This thesis is about the problem of compositionality in distributional semantics. Distributional semantics presupposes that the meanings of words are a function of their occurrences in textual contexts. It models words as distributions over these contexts and represents them as vectors in high dimensional spaces. The problem of compositionality for such models concerns itself with how to produce distributional representations for larger units of text (such as a verb and its arguments) by composing the distributional representations of smaller units of text (such as individual words). This thesis focuses on a particular approach to this compositionality problem, namely using the categorical framework developed by Coecke, Sadrzadeh, and Clark, which combines syntactic analysis formalisms with distributional semantic representations of meaning to produce syntactically motivated composition operations. This thesis shows how this approach can be theoretically extended and practically implemented to produce concrete compositional distributional models of natural language semantics. It furthermore demonstrates that such models can perform on par with, or better than, other competing approaches in the field of natural language processing. There are three principal contributions to computational linguistics in this thesis. The first is to extend the DisCoCat framework on the syntactic front and semantic front, incorporating a number of syntactic analysis formalisms and providing learning procedures allowing for the generation of concrete compositional distributional models. The second contribution is to evaluate the models developed from the procedures presented here, showing that they outperform other compositional distributional models present in the literature. The third contribution is to show how using category theory to solve linguistic problems forms a sound basis for research, illustrated by examples of work on this topic, that also suggest directions for future research.

Background: MicrO is an ontology of microbiological terms, including prokaryotic qualities and processes, material entities (such as cell components), chemical entities (such as microbiological culture media and medium ingredients), and assays. The ontology was built to support the ongoing development of a natural language processing algorithm, MicroPIE (or, Microbial Phenomics Information Extractor). During the MicroPIE design process, we realized there was a need for a prokaryotic ontology which would capture the evolutionary diversity of phenotypes and metabolic processes across the tree of life, capture the diversity of synonyms and information contained in the taxonomic literature, and relate microbiological entities and processes to terms in a large number of other ontologies, most particularly the Gene Ontology (GO), the Phenotypic Quality Ontology (PATO), and the Chemical Entities of Biological Interest (ChEBI). We thus constructed MicrO to be rich in logical axioms and synonyms gathered from the taxonomic literature. Results: MicrO currently has similar to 14550 classes (similar to 2550 of which are new, the remainder being microbiologically-relevant classes imported from other ontologies), connected by similar to 24,130 logical axioms (5,446 of which are new), and is available at (http://purl.obolibrary.org/obo/MicrO.owl) and on the project website at https://github.com/carrineblank/MicrO. MicrO has been integrated into the OBO Foundry Library (http://www.obofoundry.org/ontology/micro.html), so that other ontologies can borrow and re-use classes. Term requests and user feedback can be made using MicrO's Issue Tracker in GitHub. We designed MicrO such that it can support the ongoing and future development of algorithms that can leverage the controlled vocabulary and logical inference power provided by the ontology. Conclusions: By connecting microbial classes with large numbers of chemical entities, material entities, biological processes, molecular functions, and qualities using a dense array of logical axioms, we intend MicrO to be a powerful new tool to increase the computing power of bioinformatics tools such as the automated text mining of prokaryotic taxonomic descriptions using natural language processing. We also intend MicrO to support the development of new bioinformatics tools that aim to develop new connections between microbial phenotypes and genotypes (i.e., the gene content in genomes). Future ontology development will include incorporation of pathogenic phenotypes and prokaryotic habitats.

