Title : ( Three approaches to measuring recall on the Web: a systematic review )
Authors: Mahdi Zeynali Tazehkandi , Mohsen Nowkarizi ,Abstract
Object: The object of this paper is to analyze the different definitions and formulas of the recall. Then, the relationship between them and the components of the search engine will be revealed. Method: This is a review article that investigates different views of researchers about recall metric. Result: There are different approaches to calculate recall in search engines. First group believe that recall refers to completeness, but they do not specify where all relevant documents are located. For the second group, recall refers to retrieve all the documents from the collection. It seems that the term “collection” is ambiguous. The third group states that the collection meaning is the search engine’s index. For the Fourth group, the collection meaning refers to the Web. Therefore, recall refers to retrieve the relevant documents existing in the search engine’s index or web according to views of the third, and forth groups, respectively While for the fifth, ranking of retrieved documents should also be taken into account in calculating the recall. Theoretical foundations (descriptions) determine practical techniques (formulas). Hence the third, fourth and fifth descriptions have their analogous formulas, but the first and second definitions are not clear, so they do not lead to the formula. It can be said that these definitions are ambiguous to others, but their authors agree with one of others. Conclusion: Using different definitions and formulas leads to evaluate different components of the information retrieval systems such as search engines.
Keywords
, Information retreival, webometrics, search engine effectiveness, relevance judgement, evaluation metrics@article{paperid:1080034,
author = {Zeynali Tazehkandi, Mahdi and Nowkarizi, Mohsen},
title = {Three approaches to measuring recall on the Web: a systematic review},
journal = {Electronic Library},
year = {2020},
volume = {38},
number = {3},
month = {June},
issn = {0264-0473},
pages = {477--492},
numpages = {15},
keywords = {Information retreival; webometrics; search engine effectiveness; relevance judgement; evaluation metrics},
}
%0 Journal Article
%T Three approaches to measuring recall on the Web: a systematic review
%A Zeynali Tazehkandi, Mahdi
%A Nowkarizi, Mohsen
%J Electronic Library
%@ 0264-0473
%D 2020