Under the heading Concretely Annotated, we are processing a variety of standard corpora with multiple popular NLP tool-chains, collected together under a single data schema we have created that we refer to as CONCRETE. We envision a multimodal workflow, where, e.g., knowledge can be extracted from both text and audio. We developed CONCRETE to record and share annotations on structured human language data – both text and speech.

A supervised method relies on simple, lightweight features in order to distinguish Wikipedia articles that are classes (Shield volcano) from other articles (Kilauea). The features are lexical or semantic in nature. Experimental results in multiple languages over multiple evaluation sets demonstrate the superiority of the proposed method over previous work.

Hand-engineered feature sets are a well understood method for creating robust NLP models, but they require a lot of expertise and effort to create. In this work we describe how to automatically generate rich feature sets from simple units called featlets, requiring less engineering. Using information gain to guide the generation process, we train models which rival the state of the art on two standard Semantic Role Labeling datasets with almost no task or linguistic insight.

Computer Assisted Discovery Extraction and Translation (CADET) is a workbench for helping knowledge workers find, label, and translate documents of interest. It combines a multitude of analytics together with a flexible environment for customizing the workflow for different users. This open-source framework allows for easy development of new research prototypes using a micro-service architecture based atop Docker and Apache Thrift.

Existing Knowledge Base Population methods extract relations from a closed relational schema with limited coverage leading to sparse KBs. We propose Pocket Knowledge Base Population (PKBP), the task of dynamically constructing a KB of entities related to a query and finding the best characterization of relationships between entities. We describe novel Open Information Extraction methods which leverage the PKB to find informative trigger words. We evaluate using existing KBP shared-task data as well anew annotations collected for this work. Our methods produce high quality KB from just text with many more entities and relationships than existing KBP systems.

Global features have proven effective in a wide range of structured prediction problems but come with high inference costs.Imitation learning is a common method for training models when exact inference isn't feasible.We study imitation learning for Semantic Role Labeling (SRL) and analyze the effectiveness of the Violation Fixing Perceptron (VFP) (Huang et al., 2012) and Locally Optimal Learning to Search (LOLS) (Chang et al., 2015) frameworks with respect to SRL global features.We describe problems in applying each framework to SRL and evaluate the effectiveness of some solutions.We also show that action ordering, including easy first inference, has a large impact on the quality of greedy global models.

We present a joint model for predicate argument alignment. We leverage multiple sources of semantic information, including temporal ordering constraints between events. These are combined in a max-margin framework to find a globally consistent view of entities and events across multiple documents, which leads to improvements over a very strong local baseline.

Most work on building knowledge bases has focused on collecting entities and facts from as large a collection of documents as possible. We argue for and describe a new paradigm where the focus is on a high-recall extraction over a small collection of documents under the supervision of a human expert, that we call Interactive Knowledge Base Population (IKBP).

Nanyun Peng, Francis Ferraro, Mo Yu, Nicholas Andrews, Jay DeYoung, Max Thomas, Matthew R. Gormley, Travis Wolfe, Craig Harman, Benjamin Van Durme, Mark Dredze. Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations. 2015.

We increase the lexical coverage of FrameNet through automatic paraphrasing. We use crowdsourcing to manually filter out bad paraphrases in order to ensure a high-precision resource. Our expanded FrameNet contains an additional 22K lexical units, a 3-fold increase over the current FrameNet, and achieves 40% better coverage when evaluated in a practical setting on New York Times data.

In either setting, it is common for a research group to generate bulk annotations over a preferred corpus internally, using their own tools, programming languages and formats, but then reporting on this as merely an engineering pre-processing step not worth describing in significant detail. Worse, these annotated collections are often not available to the rest of the community, making it difficult to perform apples-to-apples comparison of the “real research”.

Richly annotated documents are a part of many knowledge extraction efforts. Such efforts may include knowledge base population (KBP), a task where it is important to recognize certain relationships between entities in text in order to add those facts to an entity-centric database. Document annotations might also serve as grist for training various distributional models of meaning.In either setting, it is common for a research group to generate bulk annotations over a preferred corpus internally, using their own tools, programming languages and formats, but then reporting on this as merely an engineering pre-processing step not worth describing in significant detail. Worse, these annotated collections are often not available to the rest of the community, making it difficult to perform apples-to-apples comparison of the “real research”.

Effectively exploring and analyzing large text corpora requires visualizations that provide a high level summary. Past work has relied on faceted browsing of document metadata or on natural language processing of document text. In this paper, we present a new web-based tool that integrates topics learned from an unsupervised topic model in a faceted browsing experience. The user can manage topics, filter documents by topic and summarize views with metadata and topic graphs. We report a user study of the usefulness of topics in our tool.

We introduce PARMA, a system for crossdocument, semantic predicate and argument alignment. Our system combines a number of linguistic resources familiar to researchers in areas such as recognizing textual entailment and question answering, integrating them into a simple discriminative model. PARMA achieves state of the art results on an existing and a new dataset. We suggest that previous efforts have focussed on data that is biased and too easy, and we provide a more difficult dataset based on translation data with a low baseline which we beat by 17% F1.

We describe a system for recommending news articles, called NewsPer, which learns news-reading preferences of its users and suggests recently published articles that may be of interest to specific readers based on their interest profiles. The underlying algorithm is based on representing articles by bags of words and named entities, and applying support vector machines to this representation. We present this algorithm and give initial empirical results. We also discuss broader issues in the news personalization and the challenges of performance evaluation based on historical data.