Indexing of subjects is the action of describing or classifying documents based on index terms or other symbols to indicate what document about, to summarize its content or to improve its capabilities. In other words, it is about identifying and describing the subject document. Indexes are organized, separately, on three different levels: terms in documents such as books; objects in collections such as libraries; and documents (such as books and articles) in the field of knowledge.
Subject indexing is used in information retrieval primarily to create a bibliographic index for retrieving documents on a particular subject. Examples of academic indexing services are Zentralblatt MATH, Chemical Abstracts and PubMed. The term index is largely commissioned by experts but the author's keywords are also common.
The indexing process begins with the document subject analysis. The indexer should then identify the exact term identifying the subject either by extracting the words directly from the document or assigning words from the controlled vocabulary. The terms in the index are then presented in a systematic order.
The indexer must decide how many terms to include and how specific the terms should be. Together this provides the depth of indexing.
Video Subject indexing
Analisis subjek
The first step in indexing is to decide the subject of the document. In manual indexing, the indexer will consider the subject matter in terms of answers to a series of questions such as "Does the document handle certain products, conditions, or phenomena?". Since the analysis is influenced by the knowledge and experience of the indexer, two indexers can analyze the content differently and thereby arise different index terms. This will have an impact on the success of retrieval.
Automatic vs. subject analysis manual
The following automatic indexing regulates the process of analyzing the frequency of word patterns and comparing results to other documents to assign to subject categories. It does not require an understanding of indexed material. This therefore leads to more uniform indexing but this at the expense of the true meaning is being interpreted. The computer program will not understand the meaning of the statement and may therefore fail to establish some relevant terms or set false. Human indexers focus their attention on certain parts of the document such as title, abstract, summary and conclusion, because analyzing full text in depth is costly and time-consuming. The system automatically eliminates the time limit and allows all documents to be analyzed, but also has the option to be directed to specific parts of the document.
Maps Subject indexing
Selection of terms
The second stage of indexing involves translating subject analysis into a set of index terms. This can involve extracting from documents or assigning from controlled vocabulary. With the ability to perform full text search widely available, many people have relied on their own expertise in searching information and full text search has become very popular. Indexing of subjects and experts, professional indexers, catalogers, and librarians, remains important for information and retrieval organizations. These experts understand the controlled vocabulary and can find information that can not be found by full text search. The cost of expert analysis to create subject indexing is not easy compared to the cost of hardware, software, and labor to produce a fully searchable full text set. With a new web application that allows every user to annotate documents, social tagging has gained popularity especially on the Web.
One indexing app, book index, remains relatively unchanged despite the information revolution.
Indexing/Indexing
Indexing extraction involves taking words directly from a document. It uses natural language and matches the automatic technique in which the word frequency is calculated and the frequency with a predetermined threshold is used as an index term. A stop-list containing common words (such as "", "and") will be referenced and such quit words will be excluded as index terms.
Automatic extraction indexing can lead to the loss of the term meaning by indexing single words as opposed to a phrase. While it is possible to extract frequent phrases, it becomes more difficult if key concepts are inconsistently written in phrases. Automatic extraction indexing also has the problem that, even with the use of stop-lists to remove common words, some frequent words may not be useful to allow discrimination between documents. For example, the term glucose tends to occur frequently in every document associated with diabetes. Therefore the use of this term is likely to return most or all of the documents in the database. Post-coordinate indexing where terms are combined at the time of search will reduce this effect but the responsibility will be in the searcher to connect the exact term that goes against the information professional. In addition rare terms may be very significant eg new drugs may be mentioned rarely but the novelty of the subject makes a significant reference. One method of allowing a more rare term to include and common words to be excluded by automatic techniques is the relative frequency approach in which the frequency of a word in a document is compared with the frequency in the database as a whole. Therefore, the more frequent terms in the document than might be expected based on the rest of the database can then be used as an index term, and terms that often occur will be excluded. Another problem with automatic extraction is that it does not recognize when a concept is discussed but is not identified in the text by indexable keywords.
Task indexing
An alternative is indexing tasks where the index term is taken from a controlled vocabulary. It has the advantage of controlling synonyms as a preferred term indexed and synonyms or related terms directing users to the desired terms. This means users can find articles regardless of the specific terms used by the author and keep users from having to know and check all possible synonyms. It also removes the confusion caused by the homograph by entering the qualifying term. A third advantage is to allow linking of related terms whether they are connected to a hierarchy or association, e.g. index entries for oral medications may list other oral agents as related terms at the same level of the hierarchy but will also link to broader terms such as treatment. Task indexing is used in manual indexing to improve inter-indexer consistency because different indexers will have a set of terms that are controlled to choose from. Controlled vocabulary does not completely remove the inconsistency because two indexers may still interpret the subject differently.
