关于LDA的文章

转：http://www.zhizhihu.com/html/y2011/3228.html

l  Theory

n  Introduction

u  Unsupervised learning by probabilistic latent semantic analysis.

u  Latent dirichlet allocation.

u  Finding scientific topics.

u  Rethinking LDA: Why Priors Matter

u  On an equivalence between PLSI and LDA

n  Variations

u  Correlated Topic Models.

u  Hierarchical topic models and the nested Chinese restaurant process.

u  Hierarchical Dirichlet processes.

u  Nonparametric Bayes pachinko allocation.

u  Topic Models with Power-Law Using Pitman-Yor Process

u  Supervised topic models.

u  Topic Models Conditioned on Arbitrary Features withDirichlet-multinomial Regression

u  Discriminative Topic Modeling based on Manifold Learning

u  Interactive Topic Modeling

u  Mixtures of hierarchical topics with pachinko allocation

u  Incorporating domain knowledge into topic modeling via DirichletForest priors

u  Conditional topic random fields

u  Markov random topic fields

u  A two-dimensional topic-aspect model for discovering multi-facetedtopics

u  Generalized component analysis for text with heterogeneousattributes

n  Inference

u  Gibbs Sampling:

l  Finding scientific topics.

l  Parameter estimation for text analysis

l  Fast collapsed gibbs sampling for latent dirichlet allocation

l  Distributed inference for latent dirichlet allocation

u  Variational EM

l  Latent dirichlet allocation.

n  Evaluation

u  Reading tea leaves: How humans interpret topic models.

u  Evaluation Methods for Topic Models

n  Online learning and scalability

u  On-line LDA: Adaptive topic models for mining text streams withapplications to topic detection and tracking

u  Online variational inference for the hierarchical Dirichlet process.

u  Online Learning for Latent Dirichlet Allocation

u  Efficient Methods for Topic Model Inference on Streaming DocumentCollections

u  Online Multiscale Dynamic Topic Models

l  Applications

n  Classification

u  DiscLDA: Discriminative learning for dimensionality reduction andclassification

u  Labeled LDA: A supervised topic model for credit attribution inmulti-labeled corpora

u  MedLDA: maximum margin supervised topic models for regression andclassification

n  Clustering

n  Network data(social network) mining

u  Link-PLSA-LDA: A new unsupervised model for topics and influence ofblogs

u  Connections between the lines: augmenting social networks with text

u  Relational topic models for document networks

u  Topic and role discovery in social networks with experiments onenron and academic email

u  Group and topic discovery from relations and text

u  Probabilistic models for discovering e-communities

u  Arnetminer: Extraction and mining of academic social networks

u  Community evolution in dynamic multi-mode networks

u  An LDA-based community structure discovery approach for large-scalesocial networks

u  Probabilistic community discovery using hierarchical latent gaussianmixture model

u  Modeling Evolutionary Behaviors for Community-based DynamicRecommendation

u  Joint group and topic discovery from relations and text

u  Social topic models for community extraction

u  Combining link and content for community detection: a discriminativeapproach

u  Topic-Link LDA: Joint Models of Topic and Author Community

u  Modeling hidden topics on document manifold

u  Topic Modeling with Network Regularization

u  Mining Topic-Level Influence in Heterogeneous Networks

u  Utilizing Context in Generative Bayesian Models for Linked Corpus

u

n  Sentiment analysis and opinion mining

u  Rated aspect summarization of short comments.

u  Learning document-level semantic properties from free-textannotations.

u  Joint sentiment/topic model for sentiment analysis.

u  Mining multi-faceted overviews of arbitrary topics in a textcollection

u  Modeling online reviews with multi-grain topic models

u  Topic sentiment mixture: modeling facets and opinions in weblogs.

u  Multiple aspect ranking using the good grief algorithm.

u  A joint model of text and aspect ratings for sentiment summarization.

u  Opinion integration through semi-supervised topic modeling

u  Holistic Sentiment Analysis Across Languages: MultilingualSupervised Latent Dirichlet Allocation.

u  Latent Aspect Rating Analysis on Review Text Data: A RatingRegression Approach

u  Aspect and Sentiment Unification Model for Online Review Analysis

u  An unsupervised aspect-sentiment model for online reviews

u  Jointly Modeling Aspects and Opinions with a MaxEnt-LDA Hybrid.

n  Evolutionary text stream mining

u  Discovering evolutionary theme patterns from text: an exploration oftemporal text mining

u  Topics over time: a non-markov continuous-time model of topicaltrends

u  Topic models over text streams: A study of batch and onlineunsupervised learning

u  Mining correlated bursty topic patterns from coordinated textstreams

u  Topic Evolution in a stream of Documents

u  Evolutionary Hierarchical Dirichlet Processes for MultipleCorrelated Time-varying Corpora

u  Studying the history of ideas using topic models

u  Mining common topics from multiple asynchronous text streams.

u  Mining Correlated Bursty Topic Patterns from Coordinated TextStreams

n  Temporal and spatial data analysis

u  A latent variable model for geographic lexical variation.

