Most of the recent efficient algorithms for dependency parsing work by factoring the dependency trees. In most of these approaches, the parser loses much of the contextual information during the process of factorization. There have been approaches to build higher order dependency parsers - second order, [Carreras2007] and third order [Koo and Collins2010]. In the thesis, the approach by Koo and Collins should be further exploited in one or more ways. Possible directions of further exploitation include but are not limited to: investigating possibilities of extension of the approach to non-projective parsing; integrating labeled parsing; joining word-senses during the parsing phase [Eisner2000]

Most of the recent efficient algorithms for dependency parsing work by factoring the dependency trees. In most of these approaches, the parser loses much of the contextual information during the process of factorization. There have been approaches to build higher order dependency parsers - second order, [Carreras2007] and third order [Koo and Collins2010]. In the thesis, the approach by Koo and Collins should be further exploited in one or more ways. Possible directions of further exploitation include but are not limited to: investigating possibilities of extension of the approach to non-projective parsing; integrating labeled parsing; joining word-senses during the parsing phase [Eisner2000].

No description available.

The most time-consuming component of designing and launching hardware products to market is the verification of Integrated Circuits (IC). An effective way of verifying a design can be achieved by adding assertions to the design. Automatic translation of hardware specifications from natural language to assertions in a formal representation has the potential to improve the verification productivity of ICs. However, natural language specifications have the characteristics of being imprecise, incomplete, and ambiguous. An automation framework can benefit verification engineers only if it is designed with the right balance between the ease of expression and precision of meaning allowed for in the input natural language specifications. This requirement introduces two major challenges for designing an effective translation framework. The first challenge is to allow the processing of expressive specifications with flexible word order variations and sentence structures. The second challenge is to assist users in writing unambiguous and complete specifications in the English language that can be accurately translated. In this dissertation, we address the first challenge by modeling semantic parsing of the input sentence as a game of BINGO that can capture the combinatorial nature of natural language semantics. BINGO parsing considers the context of each word in the input sentence to ensure high precision in the creation of semantic frames. We address the second challenge by designing a suggestion and feedback framework to assist users in writing clear and coherent specifications. Our feedback generates different ways of writing acceptable sentences when the input sentence is not understood. We evaluated our BINGO model on 316 hardware design specifications taken from the documents of AMBA, memory controller, and UART architectures. The results showed that highly expressive specifications could be handled in our BINGO model. It also demonstrated the ease of creating rules to generate the same semantic frame for specifications with the same meaning but different word order. We evaluated the suggestion and rewriting framework on 132 erroneous specifications taken from AMBA and memory controller architectures documents. Our system generated suggestions for all the specs. On manual inspection, we found that 87% of these suggestions were semantically closer to the intent of the input specification. Moreover, automatic contextual analysis of the rewritten form of the input specification allowed the translation of the input specification with different words and different order of words that were not defined in our grammar. / Doctor of Philosophy / The most time-consuming component of designing and launching hardware products to market is the verification of hardware circuits. An effective way of verifying a design is to add programming codes called assertions in the design. The creation of assertions can be time-consuming and error-prone due to the technical details needed to write assertions. Automatically translating assertion specifications written in English to program code can reduce design time and errors since the English language hides away the technical details required for writing assertions. However, sentences written in English language can have multiple and incomplete interpretations. It becomes difficult for machines to understand assertions written in the English language. In this work, we automatically generate assertions from assertion descriptions written in English. We propose techniques to write rules that can accurately translate English specifications to assertions. Our rules allow a user to write specifications with flexible use of word order and word interpretations. We have tested the understanding framework on English specifications taken from four different types of hardware design architectures. Since we cannot create rules to understand all possible ways of writing a specification, we have proposed a suggestion framework that can inform the user about the words and word structures acceptable to our translation framework. The suggestion framework was tested on specifications of AMBA and memory controller architectures.

