Semantically-Informed Syntactic Machine Translation: A Tree-Grafting Approach Machine Learning

We describe a unified and coherent syntactic framework for supporting a semantically-informed syntactic approach to statistical machine translation. Semantically enriched syntactic tags assigned to the target-language training texts improved translation quality. The resulting system significantly outperformed a linguistically naive baseline model (Hiero), and reached the highest scores yet reported on the NIST 2009 Urdu-English translation task. This finding supports the hypothesis (posed by many researchers in the MT community, e.g., in DARPA GALE) that both syntactic and semantic information are critical for improving translation quality---and further demonstrates that large gains can be achieved for low-resource languages with different word order than English.

Dialogue Management for Interactive Question Answering

AAAI Conferences

A major obstacle in building robust, user-friendly Q&A systems is the need to enable a conversation with the user in which clarifications, followup questions and context specification are made possible.

Location-Based Twitter Sentiment Analysis for Predicting the U.S. 2016 Presidential Election

AAAI Conferences

We seek to determine the effectiveness of using location-based social media to predict the outcome of the 2016 presidential election. To this aim, we create a dataset consisting of approximately 3 million tweets ranging from September 22nd to November 8th related to either Donald Trump or Hillary Clinton. Twenty-one states are chosen, with eleven categorized as swing states, five as Clinton favored and five as Trump favored. We incorporate two metrics in polling voter opinion for election outcomes: tweet volume and positive sentiment. Our data is labeled via a convolutional neural network trained on the sentiment140 dataset. To determine whether Twitter is an indicator of election outcome, we compare our results to the election outcome per state and across the nation. We use two approaches for determining state victories: winner-take-all and shared elector count. Our results show tweet sentiment mirrors the close races in the swing states; however, the differences in distribution of positive sentiment and volume between Clinton and Trump are not significant using our approach. Thus, we conclude neither sentiment nor volume is an accurate predictor of election results using our collection of data and labeling process.

A provable SVD-based algorithm for learning topics in dominant admixture corpus Machine Learning

Topic models, such as Latent Dirichlet Allocation (LDA), posit that documents are drawn from admixtures of distributions over words, known as topics. The inference problem of recovering topics from admixtures, is NP-hard. Assuming separability, a strong assumption, [4] gave the first provable algorithm for inference. For LDA model, [6] gave a provable algorithm using tensor-methods. But [4,6] do not learn topic vectors with bounded $l_1$ error (a natural measure for probability vectors). Our aim is to develop a model which makes intuitive and empirically supported assumptions and to design an algorithm with natural, simple components such as SVD, which provably solves the inference problem for the model with bounded $l_1$ error. A topic in LDA and other models is essentially characterized by a group of co-occurring words. Motivated by this, we introduce topic specific Catchwords, group of words which occur with strictly greater frequency in a topic than any other topic individually and are required to have high frequency together rather than individually. A major contribution of the paper is to show that under this more realistic assumption, which is empirically verified on real corpora, a singular value decomposition (SVD) based algorithm with a crucial pre-processing step of thresholding, can provably recover the topics from a collection of documents drawn from Dominant admixtures. Dominant admixtures are convex combination of distributions in which one distribution has a significantly higher contribution than others. Apart from the simplicity of the algorithm, the sample complexity has near optimal dependence on $w_0$, the lowest probability that a topic is dominant, and is better than [4]. Empirical evidence shows that on several real world corpora, both Catchwords and Dominant admixture assumptions hold and the proposed algorithm substantially outperforms the state of the art [5].

Modeling Polarizing Topics: When Do Different Political Communities Respond Differently to the Same News?

AAAI Conferences

Political discourse in the United States is getting increasingly polarized. This polarization frequently causes different communities to react very differently to the same news events. Political blogs as a form of social media provide an unique insight into this phenomenon. We present a multitarget, semisupervised latent variable model, MCR-LDA to model this process by analyzing political blogs posts and their comment sections from different political communities jointly to predict the degree of polarization that news topics cause. Inspecting the model after inference reveals topics and the degree to which it triggers polarization. In this approach, community responses to news topics are observed using sentiment polarity and comment volume which serves as a proxy for the level of interest in the topic. In this context, we also present computational methods to assign sentiment polarity to the comments which serve as targets for latent variable models that predict the polarity based on the topics in the blog content. Our results show that the joint modeling of communities with different political beliefs using MCR-LDA does not sacrifice accuracy in sentiment polarity prediction when compared to approaches that are tailored to specific communities and additionally provides a view of the polarization in responses from the different communities.