Question Answering (QA), as a research field, has primarily focused on either knowledge bases (KBs) or free text as a source of knowledge. These two sources have historically shaped the kinds of questions that are asked over these sources, and the methods developed to answer them. In this work, we look towards a practical use-case of QA over user-instructed knowledge that uniquely combines elements of both structured QA over knowledge bases, and unstructured QA over narrative, introducing the task of multi-relational QA over personal narrative. As a first step towards this goal, we make three key contributions: (i) we generate and release TextWorldsQA, a set of five diverse datasets, where each dataset contains dynamic narrative that describes entities and relations in a simulated world, paired with variably compositional questions over that knowledge, (ii) we perform a thorough evaluation and analysis of several state-of-the-art QA models and their variants at this task, and (iii) we release a lightweight Python-based framework we call TextWorlds for easily generating arbitrary additional worlds and narrative, with the goal of allowing the community to create and share a growing collection of diverse worlds as a test-bed for this task.

As humans, we often rely on language to learn language.For example, when corrected in a conversation, we may learn from that correction, over time improving our language fluency.Inspired by this observation, we propose a learning algorithm for training semantic parsers from supervision (feedback) expressed in natural language.Our algorithm learns a semantic parser from users' corrections such as "no, what I really meant was before his job, not after", by also simultaneously learning to parse this natural language feedback in order to leverage it as a form of supervision.Unlike supervision with gold-standard logical forms, our method does not require the user to be familiar with the underlying logical formalism, and unlike supervision from denotation, it does not require the user to know the correct answer to their query.This makes our learning algorithm naturally scalable in settings where existing conversational logs are available and can be leveraged as training data.We construct a novel dataset of natural language feedback in a conversational setting, and show that our method is effective at learning a semantic parser from such natural language supervision.

This paper focuses on how to take advantage of external knowledge bases (KBs) to improve recurrent neural networks for machine reading.Traditional methods that exploit knowledge from KBs encode knowledge as discrete indicator features.Not only do these features generalize poorly, but they require task-specific feature engineering to achieve good performance.We propose KBLSTM, a novel neural model that leverages continuous representations of KBs to enhance the learning of recurrent neural networks for machine reading.To effectively integrate background knowledge with information from the currently processed text, our model employs an attention mechanism with a sentinel to adaptively decide whether to attend to background knowledge and which information from KBs is useful.Experimental results show that our model achieves accuracies that surpass the previous state-of-the-art results for both entity extraction and event extraction on the widely used ACE2005 dataset.

We introduce a new method for framesemantic parsing that significantly improves the prior state of the art.Our model leverages the advantages of a deep bidirectional LSTM network which predicts semantic role labels word by word and a relational network which predicts semantic roles for individual text expressions in relation to a predicate.The two networks are integrated into a single model via knowledge distillation, and a unified graphical model is employed to jointly decode frames and semantic roles during inference.Experiments on the standard FrameNet data show that our model significantly outperforms existing neural and non-neural approaches, achieving a 5.7 F1 gain over the current state of the art, for full frame structure extraction.

Structured Prediction Energy Networks (SPENs) are a simple, yet expressive family of structured prediction models (Belanger and McCallum, 2016). An energy function over candidate structured outputs is given by a deep network, and predictions are formed by gradient-based optimization. This paper presents end-to-end learning for SPENs, where the energy function is discriminatively trained by back-propagating through gradient-based prediction. In our experience, the approach is substantially more accurate than the structured SVM method of Belanger and McCallum (2016), as it allows us to use more sophisticated non-convex energies. We provide a collection of techniques for improving the speed, accuracy, and memory requirements of end-to-end SPENs, and demonstrate the power of our method on 7-Scenes image denoising and CoNLL-2005 semantic role labeling tasks. In both, inexact minimization of non-convex SPEN energies is superior to baseline methods that use simplistic energy functions that can be minimized exactly.

Identifying individual's political ideology from their speeches and written texts is important for analyzing political opinions and user behavior on social media. Traditional opinion mining methods rely on bag-of-words representations to classify texts into different ideology categories. Such methods are too coarse for understanding political ideologies. The key to identify different ideologies is to recognize different opinions expressed toward a specific topic. To model this insight, we classify ideologies based on the distribution of opinions expressed towards real-world entities or topics. Specifically, we propose a novel approach to political ideology detection that makes predictions based on an opinion-aware knowledge graph. We show how to construct such graph by integrating the opinions and targeted entities extracted from text into an existing structured knowledge base, and show how to perform ideology inference by information propagation on the graph. Experimental results demonstrate that our method achieves high accuracy in detecting ideologies compared to baselines including LR, SVM and RNN.