Ever since we entered the digital communication era, the ease of information sharing through the internet has encouraged online literature searching. With this comes the potential risk of a rise in academic misconduct and intellectual property theft. As concerns over plagiarism grow, more attention has been directed towards automatic plagiarism detection. This is a computational approach which assists humans in judging whether pieces of texts are plagiarised. However, most existing plagiarism detection approaches are limited to super cial, brute-force stringmatching techniques. If the text has undergone substantial semantic and syntactic changes, string-matching approaches do not perform well. In order to identify such changes, linguistic techniques which are able to perform a deeper analysis of the text are needed. To date, very limited research has been conducted on the topic of utilising linguistic techniques in plagiarism detection. This thesis provides novel perspectives on plagiarism detection and plagiarism direction identi cation tasks. The hypothesis is that original texts and rewritten texts exhibit signi cant but measurable di erences, and that these di erences can be captured through statistical and linguistic indicators. To investigate this hypothesis, four main research objectives are de ned. First, a novel framework for plagiarism detection is proposed. It involves the use of Natural Language Processing techniques, rather than only relying on the vii traditional string-matching approaches. The objective is to investigate and evaluate the in uence of text pre-processing, and statistical, shallow and deep linguistic techniques using a corpus-based approach. This is achieved by evaluating the techniques in two main experimental settings. Second, the role of machine learning in this novel framework is investigated. The objective is to determine whether the application of machine learning in the plagiarism detection task is helpful. This is achieved by comparing a thresholdsetting approach against a supervised machine learning classi er. Third, the prospect of applying the proposed framework in a large-scale scenario is explored. The objective is to investigate the scalability of the proposed framework and algorithms. This is achieved by experimenting with a large-scale corpus in three stages. The rst two stages are based on longer text lengths and the nal stage is based on segments of texts. Finally, the plagiarism direction identi cation problem is explored as supervised machine learning classi cation and ranking tasks. Statistical and linguistic features are investigated individually or in various combinations. The objective is to introduce a new perspective on the traditional brute-force pair-wise comparison of texts. Instead of comparing original texts against rewritten texts, features are drawn based on traits of texts to build a pattern for original and rewritten texts. Thus, the classi cation or ranking task is to t a piece of text into a pattern. The framework is tested by empirical experiments, and the results from initial experiments show that deep linguistic analysis contributes to solving the problems we address in this thesis. Further experiments show that combining shallow and viii deep techniques helps improve the classi cation of plagiarised texts by reducing the number of false negatives. In addition, the experiment on plagiarism direction detection shows that rewritten texts can be identi ed by statistical and linguistic traits. The conclusions of this study o er ideas for further research directions and potential applications to tackle the challenges that lie ahead in detecting text reuse.

Open-domain question answering (QA) is an established NLP task which enables users to search for speciVc pieces of information in large collections of texts. Instead of using keyword-based queries and a standard information retrieval engine, QA systems allow the use of natural language questions and return the exact answer (or a list of plausible answers) with supporting snippets of text. In the past decade, open-domain QA research has been dominated by evaluation fora such as TREC and CLEF, where shallow techniques relying on information redundancy have achieved very good performance. However, this performance is generally limited to simple factoid and deVnition questions because the answer is usually explicitly present in the document collection. Current approaches are much less successful in Vnding implicit answers and are diXcult to adapt to more complex question types which are likely to be posed by users. In order to advance the Veld of QA, this thesis proposes a shift in focus from simple factoid questions to encyclopaedic questions: list questions composed of several constraints. These questions have more than one correct answer which usually cannot be extracted from one small snippet of text. To correctly interpret the question, systems need to combine classic knowledge-based approaches with advanced NLP techniques. To Vnd and extract answers, systems need to aggregate atomic facts from heterogeneous sources as opposed to simply relying on keyword-based similarity. Encyclopaedic questions promote QA systems which use basic reasoning, making them more robust and easier to extend with new types of constraints and new types of questions. A novel semantic architecture is proposed which represents a paradigm shift in open-domain QA system design, using semantic concepts and knowledge representation instead of words and information retrieval. The architecture consists of two phases, analysis – responsible for interpreting questions and Vnding answers, and feedback – responsible for interacting with the user. This architecture provides the basis for EQUAL, a semantic QA system developed as part of the thesis, which uses Wikipedia as a source of world knowledge and iii employs simple forms of open-domain inference to answer encyclopaedic questions. EQUAL combines the output of a syntactic parser with semantic information from Wikipedia to analyse questions. To address natural language ambiguity, the system builds several formal interpretations containing the constraints speciVed by the user and addresses each interpretation in parallel. To Vnd answers, the system then tests these constraints individually for each candidate answer, considering information from diUerent documents and/or sources. The correctness of an answer is not proved using a logical formalism, instead a conVdence-based measure is employed. This measure reWects the validation of constraints from raw natural language, automatically extracted entities, relations and available structured and semi-structured knowledge from Wikipedia and the Semantic Web. When searching for and validating answers, EQUAL uses the Wikipedia link graph to Vnd relevant information. This method achieves good precision and allows only pages of a certain type to be considered, but is aUected by the incompleteness of the existing markup targeted towards human readers. In order to address this, a semantic analysis module which disambiguates entities is developed to enrich Wikipedia articles with additional links to other pages. The module increases recall, enabling the system to rely more on the link structure of Wikipedia than on word-based similarity between pages. It also allows authoritative information from diUerent sources to be linked to the encyclopaedia, further enhancing the coverage of the system. The viability of the proposed approach was evaluated in an independent setting by participating in two competitions at CLEF 2008 and 2009. In both competitions, EQUAL outperformed standard textual QA systems as well as semi-automatic approaches. Having established a feasible way forward for the design of open-domain QA systems, future work will attempt to further improve performance to take advantage of recent advances in information extraction and knowledge representation, as well as by experimenting with formal reasoning and inferencing capabilities.