With the prevalence of large data stored in the cloud, including unstructured information in the form of text, there is now an increased emphasis on text mining. A broad range of techniques are now used for text mining, including algorithms adapted from machine learning, NLP, computational linguistics, and data mining. Applications are also multi-fold, including classification, clustering, segmentation, relationship discovery, and practically any task that discovers latent information from written natural language. Classical mining algorithms have traditionally focused on shallow representations such as bag-of-words and similar feature-based models. With the advent of modern high performance computing, deep sentence level linguistic analysis of large scale text corpora has become practical. In this dissertation, we evaluate the utility of dependency parses as textual features for different text mining applications. Dependency parsing is one form of syntactic parsing, based on the dependency grammar implicit in sentences. While dependency parsing has traditionally been used for text understanding, we investigate here its application to supply features for text mining applications. We specifically focus on three methods to construct textual features from dependency parses. First, we consider a dependency parse as a general feature akin to a traditional bag-of-words model. Second, we consider the dependency parse as the basis to build a feature graph representation. Finally, we use dependency parses in a supervised collocation mining method for feature selection. To investigate these three methods, several applications are studied, including: (i) movie spoiler detection, (ii) text segmentation, (iii) query expansion, and (iv) recommender systems. / Ph. D.

The use of Long Short Term Memory (LSTM) networks continues to yield better results in natural language processing tasks. One area which recently has seen significant improvements is semantic dependency parsing, where the current state-of-the-art model uses a multilayer LSTM combined with an attention-based scoring function to predict the dependencies. In this thesis the state of the art model is first replicated and then extended to include features based on syntactical trees, which was found to be useful in a similar model. In addition, the effect of part-of-speech tags is studied. The replicated model achieves a labeled F1 score of 93.6 on the in-domain data and 89.2 on the out-of-domain data on the DM dataset, which shows that the model is indeed replicable. Using multiple features extracted from syntactic gold standard trees of the DELPH-IN Derivation Tree (DT) type increased the labeled scores to 97.1 and 94.1 respectively, while the use of predicted trees of the Stanford Basic (SB) type did not improve the results at all. The usefulness of part-of-speech tags was found to be diminished in the presence of other features.

This thesis deals with automatic syntactic analysis of natural languagetext, also known as parsing. The parsing approach is data-driven, whichmeans that parsers are constructed by means of machine learning, lookingat training data in the form of annotated natural language sentences. The syntactic framework used in the thesis is dependency-based. Robustness is one of the characteristics of the data-driven approaches investigated here.The overall aim of this thesis is to maintain robustness while increasing accuracy.The content of the thesis falls naturally into two tracks, a transformation track and a combination track. The  rst type of transformation investigatedis called pseudo-projective, because it enables strictly projective dependency parsers to recover non-projective dependency relations. Informally,a non-projective dependency tree contains crossing binary directed relations, when drawn above the sentence. Experimental results show that pseudo-projective transformations can improve accuracy significantly for a range of languages. The second type of transformation aims to facilitate the processing of specific linguistic constructions such as coordination and verb groups. Experimental results again show a positive effect on parsing accuracy for several languages, often greater than for the pseudo-projective transformations. However, the improvement of the transformations dependson the internal structure of the base parser, which is not the case for thepseudo-projective transformations. The combination track compares various approaches for combining data driven dependency parsers, again as a means of improving accuracy. As different parsers have different strengths and weaknesses, making parsers collaborate in order to  nd one single syntactic analysis may result in higher accuracy than any of the syntactic analyzers can produce by itself. The experimental results show that accuracy improves across languages, giventhat appropriate parsers are combined. The thesis ends with an attempt to combine the two tracks, showing that combining parsers with different tree transformations also increases accuracy. Moreover, this experiment indicates that high diversity among a small set of parsers is much more important than a large number of parsers with low diversity.

No description available.