Events and entities are closely related; entities are often actors or participants in events and events without entities are uncommon.The interpretation of events and entities is highly contextually dependent.Existing work in information extraction typically models events separately from entities, and performs inference at the sentence level, ignoring the rest of the document.In this paper, we propose a novel approach that models the dependencies among variables of events, entities, and their relations, and performs joint inference of these variables across a document.The goal is to enable access to document-level contextual information and facilitate contextaware predictions.We demonstrate that our approach substantially outperforms the stateof-the-art methods for event extraction as well as a strong baseline for entity extraction.

We present a novel hierarchical distance-dependent Bayesian model for event coreference resolution. While existing generative models for event coreference resolution are completely unsupervised, our model allows for the incorporation of pairwise distances between event mentions — information that is widely used in supervised coreference models to guide the generative clustering processing for better event clustering both within and across documents. We model the distances between event mentions using a feature-rich learnable distance function and encode them as Bayesian priors for nonparametric clustering. Experiments on the ECB+ corpus show that our model outperforms state-of-the-art methods for both within- and cross-document event coreference resolution.

This paper proposes a novel context-aware method for analyzing sentiment at the level of individual sentences. Most existing machine learning approaches suffer from limitations in the modeling of complex linguistic structures across sentences and often fail to capture nonlocal contextual cues that are important for sentiment interpretation. In contrast, our approach allows structured modeling of sentiment while taking into account both local and global contextual information. Specifically, we encode intuitive lexical and discourse knowledge as expressive constraints and integrate them into the learning of conditional random field models via posterior regularization. The context-aware constraints provide additional power to the CRF model and can guide semi-supervised learning when labeled data is limited. Experiments on standard product review datasets show that our method outperforms the state-of-theart methods in both the supervised and semi-supervised settings.

While relation extraction has traditionally been viewed as a task relying solely on textual data, recent work has shown that by taking as input existing facts in the form of entity-relation triples from both knowl-edge bases and textual data, the perfor-mance of relation extraction can be im-proved significantly. Following this new paradigm, we propose a tensor decompo-sition approach for knowledge base em-bedding that is highly scalable, and is es-pecially suitable for relation extraction. By leveraging relational domain knowl-edge about entity type information, our learning algorithm is significantly faster than previous approaches and is better able to discover new relations missing from the database. In addition, when ap-plied to a relation extraction task, our ap-proach alone is comparable to several ex-isting systems, and improves the weighted mean average precision of a state-of-the-art method by 10 points when used as a subcomponent. 1

In this paper, we study the problems of opinion expression extraction and expression-level polarity and intensity classification. Traditional fine-grained opinion analysis systems address these problems in isolation and thus cannot capture interactions among the textual spans of opinion expressions and their opinion-related properties. We present two types of joint approaches that can account for such interactions during 1) both learning and inference or 2) only during inference. Extensive experiments on a standard dataset demonstrate that our approaches provide substantial improvements over previously published results. By analyzing the results, we gain some insight into the advantages of different joint models.

Labeling text data is quite time-consuming but essential for automatic text classification. Especially, manually creating multiple labels for each document may become impractical when a very large amount of data is needed for training multi-label text classifiers. To minimize the human-labeling efforts, we propose a novel multi-label active learning approach which can reduce the required labeled data without sacrificing the classification accuracy. Traditional active learning algorithms can only handle single-label problems, that is, each data is restricted to have one label. Our approach takes into account the multi-label information, and select the unlabeled data which can lead to the largest reduction of the expected model loss. Specifically, the model loss is approximated by the size of version space, and the reduction rate of the size of version space is optimized with Support Vector Machines (SVM). An effective label prediction method is designed to predict possible labels for each unlabeled data point, and the expected loss for multi-label data is approximated by summing up losses on all labels according to the most confident result of label prediction. Experiments on several real-world data sets (all are publicly available) demonstrate that our approach can obtain promising classification result with much fewer labeled data than state-of-the-art methods.

Mobile communication data analysis has been often used as a background application to motivate many data mining problems. However, very few data mining researchers have a chance to see a working data mining system on real mobile communication data. In this demo, we showcase our new system MobileMiner on a real mobile communication data set, which presents a case study of business solutions using state-of-the-art data mining techniques. MobileMiner adaptively profiles users' behavior from their calling and moving record streams. Customer segmentation and social community analysis can be conducted based on user profiles. We show how data mining techniques can help in mobile communication data analysis. Moreover, we also show some interesting observations which still cannot be mined by the current techniques, and thus may motivate new research and development.