Articles

To build a coreference resolver for a new language, the typical approach is to first coreference-annotate documents from this tar- get language and then train a resolver on these annotated documents using supervised learn- ing techniques. However, the high cost asso- ciated with manually coreference-annotating documents needed by a supervised …...

Text input aids such as automatic correction systems play an increasingly important role in facilitating fast text entry and efficient com- munication between text message users. Al- though these tools are beneficial when they work correctly, they can cause significant communication problems when they fail. To improve its autocorrection performance, …...

The accuracy of many natural language pro- cessing tasks can be improved by a reranking step, which involves selecting a single output from a list of candidate outputs generated by a baseline system. We propose a novel fam- ily of reranking algorithms based on learning separate low-dimensional embeddings of the …...

We present a general framework containing a graded spectrum of Expectation Maximization (EM) algorithms called Unified Expectation Maximization (UEM.) UEM is parameterized by a single parameter and covers existing al- gorithms like standard EM and hard EM, con- strained versions of EM such as Constraint- Driven Learning (Chang et al., …...

We present novel methods to construct com- pact natural language lexicons within a graph- based semi-supervised learning framework, an attractive platform suited for propagating soft labels onto new natural language types from seed data. To achieve compactness, we induce sparse measures at graph vertices by incorporating sparsity-inducing penalties in Gaussian …...

Existing work in fine-grained sentiment anal- ysis focuses on sentences and phrases but ig- nores the contribution of individual words and their grammatical connections. This is because of a lack of both (1) annotated data at the word level and (2) algorithms that can leverage syn- tactic information in a …...

Microblog streams often contain a consider- able amount of information about local, re- gional, national, and global events. Most ex- isting microblog search capabilities are fo- cused on recent happenings and do not provide the ability to search and explore past events. This paper proposes the problem of structured retrieval …...

Natural Language Generation (NLG) systems often use a pipeline architecture for sequen- tial decision making. Recent studies how- ever have shown that treating NLG decisions jointly rather than in isolation can improve the overall performance of systems. We present a joint learning framework based on Hierar- chical Reinforcement Learning (HRL) …...

We introduce the social study of bullying to the NLP community. Bullying, in both physi- cal and cyber worlds (the latter known as cy- berbullying), has been recognized as a seri- ous national health issue among adolescents. However, previous social studies of bully- ing are handicapped by data scarcity, while …...

Generating referring expressions has received considerable attention in Natural Language Generation. In recent years we start seeing deployments of referring expression genera- tors moving away from limited domains with custom-made ontologies. In this work, we ex- plore the feasibility of using large scale noisy ontologies (folksonomies) for open domain referring …...

In this paper, we investigate the use of tempo- ral information for improving extractive sum- marization of historical articles. Our method clusters sentences based on their timestamps and temporal similarity. Each resulting clus- ter is assigned an importance score which can then be used as a weight in traditional sentence …...

It is well known that the output quality of statistical machine translation (SMT) systems increases with more training data. To ob- tain more parallel text for translation mod- eling, researchers have turned to the web to mine parallel sentences, but most previous ap- proaches have avoided the difficult problem of …...

Given a parallel corpus, if two distinct words in language A, a1 and a2, are aligned to the same word b1 in language B, then this might signal that b1 is polysemous, or it might sig- nal a1 and a2 are synonyms. Both assump- tions with successful work have been …...

Noticing that different information sources of- ten provide complementary coverage of word sense and meaning, we propose a simple and yet effective strategy for measuring lexical se- mantics. Our model consists of a committee of vector space models built on a text cor- pus, Web search results and thesauruses, and …...

This paper compares a number of recently pro- posed models for computing context sensitive word similarity. We clarify the connections between these models, simplify their formula- tion and evaluate them in a unified setting. We show that the models are essentially equivalent if syntactic information is ignored, and that the …...

Microblogging networks serve as vehicles for reaching and influencing users. Predicting whether a message will elicit a user response opens the possibility of maximizing the viral- ity, reach and effectiveness of messages and ad campaigns on these networks. We propose a discriminative model for predicting the like- lihood of a …...

Although first names and nicknames in the United States have been well documented, there has been almost no quantitative analysis on the usage and association of these names amongst themselves. In this paper we in- troduce the Intelius Nickname Collection, a quantitative compilation of millions of name- nickname associations based …...

Sentiment analysis of citations in scientific pa- pers and articles is a new and interesting prob- lem which can open up many exciting new ap- plications in bibliographic search and biblio- metrics. Current work on citation sentiment detection focuses on only the citation sen- tence. In this paper, we address …...

Public debate functions as a forum for both expressing and forming opinions, an impor- tant aspect of public life. We present results for automatically classifying posts in online debate as to the position, or STANCE that the speaker takes on an issue, such as Pro or Con. We show that …...

Are word-level affect lexicons useful in de- tecting emotions at sentence level? Some prior research finds no gain over and above what is obtained with ngram features—arguably the most widely used features in text classifica- tion. Here, we experiment with two very dif- ferent emotion lexicons and show that even …...