What is Indexing

In bibliography

The indexing of a text is to find in it certain words or expressions which are particularly significant (called terms) in a given context, and create a link between these terms and the original text. For example, the index pages of a book resume (sometimes) the significant terms appearing in the book, are linked to pages of the book where these terms (or their synonyms) appear.

This helps the reader to the location of the pages or sections where you mention a particular topic. Similarly, the contents of a book are a (very rough) form of index.

Many large sites are indexed by an “index”, which you probably helped to find without reading all the previous pages of “A” to “I”. On the other hand, it can index objects that are not necessarily text: a catalog of indexes are museum objects and may include an index of terms related to these objects.

Automatic indexing

Indexing can be manual (by a human) or automatic. The method often most effective when using batch files is indexing (it is also suitable for other types of data stored in memory).

The components are stored in the order of their creation. It then uses an array index, giving first place to the number of the first component, then the second etc. The advantage of this method is that the addition of components is optimal, we add value to the file, and it updates the table index.

Any movement of a component will be replaced by modifying the table index, without physically moving the value in the file. In general, this table can fit in the memory, which allows a rapid change in general. One can also save on magnetic media before leaving the program, thus avoiding the need to recreate (eg re-sort) to the next use.

You can also use an index list if travel is common (but then it becomes a sequential access). The second advantage of this method is that you can use multiple indexes: for example a list of people, we can create an index for the alphabetic names, one on the cities you will access more quickly, to all indexed fields, while non-indexed fields will be satisfied with a sequential search, and without modification in the file (sort by name then by city is necessary without indexing).

The cons are any changes require updating of all tables indexes. The removal by cons is problematic. In general, to avoid gaps in the files, we prefer a mark to distinguish the deleted fields (e.g. a name which is not alphabetic or blank), and then update the index which will point on the field. It can also (as in the method of a super-table) create a list of empty fields, which will be accessed faster than a sequential search at the next insertion.

On an indexed file, you can again afford algorithms using integration, since it affects only the index (fast access). To sort by example, you can use the insertion sort, provided to optimize the search for insertion position (weighted by dichotomy for example), since it requires the reading of fields in the file while the insertion does not involve shifts in a painting, a term generally negligible compared to the time taken for research.

You can also use an index list instead of a table if necessary. The manual indexing information is usually costly: to index properly scientific text of a certain level, it must involve people who are themselves able to understand the text, which imposes a significant cost.

While indexing is based on developed techniques, you can have several different indexing of the same text, equally valid than for others, depending on the use it must be made public and addressed. For example, imagine a text that is a description of selected gardens in England, indexing will be very different depending on whether the public will be made up of gardeners, tourists, or surveyors.

Importance of indexing electronic documents

A major problem in the intangible economy is the exponential growth of the quantity of raw information, but only if there are really relevant processes to enable a correct indexing of this information which would draw the best.
Related terms: index, index, thesaurus, keyword, index, synonym.