Knowledge Transfer across Multilingual Corpora via Latent Topics
Lecture notes in computer sciencePublished 1 January 2011Open access
Wim De Smet, Jie Tang, Marie‐Francine Moens
Citations32
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
Experimental results on multilingual Wikipedia data show that the proposed topic model effectively discovers the topic information from the bilingual corpora, and the learned topics successfully transfer classification knowledge to other languages, for which no labeled training data are available.
Abstract
This paper explores bridging the content of two different languages via latent topics. Specifically, we propose a unified probabilistic model to simultaneously model latent topics from bilingual corpora that discuss comparable content and use the topics as features in a cross-lingual, dictionary-less text categorization task. Experimental results on multilingual Wikipedia data show that the proposed topic model effectively discovers the topic information from the bilingual corpora, and the learned topics successfully transfer classification knowledge to other languages, for which no labeled training data are available.
Keywords
Computer Science
Journal of Machine Learning ResearchLatent dirichlet allocation
27,049 Citations2003David M. Blei, Andrew Y. Ng +1 more
Probabilistic latent semantic indexing
3,916 Citations1999Thomas Hofmann
Probabilistic Latent Semantic Indexing is a novel approach to automated document indexing which is based on a statistical latent class model for factor analysis of count data.
arXiv (Cornell University)Probabilistic Latent Semantic Analysis
2,092 Citations2013Thomas Hofmann
This work proposes a widely applicable generalization of maximum likelihood model fitting by tempered EM, based on a mixture decomposition derived from a latent class model which results in a more principled approach which has a solid foundation in statistics.
Boosting for transfer learning
1,709 Citations2007Wenyuan Dai, Qiang Yang +2 more
This paper presents a novel transfer learning framework called TrAdaBoost, which extends boosting-based learning algorithms and shows that this method can allow us to learn an accurate model using only a tiny amount of new data and a large amount of old data, even when the new data are not sufficient to train a model alone.
Domain adaptation with structural correspondence learning
1,562 Citations2006John Blitzer, Ryan McDonald +1 more
This work introduces structural correspondence learning to automatically induce correspondences among features from different domains in order to adapt existing models from a resource-rich source domain to aresource-poor target domain.
Self-taught learning
1,533 Citations2007Rajat Raina, Alexis Battle +3 more
An approach to self-taught learning that uses sparse coding to construct higher-level features using the unlabeled data to form a succinct input representation and significantly improve classification performance.
The MIT Press eBooksMulti-Task Feature Learning
1,369 Citations2007Andreas A. Argyriou, Theodoros Evgeniou +1 more
Edinburgh Research ExplorerMulti-task Gaussian Process Prediction
851 Citations2007Edwin V. Bonilla, Kian Ming A. Chai +1 more
A model that learns a shared covariance function on input-dependent features and a "free-form" covariance matrix over tasks allows for good flexibility when modelling inter-task dependencies while avoiding the need for large amounts of data for training.
Co-training for cross-lingual sentiment classification
499 Citations2009Xiaojun Wan
A cotraining approach is proposed to making use of unlabeled Chinese data for cross-lingual sentiment classification, which leverages an available English corpus for Chinese sentiment classification by using the English corpus as training data.
Learning Multilingual Subjective Language via Cross-Lingual Projections
399 Citations2007Rada Mihalcea, Carmen Banea +1 more
This paper discusses learning multilingual subjective language via cross-lingual projections with a focus on English as a second language.
Polylingual topic models
338 Citations2009David Mimno, Hanna Wallach +3 more
This work introduces a polylingual topic model that discovers topics aligned across multiple languages and demonstrates its usefulness in supporting machine translation and tracking topic trends across languages.
Knowledge transfer via multiple model local structure mapping
329 Citations2008Jing Gao, Wei Fan +2 more
A locally weighted ensemble framework to combine multiple models for transfer learning, where the weights are dynamically assigned according to a model's predictive power on each test example, is proposed.
Cross-language information retrieval based on parallel texts and automatic mining of parallel texts from the Web
310 Citations1999Jian‐Yun Nie, Michel Simard +2 more
It is shown that using a probabilistic model, it is able to obtain performances close to those using an MT system, and the possibility of automatically gather parallel texts from the Web in an attempt to construct a reasonable training corpus is investigated.
