S M M Verstappen, A Boonen, H Verkleij, J W J Bijlsma, E Buskens, J W G Jacobs, on behalf of the Utrecht Rheumatoid Arthritis Cohort Study Group (SRU) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...

Three papers in this session concern gathering spontaneous speech data. Two of the papers (AT&T and SRU) used a Wizard of Oz paradigm to interact with users via telephone. In the WOZ paradigm, subjects are led to believe that they are interacting with a machine when an experimenter is actually performing the machine's task. While this technique is used by many sites collecting spontaneous speec...

In this paper a fast and effective algorithm named equal feature variance sum (EFVS) framesynchronous searching is presented for state decoding. EFVS controls the state transition by using only the feature variance of the speech, instead of by using the state dwell distribution. The basic hypothesis of this new algorithm is the equality of feature variance sum in each state of the speech. Given...

A successful text categorization experiment divides a textual collection into pre-defined classes. A true representative for each class is generally obtained during training of the categorizer. In this paper, we report on our experiments on training and categorization of optically recognized documents. In particular, we will address the issues regarding the effects OCR errors may have on traini...

Information retrieval diiers signiicantly from function approximation in that the goal is for the system to achieve the same ranking function of documents relative to queries as the user: the outputs of the system relative to one another must be in the proper order. We hypothesize that a particular rank-order statistic, Guttman's point alienation, is the proper objective function for such a sys...