<p>This licentiate thesis presents a software architecture for inductive labeled dependency parsing of unrestricted natural language text, which achieves a strict modularization of parsing algorithm, feature model and learning method such that these parameters can be varied independently. The architecture is based on the theoretical framework of inductive dependency parsing by Nivre \citeyear{nivre06c} and has been realized in MaltParser, a system that supports several parsing algorithms and learning methods, for which complex feature models can be defined in a special description language. Special attention is given in this thesis to learning methods based on support vector machines (SVM).</p><p>The implementation is validated in three sets of experiments using data from three languages (Chinese, English and Swedish). First, we check if the implementation realizes the underlying architecture. The experiments show that the MaltParser system outperforms the baseline and satisfies the basic constraints of well-formedness. Furthermore, the experiments show that it is possible to vary parsing algorithm, feature model and learning method independently. Secondly, we focus on the special properties of the SVM interface. It is possible to reduce the learning and parsing time without sacrificing accuracy by dividing the training data into smaller sets, according to the part-of-speech of the next token in the current parser configuration. Thirdly, the last set of experiments present a broad empirical study that compares SVM to memory-based learning (MBL) with five different feature models, where all combinations have gone through parameter optimization for both learning methods. The study shows that SVM outperforms MBL for more complex and lexicalized feature models with respect to parsing accuracy. There are also indications that SVM, with a splitting strategy, can achieve faster parsing than MBL. The parsing accuracy achieved is the highest reported for the Swedish data set and very close to the state of the art for Chinese and English.</p> / <p>Denna licentiatavhandling presenterar en mjukvaruarkitektur för</p><p>datadriven dependensparsning, dvs. för att automatiskt skapa en</p><p>syntaktisk analys i form av dependensgrafer för meningar i texter</p><p>på naturligt språk. Arkitekturen bygger på idén att man ska kunna variera parsningsalgoritm, särdragsmodell och inlärningsmetod oberoende av varandra. Till grund för denna arkitektur har vi använt det teoretiska ramverket för induktiv dependensparsning presenterat av Nivre \citeyear{nivre06c}. Arkitekturen har realiserats i programvaran MaltParser, där det är möjligt att definiera komplexa särdragsmodeller i ett speciellt beskrivningsspråk. I denna avhandling kommer vi att lägga extra tyngd vid att beskriva hur vi har integrerat inlärningsmetoden supportvektor-maskiner (SVM).</p><p>MaltParser valideras med tre experimentserier, där data från tre språk används (kinesiska, engelska och svenska). I den första experimentserien kontrolleras om implementationen realiserar den underliggande arkitekturen. Experimenten visar att MaltParser utklassar en trivial metod för dependensparsning (\emph{eng}. baseline) och de grundläggande kraven på välformade dependensgrafer uppfylls. Dessutom visar experimenten att det är möjligt att variera parsningsalgoritm, särdragsmodell och inlärningsmetod oberoende av varandra. Den andra experimentserien fokuserar på de speciella egenskaperna för SVM-gränssnittet. Experimenten visar att det är möjligt att reducera inlärnings- och parsningstiden utan att förlora i parsningskorrekthet genom att dela upp träningsdata enligt ordklasstaggen för nästa ord i nuvarande parsningskonfiguration. Den tredje och sista experimentserien presenterar en empirisk undersökning som jämför SVM med minnesbaserad inlärning (MBL). Studien använder sig av fem särdragsmodeller, där alla kombinationer av språk, inlärningsmetod och särdragsmodell</p><p>har genomgått omfattande parameteroptimering. Experimenten visar att SVM överträffar MBL för mer komplexa och lexikaliserade särdragsmodeller med avseende på parsningskorrekthet. Det finns även vissa indikationer på att SVM, med en uppdelningsstrategi, kan parsa en text snabbare än MBL. För svenska kan vi rapportera den högsta parsningskorrektheten hittills och för kinesiska och engelska är resultaten nära de bästa som har rapporterats.</p>

Cross-lingual transfer has been shown effective for dependency parsing of some low-resource languages. It typically requires closely related high-resource languages. Pre-trained deep language models significantly improve model performance in cross-lingual tasks. We evaluate cross-lingual model transfer on parsing Marathi, a low-resource language that does not have a closely related highresource language. In addition, we investigate monolingual modeling for comparison. We experiment with two state-of-the-art language models: mBERT and XLM-R. Our experimental results illustrate that the cross-lingual model transfer approach still holds with distantly related source languages, and models benefit most from XLM-R. We also evaluate the impact of multi-task learning by training all UD tasks simultaneously and find that it yields mixed results for dependency parsing and degrades the transfer performance of the best performing source language Ancient Greek.