1,721,034 research outputs found
Extremal formulations for elastic and elastic-plastic analysis by boundary integral equations
No full textA Learning Algorithm for Web Page Scoring Systems
No full textHyperlink analysis is a successful approach to define algorithms which compute the relevance of a document on the basis of the citation graph. In this paper we propose a technique to learn the parameters of the page ranking model using a set of pages labeled as relevant or not relevant by a supervisor. In particular we describe a learning algorithm applied to a scheme similar to PageRank. The ranking algorithm is based on a probabilistic Web surfer model and its parameters are optimized in order to increase the probability of the surfer to visit a page labeled as relevant and to reduce it for the pages labeled as not relevant. The experimental results show the effectiveness of the proposed technique in reorganizing the page ordering in the ranking list accordingly to the examples provided in the learning set
Implementation of a symmetric boundary element method in transient heat conduction with semi-analytical integrations
No full textHidden Tree Markov Models for Image Document Classification
No full textClassification is an important problem in image document processing and is often a preliminary step toward recognition, understanding, and information extraction. In this paper, the problem is formulated in the framework of concept learning and each category corresponds to the set of image documents with similar physical structure. We propose a solution based on two algorithmic ideas. First, we obtain a structured representation of images based on labeled XY-trees (this representation informs the learner about important relationships between image subconstituents). Second, we propose a probabilistic architecture that extends hidden Markov models for learning probability distributions defined on spaces of labeled trees. Finally, a successful application of this method to the categorization of commercial invoices is presented
Adaptive graphical pattern recognition: The joint role of structure, learning
No full textIn this paper we introduced methodological and practical issues on adaptive graphical pattern recognition, a new approach to process patterns for which neither a purely symbolic nor a purely sub-symbolic representation seems to be adequate. Adaptive graphical pattern recognition is based on appropriate graphical representations of patterns which are subsequently processed by recursive neural networks, equipped with connectionist-based learning algorithms.
The preliminary ideas and results given in [1] are extended and some properties, like the capabilities of incorporating scale and rotation invariance and dealing with noise, are emphasized and related to competing approaches. It turns out that adaptive graphical pattern recognition bridges nicely the gap between traditional connectionist models and syntactic pattern recognition and appears to be a new challenging approach to many pattern recognition problems
Classification of HTML documents by Hidden Tree-Markov Models
No full textContent-based search and organization of Web documents poses new issues in information retrieval. We propose a novel approach for the classification of HTML documents based on a structured representation of their contents which are split into logical contexts (paragraphs, sections, anchors, etc.). The classification is performed using Hidden Tree-Markov Models (HTMMs), an extension of Hidden Markov Models for processing structured objects. We report some promising experimental results showing that the use of the structured representation improves the classification accuracy in most of the cases
Automatic Generation of Crossword Puzzles
No full textCrossword puzzles are used everyday by millions of people for entertainment, but have ap- plications also in educational and rehabilitation contexts. Unfortunately, the generation of ad-hoc puzzles, especially on specific subjects, typically requires a great deal of human expert work. This paper presents the architecture of WebCrow-generation, a system that is able to generate crosswords with no human intervention, including clue generation and crossword compilation. In particular, the proposed system crawls information sources on the Web, extracts definitions from the downloaded pages using state-of-the-art natural language processing techniques and, finally, compiles the crossword schema with the ex- tracted definitions by constraint satisfaction programming. The system has been tested on the creation of Italian crosswords, but the extensive use of machine learning makes the system easily portable to other languages
- …