Index presentation
The last phase of indexing is to present entries in a systematic order. This may involve linking entries. In a coordinated index before, the indexer determines the order in which terms are connected in an entry by considering how users can define their search. In the post-coordinated index, entries are presented singly and users can link entries via search, most often done by computer software. Post coordination results in loss of precision compared to pre-coordination
Indexing Depth
The indexer should make decisions about which entries should be entered and how many entries the index should include. Indexing depth illustrates the accuracy of the indexing process with reference to electrical and specificity
Fatigue
The full index is one that lists all possible index terms. Greater flexibility provides higher memory, or greater possibility of all relevant articles being retrieved, however, this happens at the expense of precision. This means that users can retrieve more irrelevant documents or documents that only deal with the subject in depth. In a manual system, a larger electrical level carries a larger cost because it takes more hours. The additional time taken in the automated system will be much less significant. At the other end of the scale, in selective indices only the most important aspects are discussed. Remember reduced in selective indexes as if the indexer does not include enough terms, highly relevant articles may be ignored. Therefore, the indexer should try to balance and consider what documents can be used. They may also have to consider the implications of time and cost.
Specificity
Specificity illustrates how closely the index term corresponds to the topic they represent Index is said to be special if the indexer uses the parallel descriptor with the concept of the document and reflects the concept appropriately. Specificity tends to increase with fatigue because the more terms you enter, the narrower the terms are.
Indexing theory
HjÃÆ'¸rland (2011) found that indexing theories are at the deepest level connected with different theories of knowledge:
Indexing rationalization theory (such as Ranganathan's theory) shows that subjects are logically arranged from a set of fundamental categories. The basic method of subject analysis is then "synthetic-analytic", to isolate a set of basic categories (= analysis) and then construct the subject of each given document by combining the categories according to some rules (= synthesis). The empirical theory of indexing is based on the selection of similar documents based on their properties, in particular by applying numerical statistical techniques. Historical theory and indexing hermeneutics show that the subject of a given document relative to a particular discourse or domain, why indexing should reflect the needs of a particular discourse or domain. According to hermeneutics is a document that is always written and interpreted from a certain horizon. The same thing happens with the knowledge organization system and with all users looking for such a system. Any questions put forward to such a system are laid out from certain horizons. All of these horizons may be more or less in consensus or in conflict. To index the document is to try to contribute to retrieving "relevant" documents by knowing about different horizons. Pragmatic and critical indexing theory (such as HjÃÆ'¸rland, 1997) conforms to the historical point of view that the subject is relative to a particular discourse but emphasizes that the subject's analysis must support established goals and values ​​and should consider the consequences of indexing one or other ways. These theories believe that indexing can not be neutral and it is the wrong goal to try to index in a neutral way. Indexing is an action (and computer-based indexing acts in accordance with the programmer's intent). Acts act for human purposes. Libraries and information services also serve human purposes, why their indexing should be done in a way that supports this goal as much as possible. At first glance this looks odd because the purpose of libraries and information services is to identify a document or piece of information. Nonetheless, certain ways of indexing always support some types of usage at the expense of others. The documents to be indexed intend to serve some specific purpose within a community. Basically indexing should intend to serve the same purpose. The primary and secondary documents and information services are part of the same social system as a whole. In such systems, theories, epistemology, worldviews, etc. Can be played and the user must be able to adjust and navigate between the different views. This calls for different epistemological mapping in the field and the classification of a single document into the map. Excellent examples of different paradigms and their consequences for indexing and classification systems are provided in the art domain by ÃÆ'ËÅ"rom (2003) and in music by Abrahamsen (2003).
The essence of indexing is, as stated by Rowley & amp; Farrow to evaluate the contribution of papers to the knowledge and index it accordingly. Or, with the words HjÃÆ'¸rland (1992, 1997) to index its informative potential.
"To achieve consistent good indexing, the indexer should have a thorough appreciation of the subject structure and the nature of the contributions that the document makes for the advancement of knowledge." (Rowley & Farrow, 2000, p.Ã, 99).
See also
- Indexing and abstract services
- Document classification
- Metadata
- Overcategorization
- Thomas of Ireland, the medieval pioneer in indexing the subject
References
- Fugman, Robert (1993). Analysis and indexing of subjects. Theoretical foundation and practical advice . Frankfurt/Main: Index Verlag.
- Frohmann, B. (1990). "Rules of Indexing: Criticism of Mentalism in the Theory of Information Taking". Documentation Journal . 46 (2): 81-101. doi: 10.1108/eb026855.
Source of the article : Wikipedia
0 Comments