and functions are based on the following things, and compatibility is sought in these areas:

1. Facts and ideas are made based on what information users of the system are looking for;

2. Give users accurate quantitative and qualitative information that meets their needs;

3. Processes of making, repackaging, or moving information through certain channels;

4. Processes for making certain kinds of documents that searchers want to use as communication channels;

5. Processes for the flow of information carriers to meet immediate communication needs;

6. Information banks, their parameters and structures, and the technological methods of storing, retrieving, and sharing information;

7. Humans are an important part of the systems that create and use information. They play the roles of information creators, information receivers, and information brokers.

8. Elements, structural relationships, and behaviour of an information system;

9. Organization, management, and control of information systems, networks, and the information that flows between them;

10. compatibility of communication processes and operations from a technical point of view;

11. Information languages, both natural and made, in terms of their syntax, meaning, and use.

5.0 Write short notes on any two of the following: (10)

(a) Semantic Web

Ans) People have said that the web won't be able to reach its full potential until both people and computers can share, process, and understand data on it. But the truth is that even though search engines like Google, Yahoo!, etc. give us a lot of hits, they mostly give us information that has nothing to do with what we searched for. The problem with these search engines is that they mostly use statistical methods like the number of times a word is used, how often it is used with other words, etc. This means that when one search for something on the Web, one get a lot of irrelevant results. Even though Google and Yahoo! use entries that have been edited by people, they still come up with a lot of wrong hits. The "semantic web" is a way to add more semantic information to the web. Tim Berners-Lee, who made HTML, HTTP, URI, and the World Wide Web, came up with the idea of the Semantic Web.

(b) Farradane’s Relational Operators

Ans) Relational operators are special symbols that link the isolates to show how they are related. Each operator is shown by a slash and a special symbol that has its own meaning. So, the System of Relational Analysis is what Farradane's system is called. He got the idea for relational analysis by thinking about how people learn, especially how children learn. Analet is a subject statement that is made up of two or more isolates that are connected by relational operators.

Two types of relationships have been found, and they are based on how children learn by developing the ability to tell time and space apart, as well as the stages of this ability in each of the following areas:

1. The time sense is made up of three stages of separation that lead to full association.

2. There are three steps to becoming aware of the degree of difference between two terms in space.