Author name cluster

Frank Klassner

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

1 author row

AIJ Journal 2000 Journal Article

BIG: An agent for resource-bounded information gathering and decision making

Victor Lesser
BRYAN HORLING
Frank Klassner
Anita Raja
Thomas Wagner
Shelley XQ. Zhang

The World Wide Web has become an invaluable information resource but the explosion of available information has made Web search a time consuming and complex process. The large number of information sources and their different levels of accessibility, reliability and associated costs present a complex information gathering control problem. This paper describes the rationale, architecture, and implementation of a next generation information gathering system—a system that integrates several areas of Artificial Intelligence research under a single umbrella. Our solution to the information explosion is an information gathering agent, BIG, that plans to gather information to support a decision process, reasons about the resource trade-offs of different possible gathering approaches, extracts information from both unstructured and structured documents, and uses the extracted information to refine its search and processing activities.

Details DOI

AAAI Conference 1998 Conference Paper

BIG: A Resource-Bounded Information Gathering Agent

Victor Lesser
Frank Klassner
Thomas Wagner

Effective information gathering on the WWW is a complex task requiring planning, scheduling, text processing, and interpretation-style reasoning about extracted data to resolve inconsistencies and to refine hypotheses about the data. This paper describes the rationale, architecture, and implementation of a next generation information gathering system – a system that integrates several areas of AI research under a single research umbrella. The goal of this system is to exploit the vast number of information sources available today on the NII including a growing number of digital libraries, independent news agencies, government agencies, as well as human experts providing a variety of services. The large number of information sources and their different levels of accessibility, reliability and associated costs present a complex information gathering coordination problem. Our solution is an information gathering agent, BIG, that plans to gather information to support a decision process, reasons about the resource tradeoffs of different possible gathering approaches, extracts information from both unstructured and structured documents, and uses the extracted information to refine its search and processing activities.

PDF Details

AAAI Conference 1998 Conference Paper

The Role of Data Reprocessing in Complex Acoustic Environments

Frank Klassner

The Integrated Processing and Understanding of Signals (IPUS) architecture is a general blackboard framework for structuring bidirectional interaction between front-end signal processing algorithms (SPAs) and signal understanding processes. To date, reported work on the architecture has focused on proof-of-concept demonstrations of how well a sound-understandingtestbed(SUT) basedon IPUS coulduse small libraries of sound models and small sets of SPAs to analyze acoustic scenarios. In this paper we evaluate how well the architecture scales up to more complex environments. We describe knowledge-representation and control-strategy issuesinvolvedin scaling upan IPUS-basedSUT for usewith a library of 40 sound models, and present empirical evaluation that shows (a) the IPUS data reprocessing paradigm can increase interpretation accuracy by 25% – 50% in complex scenarios, and (b) the benefit increases with increasing complexity of the environment.

PDF Details

AAAI Conference 1997 Conference Paper

Combining Approximate Front End Signal Processing with Selective Reprocessing in Auditory Perception

Frank Klassner

When dealing with signals from complex environments, where multiple time-dependent signal signatures can interfere with each other in stochastically unpredictable ways, traditional perceptual systems tend to fall back on a strategy of always performing finelydetailed, costly analysis of the signal with a comprehensive front end set of signal processing algorithms (SPAS), whether or not the current scenario requires the extra detail. Approximate SPAS (ASPAs) - algorithms whose processing time can be limited in order to trade off precision in their outputs for reduced execution time - can play a role in producing adaptive, less-costly front ends, but their outputs tend to require context-dependent analysis for use as evidence in interpretation. This paper examines the IPUS (Integrated Processing and Understanding of Signals) architecture’s ability to serve as a support framework for applying ASPAs in interpretation problems. Specifically, our work shows that it is feasible to include an approximate version of the Short-Time Fourier Transform in an IPUS-based sound-understanding testbed.

PDF Details

AAAI Conference 1993 Conference Paper

IPUS: An Architecture for Integrated Signal Processing and Signal Interpretation in Complex Environments

Victor Lesser
Frank Klassner

This paper presents the IPUS (Integratecl Processing and Understanding of Signals) architecture to address the traditional perceptual paradigm’ s shortcomings in complex environments. It has two premises: (1) the search for correct interpretations of signal processing algorithms’ (SPAS) outputs requires concurrent search for SPAS and control parameters appropriate for the environment, and (2) interaction between these search processes must be structured by a formal theory of how inappropriate SPA usage can distort SPA output. We describe IPUS’ s key components (discrepancy detection, diagnosis, reprocessing, and differential diagnosis) and their instantiation in an acoustic interpretation system. This application, along with another in the radar domain, supports our claim that the IPUS paradigm is feasible and generic.

PDF Details