We present the notion of Ranking for evaluation of two-class classifiers. Ranking is based on using the ordering information contained in the output of a scoring model, rather than just setting a classification threshold. Using this ordering information, we can evaluate the model's performance with regard to complex goal functions, such as the cor rect identification of the k most likely and/or...

Analysis of ranking data is often required in various fields of study, for example politics, market research and psychology. Over the years, many statistical models for ranking data have been developed. Among them, distance-based ranking models postulate that the probability of observing a ranking of items depends on the distance between the observed ranking and a modal ranking. The closer to t...

In this paper, we present our approach in the We Want Web(WWW)[1] task of NTCIR-13, for both English and Chinese languages. We implement a ranking model for traditional re-ranking problems based on learning to rank. We first process the raw data and extract text features, match features, embedding features and semantic features for each query-document pair. Then we use LamdaMART[2] to train the...

research centers are among the most important institutes in a scientific society. using the altavista search engine and webometric methods, this research tries to find the performance and impact of the top research centers of the islamic world countries. the results reveal that from 57 countries, 40 of them did not have any research centers scored in webometric ranking and the rest of them had ...

efficiency calculations for units belonging to one group according to data envelopment analysis widely have received attention; while experiences related to efficiency comparison of several groups consisted of decision making units is restricted. kao's parallel system network dea model that has been designated aimed at such a comparison, practically leads to accomplish comparison among uni...