ELSNET-list archive

Category:   E-CFP
From:   Savitha Srinivasan
Email:   savitha_(on)_almaden.ibm.com
Date received:   24 Sep 2001
Deadline:   20 Mar 2002

CALL FOR PAPERS EURASIP JOURNAL ON APPLIED SIGNAL PROCESSING Special Issue on: Unstructured Information Management from Multimedia Data Sources The recent proliferation of the world-wide web and the low cost of storage have contributed to an exposively growing volume of multimedia data. With such huge amounts of data available the next question is how to make use of it. Making use of structured data is simple, for example a database with indexing and search capability; however, most of the available data is unstructured and unorganized. Though it is possible to convert such data sources (examples are newswire text, broadcast video and audio, recorded call-center conversations, etc.) to structured form by manual processing, the high cost associated with this enables only a very small portion of the data to be processed in this fashion., There is often very valuable information buried in such unstructured data (for instance, call-center data may contain information about customer trends). Consequently, there is a great deal of research and commercial value in developing methods that automatically extract and manage information that is present in such unstructured data sources. In this special issue we focus on such methods. In general, prior approaches for information extraction and management start from streams of text. Though researchers have been working on this problem for several years, there are still several open issues and problems. An additional limitation of the applicability of these approaches to our problem is that quite often, unstructured multimedia data sources are not available in text form. Consequently the difficulty of the problem is increased by the need to deal with non-text data. With the recent advances in automatic speech recognition (ASR) capability, one possible solution is to transcribe speech sources and feed the transcriptions to text based schemes. However, the errorful, sometimes ungrammatical transcriptions produced by ASR systems further handicap traditional information retrieval techniques which have been developed to work with perfect transcriptions. The postive aspect of starting from non-text data, however, is that the non-text data contains information that is not available in the text, such as the change of prosody of a speaker, scene changes between adjacent video frames, etc. Consequently, there are several open areas of research relating to the use of traditional text-based information retrieval and management methods with multimedia sources, and the complementary problem of using information other than text to augment the performance of traditional methods. This special issue will represent a vehicle whereby researchers can present new studies and applications of information retrieval and extraction techniques from multimedia sources, thus paving the way for future developments in the field and for a better understanding of its potential. Prospective papers should be unpublished and present solid research work offering innovative contributions either from a methodological or application point of view. Topics of interest include (but are not limited to): 1. Information management from speech sources, including spoken document retrieval, information extraction (named entities and other quantities of interests), and categorization of speech documents 2. Information management from video/image sources, including semantic indexing and retrieval of video clips, and automatic annotation and categorization of video and images, and content-based image retrieval 3. Information management from text, including automatic text categorization, information extraction (named entities), and question and answering methods for retrieving information from databases Authors should follow the EURASIP JASP manuscript format described at the Journal site http://asp.hindawi.com/. Prospective authors should submit an electronic copy of their complete manuscript through the EURASIP JASP's web submission system at http://asp.hindawi.com/, according to the following timetable: Manuscript Due March 20, 2002 Acceptance Notification Sep 30, 2002 Regards, Savitha Srinivasan Manager, Multimedia Content Distribution IBM Almaden Researc Center Phone: 408-927-1430

[print/pda] [no frame] [navigation table] [navigation frame]     Page generated 14-02-2008 by Steven Krauwer Disclaimer / Contact ELSNET