1. FOA section 1.1
Founding Out About of a topic of special interest need looking for things are relevant and enables us know the meaning of things which is the semantics of words, sentences, questions and documents involved.
The basic process of FOA (Ask, Answer and Assess) would be specific to each component of search engine. And the fundamental operation of search engine is a match between description (queries) by users and documents (corpus).
2. IES section 1.1 and 1.2
IR: represent, search, and manipulate large collections of electronic text and other human-language data.
Relevance ranking is a core problem in IR <- RSV (Retrieval Status Value), Probability Ranking Principle (PRP)
A major task of a search engine is to maintain and manipulate an inverted index for a document collection.
Search result presenting: document (self-contained unit), elements (pages/paragraphs) or snippets (text passages/video segments)
Performance Evaluation: efficiency, effectiveness, specificity, exhaustivity and novelty.
3. MIR sections 1.1-1.4
The IR Problem: the primary goal of an IR system is to retrieve all the documents that are relevant to a user query while retrieving as few non-relevant documents as possible.
Web effects to search are from: the characteristics of the document collection itself; the size of the collection and the volume of user queries submitted on a daily basis; the vast size of the document collection; the fact that the Web is not just a repository of documents and data, but also a medium to do business; Web advertising and other economic incentives.
Practical Issues: security, privacy, copyright and patent rights.
No comments:
Post a Comment