TY - JOUR
T1 - Evaluation of News Search Engines Based On Information Retrieval Models
AU - Mohammad Ubaidullah Bokhari
AU - Mohd. Kashif Adhami
AU - Ahmad, Afaq
N1 - Funding Information:
The authors would like to express their great appreciations and gratitude to their respective institutions, namely, Aligarh Muslim University, India, and Sultan Qaboos University, Sultanate of Oman, for providing research facilities, technical supports and research environment that enabled us to complete this research task.
Publisher Copyright:
© 2021, The Author(s), under exclusive licence to Springer Nature Switzerland AG.
PY - 2021/8/10
Y1 - 2021/8/10
N2 - News search engines are the exclusive search services for users’ news intake. Providing relevant query to a news search engine, the user gets back a single news result page consisting of various news articles aggregated from thousands of online news sources available on the World Wide Web. The availability and use of major news search engines like Bing news, Google news and Newslookup demand retrieval efectiveness evaluation of these search systems. In this paper, core retrieval models, namely, vector space model, Okapi BM25 and latent semantic indexing are used to evaluate retrieval efectiveness of news search engines for relevance efectiveness evaluation considering these models separately. Further, Monte-Carlo cross-entropy based rank aggregation technique is used to do more comprehensive relevance efectiveness evaluation by aggregating three individual rankings. Experimental results denote Google news’s performance to be better than the other two search engines.
AB - News search engines are the exclusive search services for users’ news intake. Providing relevant query to a news search engine, the user gets back a single news result page consisting of various news articles aggregated from thousands of online news sources available on the World Wide Web. The availability and use of major news search engines like Bing news, Google news and Newslookup demand retrieval efectiveness evaluation of these search systems. In this paper, core retrieval models, namely, vector space model, Okapi BM25 and latent semantic indexing are used to evaluate retrieval efectiveness of news search engines for relevance efectiveness evaluation considering these models separately. Further, Monte-Carlo cross-entropy based rank aggregation technique is used to do more comprehensive relevance efectiveness evaluation by aggregating three individual rankings. Experimental results denote Google news’s performance to be better than the other two search engines.
KW - Aggregation
KW - Correlation coefficient
KW - Kendall’s tau
KW - Latent semantic indexing
KW - Retrieval effectiveness
KW - Spearman rank order
UR - http://www.scopus.com/inward/record.url?scp=85126336070&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85126336070&partnerID=8YFLogxK
U2 - 10.1007/s43069-021-00081-0
DO - 10.1007/s43069-021-00081-0
M3 - Article
AN - SCOPUS:85126336070
VL - 2
SP - 1
EP - 21
JO - Springer Nature Operations Research Forum
JF - Springer Nature Operations Research Forum
IS - 3
M1 - 41
ER -