Nowadays, the effective management of information is extremely important for all spheres of our lives and applications such as search engines and question answering systems help users to find the information that they need. However, even when assisted by these various applications, people sometimes struggle to find what they want. For example, when choosing a product customers can be confused by the need to consider many features before they can reach a decision. Interactive question answering (IQA) systems can help customers in this process, by answering questions about products and initiating a dialogue with the customers when their needs are not clearly defined. The focus of this thesis is how to design an interactive question answering system that will assist users in choosing a product they are looking for, in an optimal way, when a large number of similar products are available. Such an IQA system will be based on selecting a set of characteristics (also referred to as product features in this thesis), that describe the relevant product, and narrowing the search space. We believe that the order in which these characteristics are presented in terms of these IQA sessions is of high importance. Therefore, they need to be ranked in order to have a dialogue which selects the product in an efficient manner. The research question investigated in this thesis is whether product characteristics mentioned in user reviews are important for a person who is likely to purchase a product and can therefore be used when designing an IQA system. We focus our attention on products such as mobile phones; however, the proposed techniques can be adapted for other types of products if the data is available. Methods from natural language processing (NLP) fields such as coreference resolution, relation extraction and opinion mining are combined to produce various rankings of phone features. The research presented in this thesis employs two corpora which contain texts related to mobile phones specifically collected for this thesis: a corpus of Wikipedia articles about mobile phones and a corpus of mobile phone reviews published on the Epinions.com website. Parts of these corpora were manually annotated with coreference relations, mobile phone features and relations between mentions of the phone and its features. The annotation is used to develop a coreference resolution module as well as a machine learning-based relation extractor. Rule-based methods for identification of coreference chains describing the phone are designed and thoroughly evaluated against the annotated gold standard. Machine learning is used to find links between mentions of the phone (identified by coreference resolution) and phone features. It determines whether some phone feature belong to the phone mentioned in the same sentence or not. In order to find the best rankings, this thesis investigates several settings. One of the hypotheses tested here is that the relatively low results of the proposed baseline are caused by noise introduced by sentences which are not directly related to the phone and phone feature. To test this hypothesis, only sentences which contained mentions of the mobile phone and a phone feature linked to it were processed to produce rankings of the phones features. Selection of the relevant sentences is based on the results of coreference resolution and relation extraction. Another hypothesis is that opinionated sentences are a good source for ranking the phone features. In order to investigate this, a sentiment classification system is also employed to distinguish between features mentioned in positive and negative contexts. The detailed evaluation and error analysis of the methods proposed form an important part of this research and ensure that the results provided in this thesis are reliable.

Permission from the author to release this work as open access is pending. Please contact the ICS library if you would like to view this work.

Semantic interoperability is essential on the Semantic Web to enable different information systems to exchange data. Ontology matching has been recognised as a means to achieve semantic interoperability on the Web by identifying similar information in heterogeneous ontologies. Existing ontology matching approaches have two major limitations. The first limitation relates to similarity metrics, which provide a pessimistic value when considering complex objects such as strings and conceptual entities. The second limitation relates to the role of description logic reasoning. In particular, most approaches disregard implicit information about entities as a source of background knowledge. In this thesis, we first present a new similarity function, called the degree of commonality coefficient, to compute the overlap between two sets based on the similarity between their elements. The results of our evaluations show that the degree of commonality performs better than traditional set similarity metrics in the ontology matching task. Secondly, we have developed the Knowledge Organisation System Implicit Mapping (KOSIMap) framework, which differs from existing approaches by using description logic reasoning (i) to extract implicit information as background knowledge for every entity, and (ii) to remove inappropriate correspondences from an alignment. The results of our evaluation show that the use of Description Logic in the ontology matching task can increase coverage. We identify people interested in ontology matching and reasoning techniques as the target audience of this work

