Lizhen Qu

Timely YAGO: harvesting, querying, and visualizing temporal knowledge from Wikipedia

Recent progress in information extraction has shown how to automatically build large ontologies from high-quality sources like Wikipedia. But knowledge evolves over time; facts have associated validity intervals. Therefore, ontologies should include time as a first-class dimension. In this paper, we introduce Timely YAGO, which extends our previously built knowledge base YAGO with temporal aspects. This prototype system extracts temporal facts from Wikipedia infoboxes, categories, and lists in articles, and integrates these into the Timely YAGO knowledge base. We also support querying temporal facts, by temporal predicates in a SPARQL-style language. Visualization of query results is provided in order to better understand of the dynamic nature of knowledge.

Harvesting facts from textual web sources by constrained label propagation

There have been major advances on automatically constructing large knowledge bases by extracting relational facts from Web and text sources. However, the world is dynamic: periodic events like sports competitions need to be interpreted with their respective timepoints, and facts such as coaching a sports team, holding political or business positions, and even marriages do not hold forever and should be augmented by their respective timespans. This paper addresses the problem of automatically harvesting temporal facts with such extended time-awareness. We employ pattern-based gathering techniques for fact candidates and construct a weighted pattern-candidate graph. Our key contribution is a system called PRAVDA based on a new kind of label propagation algorithm with a judiciously designed loss function, which iteratively processes the graph to label good temporal facts for a given set of target relations. Our experiments with online news and Wikipedia articles demonstrate the accuracy of this method.

Making Deep Neural Networks Robust to Label Noise: A Loss Correction Approach

We present a theoretically grounded approach to train deep neural networks, including recurrent networks, subject to class-dependent label noise. We propose two procedures for loss correction that are agnostic to both application domain and network architecture. They simply amount to at most a matrix inversion and multiplication, provided that we know the probability of each class being corrupted into another. We further show how one can estimate these probabilities, adapting a recent technique for noise estimation to the multi-class setting, and thus providing an end-to-end framework. Extensive experiments on MNIST, IMDB, CIFAR-10, CIFAR-100 and a large scale dataset of clothing images employing a diversity of architectures — stacking dense, convolutional, pooling, dropout, batch normalization, word embedding, LSTM and residual layers — demonstrate the noise robustness of our proposals. Incidentally, we also prove that, when ReLU is the only non-linearity, the loss curvature is immune to class-dependent label noise.

STransE: a novel embedding model of entities and relationships in knowledge bases

Knowledge bases of real-world facts about entities and their relationships are useful resources for a variety of natural language processing tasks. However, because knowledge bases are typically incomplete, it is useful to be able to perform link prediction or knowledge base completion, i.e., predict whether a relationship not in the knowledge base is likely to be true. This paper combines insights from several previous link prediction models into a new embedding model STransE that represents each entity as a low-dimensional vector, and each relation by two matrices and a translation vector. STransE is a simple combination of the SE and TransE models, but it obtains better link prediction performance on two benchmark datasets than previous embedding models. Thus, STransE can serve as a new baseline for the more complex models in the link prediction task.

Neighborhood Mixture Model for Knowledge Base Completion

Knowledge bases are useful resources for many natural language processing tasks, however, they are far from complete. In this paper, we define a novel entity representation as a mixture of its neighborhood in the knowledge base and apply this technique on TransE-a well-known embedding model for knowledge base completion. Experimental results show that the neighborhood information significantly helps to improve the results of the TransE model, leading to better performance than obtained by other state-of-the-art embedding models on three benchmark datasets for triple classification, entity prediction and relation prediction tasks.

Using tag semantic network for keyphrase extraction in blogs

Folksonomies provide a comfortable way to search and browse the blogosphere. As the tags in the blogosphere are sparse, ambiguous and too general, this paper proposes both a supervised and an unsupervised approach that extract tags from posts using a tag semantic network. We evaluate the two methods on a blog dataset and observe an improvement in F1-measure from 0.23 to 0.50 when compared to the baseline system.

Automatic Generation of Grounded Visual Questions

In this paper, we propose the first model to be able to generate visually grounded questions with diverse types for a single image. Visual question generation is an emerging topic which aims to ask questions in natural language based on visual input. To the best of our knowledge, it lacks automatic methods to generate meaningful questions with various types for the same visual input. To circumvent the problem, we propose a model that automatically generates visually grounded questions with varying types. Our model takes as input both images and the captions generated by a dense caption model, samples the most probable question types, and generates the questions in sequel. The experimental results on two real world datasets show that our model outperforms the strongest baseline in terms of both correctness and diversity with a wide margin.

Demographic Inference on Twitter using Recursive Neural Networks.

In social media, demographic inference is a critical task in order to gain a better understanding of a cohort and to facilitate interacting with one’s audience. Most previous work has made independence assumptions over topological, textual and label information on social networks. In this work, we employ recursive neural networks to break down these independence assumptions to obtain inference about demographic characteristics on Twitter. We show that our model performs better than existing models including the state-of-the-art.

Sentiment analysis with limited training data

Sentiments are positive and negative emotions, evaluations and stances. This dissertation focuses on learning based systems for automatic analysis of sentiments and comparisons in natural language text. The proposed approach consists of three contributions: 1. Bag-of-opinions model: For predicting document-level polarity and intensity, we proposed the bag-of-opinions model by modeling each document as a bag of sentiments, which can explore the syntactic structures of sentiment-bearing phrases for improved rating prediction of online reviews. 2. Multi-experts model: Due to the sparsity of manually-labeled training data, we designed the multi-experts model for sentence-level analysis of sentiment polarity and intensity by fully exploiting any available sentiment indicators, such as phrase-level predictors and sentence similarity measures. 3. SENTI-LSSVMRAE model: To understand the sentiments regarding entities, we proposed SENTI-LSSVMRAE model for extracting sentiments and comparisons of entities at both sentence and subsentential level. Different granularity of analysis leads to different model complexity, the finer the more complex. All proposed models aim to minimize the use of hand-labeled data by maximizing the use of the freely available resources. These models explore also different feature representations to capture the compositional semantics inherent in sentiment-bearing expressions. Our experimental results on real-world data showed that all models significantly outperform the state-of-the-art methods on the respective tasks.