Many of the problems that occur in long-running systems involve the way that the system uses memory. We have developed a framework for extracting and building a model of the heap ...
In this paper, we consider the extraction of speaker identity from audio records of broadcast news without a priori acoustic information about speakers. Using an automatic speech ...
Vincent Jousse, Simon Petit-Renaud, Sylvain Meigni...
—In this paper, we describe a flexible form-reader system capable of extracting textual information from accounting documents, like invoices and bills of service companies. In th...
Francesca Cesarini, Marco Gori, Simone Marinai, Gi...
As the proliferation of the Internet, especially World Wide Web, numerous information resources have been constructed. The characteristics of information resources on the Internet...
Kangchan Lee, Jae Hong Min, Kishik Park, Kyuchul L...
Position information has been proved to be very effective in document summarization, especially in generic summarization. Existing approaches mostly consider the information of se...