Katja Filippova

Dependency tree based sentence compression

We present a novel unsupervised method for sentence compression which relies on a dependency tree representation and shortens sentences by removing subtrees. An automatic evaluation shows that our method obtains result comparable or superior to the state of the art. We demonstrate that the choice of the parser affects the performance of the system. We also apply the method to German and report the results of an evaluation with humans.

Sentence Compression by Deletion with LSTMs

We present an LSTM approach to deletion-based sentence compression where the task is to translate a sentence into a sequence of zeros and ones, corresponding to token deletion decisions. We demonstrate that even the most basic version of the system, which is given no syntactic information (no PoS or NE tags, or dependencies) or desired compression length, performs surprisingly well: around 30% of the compressions from a large test set could be regenerated. We compare the LSTM system with a competitive baseline which is trained on the same amount of data but is additionally provided with all kinds of linguistic features. In an experiment with human raters the LSTMbased model outperforms the baseline achieving 4.5 in readability and 3.8 in informativeness.

Sentence Fusion via Dependency Graph Compression

We present a novel unsupervised sentence fusion method which we apply to a corpus of biographies in German. Given a group of related sentences, we align their dependency trees and build a dependency graph. Using integer linear programming we compress this graph to a new tree, which we then linearize. We use GermaNet and Wikipedia for checking semantic compatibility of co-arguments. In an evaluation with human judges our method outperforms the fusion approach of Barzilay & McKeown (2005) with respect to readability.

Extending the Entity-grid Coherence Model to Semantically Related Entities

This paper reports on work in progress on extending the entity-based approach on measuring coherence (Barzilay & Lapata, 2005; Lapata & Barzilay, 2005) from coreference to semantic relatedness. We use a corpus of manually annotated German newspaper text (TuBa-D/Z) and aim at improving the performance by grouping related entities with the WikiRelate! API (Strube & Ponzetto, 2006).

Multi-lingual opinion mining on YouTube

We designed the first model for effectively carrying out opinion mining on YouTube comments.We propose kernel methods applied to a robust shallow syntactic structure, which improves accuracy for both languages.Our approach greatly outperforms other basic models on cross-domain settings.We created a YouTube corpus (in Italian and English) and made it available for the research community.Comments must be classified in subcategories to make opinion mining effective on YouTube. In order to successfully apply opinion mining (OM) to the large amounts of user-generated content produced every day, we need robust models that can handle the noisy input well yet can easily be adapted to a new domain or language. We here focus on opinion mining for YouTube by (i) modeling classifiers that predict the type of a comment and its polarity, while distinguishing whether the polarity is directed towards the product or video; (ii) proposing a robust shallow syntactic structure (STRUCT) that adapts well when tested across domains; and (iii) evaluating the effectiveness on the proposed structure on two languages, English and Italian. We rely on tree kernels to automatically extract and learn features with better generalization power than traditionally used bag-of-word models. Our extensive empirical evaluation shows that (i) STRUCT outperforms the bag-of-words model both within the same domain (up to 2.6% and 3% of absolute improvement for Italian and English, respectively); (ii) it is particularly useful when tested across domains (up to more than 4% absolute improvement for both languages), especially when little training data is available (up to 10% absolute improvement) and (iii) the proposed structure is also effective in a lower-resource language scenario, where only less accurate linguistic processing tools are available.

Modelling Events through Memory-based, Open-IE Patterns for Abstractive Summarization

Abstractive text summarization of news requires a way of representing events, such as a collection of pattern clusters in which every cluster represents an event (e.g., marriage) and every pattern in the clus- ter is a way of expressing the event (e.g., X married Y, X and Y tied the knot). We compare three ways of extracting event patterns: heuristics-based, compression- based and memory-based. While the for- mer has been used previously in multi- document abstraction, the latter two have never been used for this task. Compared with the first two techniques, the memory- based method allows for generating sig- nificantly more grammatical and informa- tive sentences, at the cost of searching a vast space of hundreds of millions of parse trees of known grammatical utterances. To this end, we introduce a data structure and a search method that make it possible to efficiently extrapolate from every sentence the parse sub-trees that match against any of the stored utterances.

The German Vorfeld and Local Coherence

We present a method for improving local coherence in German with a positive effect on automatically as well as human-generated texts. We demonstrate that local coherence crucially depends on which constituent occupies the initial position in a sentence. To support our hypothesis, we provide statistical evidence based on a corpus investigation and on results of an experiment with human judges. Additionally, we implement our findings in a generation module for determining the Vorfeld constituent automatically.

Company-Oriented Extractive Summarization of Financial News

The paper presents a multi-document summarization system which builds company-specific summaries from a collection of financial news such that the extracted sentences contain novel and relevant information about the corresponding organization. The users familiarity with the companys profile is assumed. The goal of such summaries is to provide information useful for the short-term trading of the corresponding company, i.e., to facilitate the inference from news to stock price movement in the next day. We introduce a novel query (i.e., company name) expansion method and a simple unsupervized algorithm for sentence ranking. The system shows promising results in comparison with a competitive baseline.

Opinion Mining on YouTube

This paper defines a systematic approach to Opinion Mining (OM) on YouTube comments by (i) modeling classifiers for predicting the opinion polarity and the type of comment and (ii) proposing robust shallow syntactic structures for improving model adaptability. We rely on the tree kernel technology to automatically extract and learn features with better generalization power than bag-of-words. An extensive empirical evaluation on our manually annotated YouTube comments corpus shows a high classification accuracy and highlights the benefits of structural models in a cross-domain setting.

Idest: Learning a Distributed Representation for Event Patterns

This paper describes IDEST, a new method for learning paraphrases of event patterns. It is based on a new neural network architecture that only relies on the weak supervision signal that comes from the news published on the same day and mention the same real-world entities. It can generalize across extractions from different dates to produce a robust paraphrase model for event patterns that can also capture meaningful representations for rare patterns. We compare it with two state-of-the-art systems and show that it can attain comparable quality when trained on a small dataset. Its generalization capabilities also allow it to leverage much more data, leading to substantial quality improvements.