arXiv subjects

Dragomir Radev

Publications and source records attributed to Dragomir Radev.

98 records · Page 6Linked to original sources

arXiv2017

Graph-based Neural Multi-Document Summarization

We propose a neural multi-document summarization (MDS) system that incorporates sentence relation graphs. We employ a Graph Convolutional Network (GCN) on the relation graphs, with sentence embeddings obtained from Recurrent Neural Networks as input node features. Through multiple layer-wise propagation, the GCN generates high-level hidden sentence features for salience estimation. We then use a greedy heuristic to extract salient sentences while avoiding redundancy. In our experiments on DUC 2004, we consider three types of sentence relation graphs and demonstrate the advantage of combining sentence relations in graphs with the representation power of deep neural networks. Our model improves upon traditional graph-based extractive approaches and the vanilla GRU sequence model with no graph, and it achieves competitive results against other state-of-the-art multi-document summarization systems.

cs.CL↗

arXiv2016

Cruciform: Solving Crosswords with Natural Language Processing

Crossword puzzles are popular word games that require not only a large vocabulary, but also a broad knowledge of topics. Answering each clue is a natural language task on its own as many clues contain nuances, puns, or counter-intuitive word definitions. Additionally, it can be extremely difficult to ascertain definitive answers without the constraints of the crossword grid itself. This task is challenging for both humans and computers. We describe here a new crossword solving system, Cruciform. We employ a group of natural language components, each of which returns a list of candidate words with scores when given a clue. These lists are used in conjunction with the fill intersections in the puzzle grid to formulate a constraint satisfaction problem, in a manner similar to the one used in the Dr. Fill system. We describe the results of several of our experiments with the system.

cs.CL↗

arXiv2016

Dependency Sensitive Convolutional Neural Networks for Modeling Sentences and Documents

The goal of sentence and document modeling is to accurately represent the meaning of sentences and documents for various Natural Language Processing tasks. In this work, we present Dependency Sensitive Convolutional Neural Networks (DSCNN) as a general-purpose classification system for both sentences and documents. DSCNN hierarchically builds textual representations by processing pretrained word embeddings via Long Short-Term Memory networks and subsequently extracting features with convolution operators. Compared with existing recursive neural models with tree structures, DSCNN does not rely on parsers and expensive phrase labeling, and thus is not restricted to sentence-level tasks. Moreover, unlike other CNN-based models that analyze sentences locally by sliding windows, our system captures both the dependency information within each sentence and relationships across sentences in the same document. Experiment results demonstrate that our approach is achieving state-of-the-art performance on several tasks, including sentiment analysis, question type classification, and subjectivity classification.

cs.CL↗

arXiv2016

Classifying Syntactic Regularities for Hundreds of Languages

This paper presents a comparison of classification methods for linguistic typology for the purpose of expanding an extensive, but sparse language resource: the World Atlas of Language Structures (WALS) (Dryer and Haspelmath, 2013). We experimented with a variety of regression and nearest-neighbor methods for use in classification over a set of 325 languages and six syntactic rules drawn from WALS. To classify each rule, we consider the typological features of the other five rules; linguistic features extracted from a word-aligned Bible in each language; and genealogical features (genus and family) of each language. In general, we find that propagating the majority label among all languages of the same genus achieves the best accuracy in label pre- diction. Following this, a logistic regression model that combines typological and linguistic features offers the next best performance. Interestingly, this model actually outperforms the majority labels among all languages of the same family.

cs.CL↗

arXiv2015

Humor in Collective Discourse: Unsupervised Funniness Detection in the New Yorker Cartoon Caption Contest

The New Yorker publishes a weekly captionless cartoon. More than 5,000 readers submit captions for it. The editors select three of them and ask the readers to pick the funniest one. We describe an experiment that compares a dozen automatic methods for selecting the funniest caption. We show that negative sentiment, human-centeredness, and lexical centrality most strongly match the funniest captions, followed by positive sentiment. These results are useful for understanding humor and also in the design of more engaging conversational agents in text and multimodal (vision+text) systems. As part of this work, a large set of cartoons and captions is being made available to the community.

cs.CL↗

arXiv2011

Improving Image Search based on User Created Communities

Tag-based retrieval of multimedia content is a difficult problem, not only because of the shorter length of tags associated with images and videos, but also due to mismatch in the terminologies used by searcher and content creator. To alleviate this problem, we propose a simple concept-driven probabilistic model for improving text-based rich-media search. While our approach is similar to existing topic-based retrieval and cluster-based language modeling work, there are two important differences: (1) our proposed model considers not only the query-generation likelihood from cluster, but explicitly accounts for the overall "popularity" of the cluster or underlying concept, and (2) we explore the possibility of inferring the likely concept relevant to a rich-media content through the user-created communities that the content belongs to. We implement two methods of concept extraction: a traditional cluster based approach, and the proposed community based approach. We evaluate these two techniques for how effectively they capture the intended meaning of a term from the content creator and searcher, and their overall value in improving image search. Our results show that concept-driven search, though simple, clearly outperforms plain search. Among the two techniques for concept-driven search, community-based approach is more successful, as the concepts generated from user communities are found to be more intuitive and appealing.

cs.IR↗

arXiv2011

The effect of linguistic constraints on the large scale organization of language

This paper studies the effect of linguistic constraints on the large scale organization of language. It describes the properties of linguistic networks built using texts of written language with the words randomized. These properties are compared to those obtained for a network built over the text in natural order. It is observed that the "random" networks too exhibit small-world and scale-free characteristics. They also show a high degree of clustering. This is indeed a surprising result - one that has not been addressed adequately in the literature. We hypothesize that many of the network statistics reported here studied are in fact functions of the distribution of the underlying data from which the network is built and may not be indicative of the nature of the concerned network.

cs.CL↗

arXiv2007

CLAIRLIB Documentation v1.03

The Clair library is intended to simplify a number of generic tasks in Natural Language Processing (NLP), Information Retrieval (IR), and Network Analysis. Its architecture also allows for external software to be plugged in with very little effort. Functionality native to Clairlib includes Tokenization, Summarization, LexRank, Biased LexRank, Document Clustering, Document Indexing, PageRank, Biased PageRank, Web Graph Analysis, Network Generation, Power Law Distribution Analysis, Network Analysis (clustering coefficient, degree distribution plotting, average shortest path, diameter, triangles, shortest path matrices, connected components), Cosine Similarity, Random Walks on Graphs, Statistics (distributions, tests), Tf, Idf, Community Finding.

cs.IR↗