TAILIEUCHUNG - Báo cáo khoa học: "A Hierarchical Model of Web Summaries"

We investigate the relevance of hierarchical topic models to represent the content of Web gists. We focus our attention on DMOZ, a popular Web directory, and propose two algorithms to infer such a model from its manually-curated hierarchy of categories. Our first approach, based on information-theoretic grounds, uses an algorithm similar to recursive feature selection. Our second approach is fully Bayesian and derived from the more general model, hierarchical LDA. | A Hierarchical Model of Web Summaries Yves Petinot and Kathleen McKeown and Kapil Thadani Department of Computer Science Columbia University New York NY 10027 ypetinot kathy kapil @ Abstract We investigate the relevance of hierarchical topic models to represent the content of Web gists. We focus our attention on DMOZ a popular Web directory and propose two algorithms to infer such a model from its manually-curated hierarchy of categories. Our first approach based on information-theoretic grounds uses an algorithm similar to recursive feature selection. Our second approach is fully Bayesian and derived from the more general model hierarchical LDA. We evaluate the performance of both models against a flat 1-gram baseline and show improvements in terms of perplexity over held-out data. 1 Introduction The work presented in this paper is aimed at leveraging a manually created document ontology to model the content of an underlying document collection. While the primary usage of ontologies is as a means of organizing and navigating document collections they can also help in inferring a significant amount of information about the documents attached to them including path-level statistical representations of content and fine-grained views on the level of specificity of the language used in those documents. Our study focuses on the ontology underlying DMOZ1 a popular Web directory. We propose two methods for crystalizing a hierarchical topic model against its hierarchy and show that the resulting models outperform a flat unigram model in its predictive power over held-out data. 1http 670 To construct our hierarchical topic models we adopt the mixed membership formalism Hofmann 1999 Blei et al. 2010 where a document is represented as a mixture over a set of word multinomials. We consider the document hierarchy H . the DMOZ hierarchy as a tree where internal nodes category nodes and leaf nodes documents as well as the edges connecting them are .

• Báo cáo khoa học
• A Hierarchical Model
• Web Summaries
• Long Papers
• báo cáo khoa học
• báo cáo ngôn ngữ
• ngôn ngữ tự nhiên
• Measuring efficiency of a university faculty
• A hierarchical network data envelopment analysis model
• Hierarchical network data
• Relatively efficient since
• Data envelopment analysis
• Efficiency measurement model
• Network Model
• Hierarchical Data Model
• Relational Model
• Hierarchical Model
• Database Schema
• Physical Data Independence
• LAN Switching
• Hierarchical Network Supports
• LAN Design
• Hierarchical network design model
• Hierarchical network model
• CCNA Exploration 4
• Lecture CCNA Exploration 4
• A Topic Similarity Model
• Hierarchical Phrase based Translation
• Xinyan Xiao
• Pattern Learning
• Relation Extraction
• Hierarchical Topic Model
• Translation Model Size Reduction
• Hierarchical Phrase based
• Statistical Machine Translation
• A Hierarchical Nonparametric Model using Speaker Identity
• Topic Segmentation
• Multiparty Conversations
• A Discriminative Hierarchical Model
• Fast Coreference
• Large Scale
• A Hybrid Hierarchical Model
• Multi Document Summarization
• Asli Celikyilmaz
• A Joint Rule Selection Model
• Dictionary Definitions
• Homograph Identification
• Generative Hierarchical Model
• xử lý ngôn ngữ tự nhiên
• A Hierarchical Bayesian
• Language Model
• Pitman Yor Processes
• mô hình ngôn ngữ
• A Hierarchical Phrase Based Model
• David Chiang
• Evaluation of lecturers’ performance
• Novel hierarchical multi criteria model
• Interval complex Neutrosophic set
• Interval valued complex neutrosophic set environment
• Interval valued complex neutrosophic
• Tehran Stock Exchange
• The hierarchical theory
• Enterprise life cycle model
• Net liabilities and retained earnings
• Long term asset
• Public health
• Integrated nested laplace approximation
• Latent gaussian model
• Bayesian hierarchical model
• Identifying risk factors
• Hierarchical Non Emitting
• Markov Models
• Eric Sven Ristad
• scientific reports
• model language
• process natural language
• Probabilistic Hierarchical Clustering
• Morphological Paradigms
• Burcu Can
• Hierarchical Bayesian
• Language Modelling
• the Linguistically Informed
• Rule Filtering by Pattern
• Efficient Hierarchical Translation
• Eduardo R
• Banga
• Mistake Driven Mixture
• Hierarchical Tag Context Trees
• Masahiko Haruno
• An Automatic Inserter System
• Hierarchical Lexica