Abstract
| Language | English |
|---|---|
| Title of host publication | Progress in WWW Research and Development Lecture Notes in Computer Science |
| Pages | 432-437 |
| Publication status | Published - 2008 |
Fingerprint
Cite this
}
The Experiments with the Linear Combination Data Fusion Method in Information Retrieval. / Wu, Shengli; Bi, Yaxin; Zeng, Xiaoqin; Han, Lixin.
Progress in WWW Research and Development Lecture Notes in Computer Science. 2008. p. 432-437.Research output: Chapter in Book/Report/Conference proceeding › Chapter
TY - CHAP
T1 - The Experiments with the Linear Combination Data Fusion Method in Information Retrieval
AU - Wu, Shengli
AU - Bi, Yaxin
AU - Zeng, Xiaoqin
AU - Han, Lixin
PY - 2008
Y1 - 2008
N2 - In data fusion, the linear combination method is a very flexible method since different weights can be assigned to different systems. However, it remains an open question that which weighting schema is good. In many cases, a simple weighting schema was used: for a system, its weight is assigned as its average performance over a group of training queries. In this paper, we empirically investigate the weighting issue. We find that, a series of power functions of average performance, which can be implemented as efficiently as the simple weighting schema, is more effective than the simple weighting schema for data fusion. We also investigate combined weights which concern both performance of component results and dissimilarity among component results. Further performance improvement on data fusion is achievable by using the combined weights.
AB - In data fusion, the linear combination method is a very flexible method since different weights can be assigned to different systems. However, it remains an open question that which weighting schema is good. In many cases, a simple weighting schema was used: for a system, its weight is assigned as its average performance over a group of training queries. In this paper, we empirically investigate the weighting issue. We find that, a series of power functions of average performance, which can be implemented as efficiently as the simple weighting schema, is more effective than the simple weighting schema for data fusion. We also investigate combined weights which concern both performance of component results and dissimilarity among component results. Further performance improvement on data fusion is achievable by using the combined weights.
M3 - Chapter
SN - 978-3-540-78848-5
SP - 432
EP - 437
BT - Progress in WWW Research and Development Lecture Notes in Computer Science
ER -