TY - GEN
T1 - A method for finding similes to support final selection from search results
AU - Ushiama, Taketoshi
AU - Matsuo, Noboru
N1 - Copyright:
Copyright 2015 Elsevier B.V., All rights reserved.
PY - 2015/1/8
Y1 - 2015/1/8
N2 - This paper introduces a method for supporting users to select the final object(s) from the results of search/recommendation. In order to realize the purpose, our method provides the user with similes that would make the user understand the characteristic features of a pair of objects easily and quickly. In order to search suitable similes, we introduced the difference feature vector for a pair of objects. The difference feature vector emphasizes the differences of the objects in the pair and downplays the common features, because objects in a search/recommendation result set have the common features that matched the search/recommendation conditions and the common features are useless for further selection. The effectiveness of the proposed method is evaluated based on the experimental user studies and it is shown that our method can search more adequate simile pairs for the given pair of objects than the baseline method based on the simple feature vectors.
AB - This paper introduces a method for supporting users to select the final object(s) from the results of search/recommendation. In order to realize the purpose, our method provides the user with similes that would make the user understand the characteristic features of a pair of objects easily and quickly. In order to search suitable similes, we introduced the difference feature vector for a pair of objects. The difference feature vector emphasizes the differences of the objects in the pair and downplays the common features, because objects in a search/recommendation result set have the common features that matched the search/recommendation conditions and the common features are useless for further selection. The effectiveness of the proposed method is evaluated based on the experimental user studies and it is shown that our method can search more adequate simile pairs for the given pair of objects than the baseline method based on the simple feature vectors.
UR - http://www.scopus.com/inward/record.url?scp=84926218873&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84926218873&partnerID=8YFLogxK
U2 - 10.1145/2701126.2701197
DO - 10.1145/2701126.2701197
M3 - Conference contribution
AN - SCOPUS:84926218873
T3 - ACM IMCOM 2015 - Proceedings
BT - ACM IMCOM 2015 - Proceedings
PB - Association for Computing Machinery, Inc
T2 - 9th International Conference on Ubiquitous Information Management and Communication, ACM IMCOM 2015
Y2 - 8 January 2015 through 10 January 2015
ER -