Recommender systems such as Amazon, offer uses recommendations, or suggestions of items to try or buy. We propose a novel classification of reasons for including explanations in recommender systems. Our focus is on the aim of effectiveness, or decision support, and we contrast it with other metrics such as satisfaction and persuasion. In user studies, we found that people varied in the features they found important, and composed a short list of features in two domains (movies and cameras). We then built a natural language explanation testbed system, considering these features as well as the limitations of using commercial data. This testbed was used in a series of experiments to test whether personalization of explanations affects effectiveness, persuasion and satisfaction. We chose a simple form of personalization which considers likely constraints of a recommender system (e.g. limited meta-data related to the user) as well as brevity. In these experiments we found that: 1. Explanations help participants to make decisions compared to recommendations without explanations, we saw as a significant decrease in opt-outs in item ratings – participants were more likely to be able to give an initial rating for an item if they were given an explanation, and the likelihood of receiving a rating increased for feature-based explanations compared to a baseline. 2. Contrary to our initial hypothesis, our method of personalization could damage effectiveness for both movies and cameras which are domains that differ with regard to two dimensions which we found affected perceived effectiveness: cost (low vs. high), and valuation type (subjective vs. objective). 3. Participants were more satisfied with feature-based than baseline explanations. If the personalization is perceived as relevant to them, then personalized feature-based explanations were preferred over non-personalized. 4. Satisfaction with explanation was also reflected in the proportion of opt-outs. The opt-out rate for the explanations was highest in the baseline for all experiments. This was the case despite the different types of explanation baselines used in the two domains.

Syntactic parsing is one of the best understood language processing applications. Since language and grammar have been formally defined, it is easy for computers to parse the syntactic structure of natural language text. Does meaning have structure as well? If it has, how can we analyze the structure? Previous systems rely on a one-to-one correspondence between syntactic rules and semantic rules. But such systems can only be applied to limited fragments of English. In this thesis, we propose a general-purpose shallow semantic parser which utilizes a semantic network (WordNet), and a frame dataset (FrameNet). Semantic relations recognized by the parser are based on how human beings represent knowledge of the world. Parsing semantic structure allows semantic units and constituents to be accessed and processed in a more meaningful way than syntactic parsing, moving the automation of understanding natural language text to a higher level.

Background: Taxonomic descriptions are traditionally composed in natural language and published in a format that cannot be directly used by computers. The Exploring Taxon Concepts (ETC) project has been developing a set of web-based software tools that convert morphological descriptions published in telegraphic style to character data that can be reused and repurposed. This paper introduces the first semi-automated pipeline, to our knowledge, that converts morphological descriptions into taxon-character matrices to support systematics and evolutionary biology research. We then demonstrate and evaluate the use of the ETC Input Creation - Text Capture - Matrix Generation pipeline to generate body part measurement matrices from a set of 188 spider morphological descriptions and report the findings. Results: From the given set of spider taxonomic publications, two versions of input (original and normalized) were generated and used by the ETC Text Capture and ETC Matrix Generation tools. The tools produced two corresponding spider body part measurement matrices, and the matrix from the normalized input was found to be much more similar to a gold standard matrix hand-curated by the scientist co-authors. Special conventions utilized in the original descriptions (e.g., the omission of measurement units) were attributed to the lower performance of using the original input. The results show that simple normalization of the description text greatly increased the quality of the machine-generated matrix and reduced edit effort. The machine-generated matrix also helped identify issues in the gold standard matrix. Conclusions: ETC Text Capture and ETC Matrix Generation are low-barrier and effective tools for extracting measurement values from spider taxonomic descriptions and are more effective when the descriptions are self-contained. Special conventions that make the description text less self contained challenge automated extraction of data from biodiversity descriptions and hinder the automated reuse of the published knowledge. The tools will be updated to support new requirements revealed in this case study.