Cross-Language Information Retrieval
226 Citations1998Gregory Grefenstette
This work focuses on the development of a model for automatic Cross-Language Information Retrieval using Latent Semantic Indexing and its application to Machine Translation Technology.
Multi-task feature and kernel selection for SVMs
221 Citations2004Tony Jebara
A multi-task representation learning approach using the maximum entropy discrimination formalism is derived and the resulting convex algorithms maintain the global solution properties of support vector machines.
Topic-bridged PLSA for cross-domain text classification
188 Citations2008Gui-Rong Xue, Wenyuan Dai +2 more
A novel cross-domain text classification algorithm which extends the traditional probabilistic latent semantic analysis (PLSA) algorithm to integrate labeled and unlabeled data, which come from different but related domains, into a unified probabilism model, called Topic-bridged PLSA, or TPLSA.
Learning a meta-level prior for feature relevance from multiple related tasks
165 Citations2007Su-In Lee, Vassil Chatalbashev +2 more
Transfer learning is used to construct an informative prior on feature relevance, and it is shown that learning feature relevance improves performance on two real data sets which illustrate such settings: predicting ratings in a collaborative filtering task, and distinguishing arguments of a verb in a sentence.
Mining multilingual topics from wikipedia
116 Citations2009Xiaochuan Ni, Jian-Tao Sun +2 more
This paper tries to leverage a large-scale and multilingual knowledge base, Wikipedia, to help effectively analyze and organize Web information written in different languages.
Cross-language information retrieval using PARAFAC2
97 Citations2007Peter A. Chew, Brett W. Bader +2 more
Artificial IntelligenceTranslingual information retrieval: learning from bilingual corpora
96 Citations1998Yiming Yang, Jaime Carbonell +2 more
The results show that using bilingual corpora for automated extraction of term equivalences in context outperforms dictionarybased methods and is comparable to that of other statistical corpus-based methods.
Cross-language linking of news stories on the web using interlingual topic modelling
56 Citations2009Wim De Smet, Marie‐Francine Moens
The Latent Dirichlet Allocation model is expanded to process documents in two languages and the validity of the learned interlingual topics in a document clustering task is demonstrated, where the evaluation is performed on Google News.
Information RetrievalCombining Multiple Strategies for Effective Monolingual and Cross-Language Retrieval
55 Citations2003Jacques Savoy
This paper describes and evaluates different retrieval strategies that are useful for search operations on document collections written in various European languages, namely French, Italian, Spanish and German and proposes a combined query translation approach that has resulted in interesting retrieval effectiveness.
Cross-language text classification
50 Citations2005J. Scott Olsson, Douglas W. Oard +1 more
The goal in cross-language text classification (CLTC) is to use English training data to classify Czech documents (although the concepts presented here are applicable to any language pair), and the authors are unaware of any previous work in this area.
Exploiting comparable corpora and bilingual dictionaries for cross-language text categorization
49 Citations2006Alfio Gliozzo, Carlo Strapparava
The approach is presented, providing a low cost solution for the Cross Language Text Categorization task, and when bilingual dictionaries are available the performance of the categorization gets close to that of monolingual text categorization.
Machine LearningA co-classification approach to learning from multilingual corpora
40 Citations2009Massih-Reza Amini, Cyril Goutte
A multiview learning, co-regularization approach is proposed, in which each language is considered as a separate source, and a joint loss is minimize that combines monolingual classification losses in each language while ensuring consistency of the categorization across languages.
Multilingual document clusters discovery
36 Citations2004B. Mathieu, Romaric Besançon +1 more
This paper proposes a cross-lingual similarity measure for the documents, using bilingual dictionaries, and presents an evaluation framework for this task, and analyzes and discusses the results obtained and proposes directions for future works.
NTCIRIntegration of PLSA into Probabilistic CLIR Model — Yokohama National University at NTCIR4 CLIR —
4 Citations2004Tetsu Muramatsu, Tatsunori Mori
An approximation method based on a two-phased retrieval model is introduced based on an integration of a probabilistic CLIR model and Probabilistic Latent Semantic Analysis in order to reduce the computational cost.