BibSonomy publications for /author/CravenThu Feb 02 16:13:22 CET 2012Artificial Intelligence1–269 - 113Learning to construct knowledge bases from the World Wide Web1182000ai bachelor:2011:bachmann naiveBayes webclassificatoion www The World Wide Web is a vast source of information accessible to computers, but understandable only to humans. The goal of the research described here is to automatically create a computer understandable knowledge base whose content mirrors that of the World Wide Web. Such a knowledge base would enable much more effective retrieval of Web information, and promote new uses of the Web to support knowledge-based inference and problem solving. Our approach is to develop a trainable information extraction system that takes two inputs. The first is an ontology that defines the classes (e.g., company, person, employee, product) and relations (e.g., employed_by, produced_by) of interest when creating the knowledge base. The second is a set of training data consisting of labeled regions of hypertext that represent instances of these classes and relations. Given these inputs, the system learns to extract information from other pages and hyperlinks on the Web. This article describes our general approach, several machine learning algorithms for this task, and promising initial results with a prototype system that has created a knowledge base describing university people, courses, and research projects.Tue Dec 27 00:00:00 CET 2011CNSMconf/cnsm/20111-9Policy refinement: Decomposition and operationalization for dynamic domains.2011dblp Fri Dec 23 00:00:00 CET 2011AAAI/IAAIconf/aaai/1998509-516Learning to Extract Symbolic Knowledge from the World Wide Web.1998dblp Fri Dec 23 00:00:00 CET 2011ACM Multimediaconf/mm/200067-74Temporal links: recording and replaying virtual environments.2000dblp Fri Dec 23 00:00:00 CET 2011MLconf/icml/1991213-217Constructive Induction in Knowledge-Based Neural Networks.1991dblp Fri Dec 23 00:00:00 CET 2011ICMLconf/icml/199437-45Using Sampling and Queries to Extract Rules from Trained Neural Networks.1994dblp Fri Dec 23 00:00:00 CET 2011ICMLconf/icml/199373-80Learning Symbolic Rules Using Artificial Neural Networks.1993dblp Wed Dec 21 00:00:00 CET 2011CHIconf/chi/200130-37Exploiting interactivity, influence, space and time to explore non-linear drama in virtual worlds.2001dblp Thu Dec 15 00:00:00 CET 2011UAIconf/uai/2008444-451Learning Hidden Markov Models for Regression using Path Aggregation.2008dblp Mon Nov 28 00:00:00 CET 2011Presence135-50Patterns of Network and User Activity in an Inhabited Television Event.102001dblp Fri Nov 25 00:00:00 CET 2011Encyclopedia of Optimizationreference/opt/20091770-1774Invexity and its Applications.2009dblp Wed Nov 09 00:00:00 CET 2011Adv. Internet of Things213-26Side-Channel Analysis for Detecting Protocol Tunneling.12011dblp Tue Nov 08 00:00:00 CET 2011First Monday4Where does Web bibliographies' author information come from?102005dblp Tue Nov 08 00:00:00 CET 2011First Monday10Changes in Metatag Descriptions Over Time.62001dblp Mon Oct 24 00:00:00 CEST 2011FPLconf/fpl/2011518-523Improved Abstractions and Turnaround Time for FPGA Design Validation and Debug.2011dblp Tue Oct 11 19:16:07 CEST 2011Fluid Phase Equilibria1-2167 - 187The Area method for pure fluids and an analysis of the two-phase region13019971997 equation-of-state equilibrium two-phase The Area method, developed recently for solving multicomponent phase equilibrium problems, has been extended to pure fluids. The method is based on maximizing a single objective function in the Helmholtz-volume surface along any given isotherm, which reduces the number of independent variables to only two: the saturated liquid and vapour volumes. Two techniques are employed to find the maximum of the objective function, the integral and iterative. The integral always finds the thermodynamically stable solution without any prior assumptions about the values of the molar volumes. This factor distinguishes the integral from the iterative technique and also from methods based on the Maxwell equal-area principle. The method has been applied to a group of high accuracy non-cubic equations of state and some of the thermodynamic inconsistencies which occur inside the two-phase region are explored. A new inequality constraint which eliminates these inconsistencies during the development of new equations of state is proposed, and initial results with fitting a preliminary Helmholtz equation of state for benzene are encouraging.Thu Sep 15 00:00:00 CEST 2011IWCMCconf/iwcmc/20111310-1314Noise tolerant symbolic learning of Markov models of tunneled protocols.2011dblp Tue Aug 09 00:00:00 CEST 2011IJCAIconf/ijcai/20111171-1177A Framework for Incorporating General Domain Knowledge into Latent Dirichlet Allocation Using First-Order Logic.2011dblp Mon Aug 08 00:00:00 CEST 2011Regulatory Genomicsconf/recomb/2004rrg52-64Lecture Notes in Computer ScienceLearning Regulatory Network Models that Represent Regulator States and Roles.33182004dblp Wed Aug 03 00:00:00 CEST 2011J. Comb. Theory, Ser. A2174-181An Application of Pólya's Theory of Counting to an Enumeration Problem Arising in Quadratic Form Theory.291980dblp