A Literature Review of Indexing and Searching Techniques Implementation in Educational Search Engines

Search Engine

Several definitions of information searching have been defined, describe searching as a process of storage, organization, representation, and access to information by the user.

As shown in Figure 1, R. K. Belew (Belew, 2002) proposed a process that today considered as a general Information Retrieval System (IRS) process.

Figure 1.

The IRS Process

The general process is divided into several parts of which three processes are of paramount importance:

• •

  Indexing Process: Indexing consists of summarizing a data by characteristic elements resulting from this thing data to be able to classify it and to bring it closer to similar things data, indexing makes it possible to group values in order to find them more quickly. There are three modes of indexing. Manual where the human is charged of indexing. Automatic where indexing is done in a fully automated process. Semi-automatic where indexing is done through a fully automated process, but the intervention of the human being is necessary to refine the index.

• •

  Searching Process: Comparison between the index representation model of the query and the document. This process is used to measure the degree of similarity between the query and document. Finally, it generates a set of documents that respond to the request.

• •

  Evaluation of the Answer: The user uses the documents reported and which meet his needs, otherwise he reformulates another request.

The indexation must support some issues that will be detailed in the subsequent pages, is concerned with recovering the important terms of the document what is called classical indexing or is interested in recovering the meaning of the important words of the document, relations and events that connect them; this type of indexing is called semantic.