u  Dynamic topic models

u  A probabilistic approach to spatiotemporal theme pattern mining onweblogs

u  Continuous time dynamic topic models

u  Dynamic mixture models for multiple time series

u  On-Line LDA: Adaptive Topic Models for Mining Text Streams

u  Topic models over text streams: A study of batch and onlineunsupervised learning

u  Spatial latent dirichlet allocation

n  Scientific publication mining

u  Finding scientific topics.

u  The author-topic model for authors and documents.

u  Statistical entity-topic models

u  Probabilistic author-topic models for information discovery

u  The author-recipient-topic model for topic and role discovery insocial networks

u  Expertise modeling for matching papers with reviewers

u  Topic evolution and social interactions: how authors effect research

u  Joint latent topic models for text and citations

u  Co-ranking authors and documents in a heterogeneous network

u  Mixed-membership models of scientific publications

u  Modeling individual differences using Dirichlet processes

u  Multi-aspect expertise matching for review assignment

u  Topic-link LDA: joint models of topic and author community

u  Group and topic discovery from relations and their attributes

u  Exploiting Temporal Authors Interests via Temporal-Author-TopicModeling, ADMA 2009

u  Topic and Trend Detection in Text Collections Using Latent DirichletAllocation, ECIR 2009

u  Mining a digital library for influential authors.

u  Bibliometric Impact Measures Leveraging Topic Analysis.

u  Context-aware Citation Recommendation

u  Detecting Topic Evolution in Scientific Literature: How CanCitations Help?

u  Latent Interest-Topic Model: Finding the causal relationships behinddyadic data

u  A topic modeling approach and its integration into the random walkframework for academic search

n  Web data mining

u  Latent topic models for hypertext

n  Information retrieval

u  LDA-based document models for ad-hoc retrieval

u  Exploring social annotations for information retrieval

u  Modeling general and specific aspects of documents with a probabilistictopic model

u  Exploring topic-based language models for effective web informationretrieval

u  Probabilistic Models for Expert Finding

n  Information extraction

u  Employing Topic Models for Pattern-based Semantic Class Discovery

u  Combining Concept Hierarchies and Statistical Topic Models

u  A Probabilistic Approach for Adapting Information ExtractionWrappers and Discovering New Attributes

u  An Unsupervised Framework for Extracting and Normalizing ProductAttributes from Multiple Web Sites

u  Learning to Adapt Web Information Extraction Knowledge andDiscovering New Attributes via a Bayesian Approach

u  Adapting Web Information Extraction Knowledge via Mining SiteInvariant and Site Dependent Features

u  Learning to Extract and Summarize Hot Item Features from MultipleAuction Web Sites"

u  Semi-supervised Extraction of Entity Aspects Using Topic Models

n  Annotations(or Tagging, Labeling) and recommendation

u  Automatic labeling of multinomial topic models.

u  Context modeling for ranking and tagging bursty features in textstreams.

u  Learning document-level semantic properties from free-textannotations.

u  Generating summary keywords for emails using topics

u  Semantic Annotation of Frequent Patterns

u  Latent dirichlet allocation for tag recommendation

u  Tag-LDA for Scalable Real-time Tag Recommendation

u  The Topic-Perspective Model for Social Tagging Systems

u  A Probabilistic Topic-Connection Model for Automatic ImageAnnotation

u  Clustering the Tagged Web

n  Summarization

u  Topical keyphrase extraction from twitter.

u  Bayesian query-focused summarization

u  Topic-based multi-document summarization with probabilistic latentsemantic analysis

u  Multi-topic based Query-oriented Summarization

u  Multi-Document Summarization using Sentence-based Topic Models

u  Generating Impact-Based Summaries for Scientific Literature

u  Generating Comparative Summaries of Contradictory Opinions in Text

u  Rated Aspect Summarization of Short Comments

u  A Hybrid Hierarchical Model for Multi-Document Summarization

u  GENERATING TEMPLATES OF ENTITY SUMMARIES WITH AN ENTITY-ASPECT MODELAND PATTERN MINING

u  Latent dirichlet allocation and singular value decomposition basedmulti-document summarization

n  Social media mining

u  A latent variable model for geographic lexical variation.

u  Empirical study of topic modeling in twitter.

u  Characterizing micorblogs with topic models.

u  TwitterRank: finding topic-sensitive influential twitterers.

u  Comparing twitter and traditional media using topic models.

n  DB

u  Topic cube: Topic modeling for olap on multidimensional textdatabases

n  NLP tasks

u  A topic model for word sense disambiguation

u  Syntactic topic models

u  Integrating topics and syntax

u  Topic modeling: beyond bag-of-words

u  A Bayesian LDA-based model for semi-supervised part-of-speechtagging

u  Topical n-grams: Phrase and topic discovery, with an application toinformation retrieval

u  A topic model for word sense disambiguation

u  Named entity recognition in query

u  Multilingual topic models for unaligned text.

u  Markov topic models.

u  Modeling Syntactic Structures of Topics with a Nested HMM-LDA

u  Topic segmentation with an aspect hidden Markov model.

u  Polylingual Topic Models

u  A Latent Dirichlet Allocation method for Selectional Preferences

u  Improving word sense disambiguation using topic features

u  Cross-Lingual Latent Topic Extraction

u  Exploiting conversation structure in unsupervised topic segmentationfor emails

u  TOPIC MODELS FOR WORD SENSE DISAMBIGUATION AND TOKEN-BASED IDIOM

DETECTION