Trainability: developing a responsive learning system
Published 9 August 2003
Steven Minton, Sorinel I. Ticrea, Jennifer Beach
Citations9
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
It is shown that developers must design the learning system to insure that the teacher can train the system without becoming frustrated, and how trainability considerations constrain the type of learning algorithm that can be employed in a practical AI application.
Abstract
In this paper, we describe the lessons we learned in developing AgentBuilder, a commercial system for rapidly creating agents that extract information from web sites.
Keywords
Computer Science
Wrapper induction for information extraction
1,044 Citations1997Nicholas Kushmerick, Daniel S. Weld
This work introduces wrapper induction, a method for automatically constructing wrappers, and identifies hlrt, a wrapper class that is e(cid:14)ciently learnable, yet expressive enough to handle 48% of a recently surveyed sample of Internet resources.
Watch what I do: programming by demonstration
1,024 Citations1993Allen Cypher, Daniel C. Halbert +5 more
Part 1 Systems: Pygmalion tinker a predictive calculator rehearsal world smallStar peridot metamouse TELS eager garnet the Turvy experience chimera the geometer's sketchpad tourmaline a history-based macro by example system mondrian triggers the AIDE project.
ACM SIGMOD RecordA brief survey of web data extraction tools
690 Citations2002Alberto H. F. Laender, Berthier Ribeiro‐Neto +2 more
A taxonomy for characterizing Web data extraction fools is proposed, a survey of major web data extraction tools described in the literature is briefly surveyed, and a qualitative analysis of them is provided.
Autonomous Agents and Multi-Agent SystemsHierarchical Wrapper Induction for Semistructured Information Sources
368 Citations2001Ion Muslea, Steven Minton +1 more
This work introduces an inductive algorithm, STALKER, that generates high accuracy extraction rules based on user-labeled training examples that can wrap information sources that could not be wrapped by existing inductive techniques.
Artificial IntelligenceLearning one subprocedure per lesson
174 Citations1987Kurt VanLehn
Sierra is a program that learns procedures incrementally from examples, where an example is a sequence of actions, where a lesson is a set of examples that is guaranteed to introduce only one subprocedure.
Selective Sampling with Redundant Views
169 Citations2000Ion Muslea, Steven Minton +1 more
The most general algorithm in the cotesting family, naive co-testing, is analyzed, which can be used with virtually any type of learner and may also boost the classification accuracy.
International Journal of Cooperative Information SystemsTHE ARIADNE APPROACH TO WEB-BASED INFORMATION INTEGRATION
164 Citations2001Craig A. Knoblock, Steven Minton +5 more
This work has developed methods for mapping web sources into a uniform representation that makes it simple and efficient to integrate multiple sources and makes it easy to maintain these agents and incorporate new sources as they become available.
Studies in fuzziness and soft computingAccurately and Reliably Extracting Data from the Web: A Machine Learning Approach
132 Citations2003Craig A. Knoblock, Kristina Lerman +2 more
A set of tools for extracting data from web sites and transforming it into a structured data format, such as XML, so that the resulting data can be used to build new applications without having to deal with unstructured data.
Metamouse: specifying graphical procedures by example
98 Citations1989David Maulsby, Ian H. Witten +1 more
Close attention is paid to user interface aspects, and Metamouse helps the user by predicting and performing actions, thus reducing the tedium of repetitive graphical editing tasks.
Programming by demonstration
69 Citations1998Tessa Lau, Daniel S. Weld
This paper analyzes two application-independent methods for performing generalization that are based on well-understood machine learning technology, and argues that TGen vs has lower sample complexity, but TGen foil can learn a much more interesting class of programs.
Instructible information agents for Web mining
39 Citations2000Mathias Bauer, Dietmar Dengler +1 more
This paper deals with the problem of training an information agent to identify and extract interesting pieces of information from online documents.
A Dataflow Approach to Agent-based Information Management
11 Citations2000Greg Barish, Daniel DiPasquo +2 more
Theseus is described, an efficient plan execution system for information agents that gathers information from the Web, a domain where information retrieval is a problem that is network-bound and is often based on interleaved data gathering and navigation.
Specification-By-Demonstration: The ViCCS Interface
7 Citations1996Steven Minton, Andrew Philpot +1 more
This paper describes several design decisions that enable the VICSS (Visual Constraint Specification System) system to make combinatorial problem-solving technology available to users who have no formal training in AI or OR, and illustrates them in the context of an extended example.
PBD SYSTEMS: WHEN WILL THEY EVER LEARN?
5 Citations1996Ian H. Witten
The tragicomic courtship between programming by demonstration and machine learning, two fields that seem made for each other but have never quite got together, is traced and the long-term prognosis is not so much wedded bliss as assimilation.