fCombMNZ: An Improved Data Fusion Algorithm

Abstract

The data-fusion techniques have been investigated by many researchers and have been used in implementing several information retrieval systems. Fusion allows leveraging of the component systems in several ways by exploiting a number of effects, amongst them an effect called "Chorus effect"; this effect suggest that for a particular document if it is retrieved by two systems; it will be "better" than another document retrieved by only one system, and if it is retrieved by three systems it will be "better" than another document retrieved by one or two systems, and so on. "Better" means the document has higher probability to be relevant. We suggest including a new simple but effective rule called "fairness rule" within well known data-fusion algorithm called CombMNZ to allow this algorithm from leveraging the component systems in an effective way, this rule is the first attempt in the literature to manipulate the normalized score values for documents under certain circumstances to maximize the CombMNZ ability to benefit from the "chorus effect" without sacrificing its ability to benefit from another effect called "skimming effect", skimming effect; suggests that relevant documents are most likely to occur on the top of the retrieved list for any retrieval system. Our results show an improvement in the mean average precision (MAP) values over the traditional CombMNZ when we used our "fairness rule", the improvement for TREC3 groups ranged from 0.66% to 2.88%, and for TREC5 ranged from 0.2% to3.7%

Analysis of Meta-Search Engines using the Meta-Meta-Search tool SSIR

“…Further research has to be focussed on improving the ranking of results in fused data. Many researches have reported that combining retrieved results has improved results through effects like chorus effect, skimming effect and dark horse effect [8]. Our primary observation of fused results also draws to similar conclusion, though further analysis is required for a more conclusive assertion.…”

“…The results are fused and duplicates are removed. The 13 current system employs a simple ranking based on the number of occurrences of the URL, [8] based on the fact that a URL coming from multiple algorithms may be worth more and less susceptible to search engine spamming.…”

Performance prediction of multivariable linear regression based on the optimal influencing factors for ranking aggregation in crowdsourcing task

“…Wu and Mcclean (2006); pointed out that number of results, overlap rate among results, average and standard deviation of performance of all results can predict the performance and performance improvement of CombSUM and CombMNZ methods. Nassar and Kanaan (2009) maximized benefit of CombMNZ method from “chorus effect” by introducing “fairness principle” without sacrificing “skimming effect”. Based on statistical principles and sampling theory, Wu (2009) pointed out that the performance of submitted results, comparable scores, and sample distribution in sample space affect the aggregation performance.…”