<p>This thesis describes and introduces novel methods for knowledge discovery and presents a software system that is able to extract information from biomedical literature, review interesting connections between various biomedical concepts and in so doing, generates new hypotheses. The experimental results obtained by using methods described in this thesis, are compared to currently published results obtained by other methods and a number of case studies are described. This thesis shows how the technology&nbsp / resented can be integrated with the researchers&rsquo / own knowledge, experimentation and observations for optimal progression of scientific research.</p>

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

The overwhelming amount of online news presents a challenge called news information overload. To mitigate this challenge we propose a system to generate a causal network of news topics. To extract this information from distributed news sources, a system called Forest was developed. Forest retrieves documents that potentially contain causal information regarding a news topic. The documents are processed at a sentence level to extract causal relations and news topic references, these are the phases used to refer to a news topic. Forest uses a machine learning approach to classify causal sentences, and then renders the potential cause and effect of the sentences. The potential cause and effect are then classified as news topic references, these are the phrases used to refer to a news topics, such as “The World Cup” or “The Financial Meltdown”. Both classifiers use an algorithm developed within our working group, the algorithm performs better than several well known classification algorithms for the aforementioned tasks. In our evaluations we found that participants consider causal information useful to understand the news, and that while we can not extract causal information for all news topics, it is highly likely that we can extract causal relation for the most popular news topics. To evaluate the accuracy of the extractions made by Forest, we completed a user survey. We found that by providing the top ranked results, we obtained a high accuracy in extracting causal relations between news topics.

Unsupervised information extraction in open domain gains more and more importance recently by loosening the constraints on the strict definition of the extracted information and allowing to design more open information extraction systems. In this new domain of unsupervised information extraction, this thesis focuses on the tasks of extraction and clustering of relations between entities at a large scale. The objective of relation extraction is to discover unknown relations from texts. A relation prototype is first defined, with which candidates of relation instances are initially extracted with a minimal criterion. To guarantee the validity of the extracted relation instances, a two-step filtering procedures is applied: the first step with filtering heuristics to remove efficiently large amount of false relations and the second step with statistical models to refine the relation candidate selection. The objective of relation clustering is to organize extracted relation instances into clusters so that their relation types can be characterized by the formed clusters and a synthetic view can be offered to end-users. A multi-level clustering procedure is design, which allows to take into account the massive data and diverse linguistic phenomena at the same time. First, the basic clustering groups similar relation instances by their linguistic expressions using only simple similarity measures on a bag-of-word representation for relation instances to form high-homogeneous basic clusters. Second, the semantic clustering aims at grouping basic clusters whose relation instances share the same semantic meaning, dealing with more particularly phenomena such as synonymy or more complex paraphrase. Different similarities measures, either based on resources such as WordNet or distributional thesaurus, at the level of words, relation instances and basic clusters are analyzed. Moreover, a topic-based relation clustering is proposed to consider thematic information in relation clustering so that more precise semantic clusters can be formed. Finally, the thesis also tackles the problem of clustering evaluation in the context of unsupervised information extraction, using both internal and external measures. For the evaluations with external measures, an interactive and efficient way of building reference of relation clusters proposed. The application of this method on a newspaper corpus results in a large reference, based on which different clustering methods are evaluated.

Unsupervised information extraction in open domain gains more and more importance recently by loosening the constraints on the strict definition of the extracted information and allowing to design more open information extraction systems. In this new domain of unsupervised information extraction, this thesis focuses on the tasks of extraction and clustering of relations between entities at a large scale. The objective of relation extraction is to discover unknown relations from texts. A relation prototype is first defined, with which candidates of relation instances are initially extracted with a minimal criterion. To guarantee the validity of the extracted relation instances, a two-step filtering procedures is applied: the first step with filtering heuristics to remove efficiently large amount of false relations and the second step with statistical models to refine the relation candidate selection. The objective of relation clustering is to organize extracted relation instances into clusters so that their relation types can be characterized by the formed clusters and a synthetic view can be offered to end-users. A multi-level clustering procedure is design, which allows to take into account the massive data and diverse linguistic phenomena at the same time. First, the basic clustering groups similar relation instances by their linguistic expressions using only simple similarity measures on a bag-of-word representation for relation instances to form high-homogeneous basic clusters. Second, the semantic clustering aims at grouping basic clusters whose relation instances share the same semantic meaning, dealing with more particularly phenomena such as synonymy or more complex paraphrase. Different similarities measures, either based on resources such as WordNet or distributional thesaurus, at the level of words, relation instances and basic clusters are analyzed. Moreover, a topic-based relation clustering is proposed to consider thematic information in relation clustering so that more precise semantic clusters can be formed. Finally, the thesis also tackles the problem of clustering evaluation in the context of unsupervised information extraction, using both internal and external measures. For the evaluations with external measures, an interactive and efficient way of building reference of relation clusters proposed. The application of this method on a newspaper corpus results in a large reference, based on which different clustering methods are evaluated.

A lemmatiser is an important component of various human language technology applicalions for any language. At present, a rule-based le~nmatiserf or Afrikaans already exists, but this lermrlatiser produces disappoinringly low accuracy figures. The performimce of the current lemmatiser serves as motivation for developing another lemmatiser based on an alternative approach than language-specific rules. The alternalive method of lemmatiser corlstruction investigated in this study is memory-based learning. Thus, in this research project we develop an automatic lemmatiser for Afrikaans called Liu "Le~?rnru-idc~)~rifisv~ir'e Arfdr(i~ku~u-n s" 'hmmatiser for Afrikaans'. In order to construct Liu, thc following research objectives are sel: i) to define the classes for Afrikaans lemmatisation, ii) to determine the influence of data size and various feature options on the performance of I h , iii) to uutomalically determine the algorithm and parameters settings that deliver the best performancc in Lcrms of linguistic accuracy, execution time and memory usage. In order to achieve the first objective, we investigate the processes of inflecrion and derivation in Afrikaans, since automatic lemmatisation requires a clear distinction between inflection and derivation. We proceed to define the inflectional calegories for Afrikaans, which represent a number of affixes that should be removed from word-forms during lemmatisation. The classes for automatic lemmatisation in Afrikaans are derived from these affixes. It is subsequently shown that accuracy as well as memory usagc and execution lime increase as the amount of training dala is increased and that Ihe various feature options bave a significant effect on the performance of Lia. The algorithmic parameters and data representation that deliver the best results are determincd by the use of I'Senrck, a programme that implements Wrapped Progre~sive Sampling in order determine a set of possibly optimal algorithmic parameters for each of the TiMBL classification algorithms. Aulornaric Lcmlnalisa~ionf or Afrikaans - - Evaluation indicates that an accuracy figure of 92,896 is obtained when training Lia with the best performing parameters for the IB1 algorithm on feature-aligned data with 20 features. This result indicates that memory-based learning is indeed more suitable than rule-based methods for Afrikaans lenlmatiser construction. / Thesis (M.Ing. (Computer and Electronical Engineering))--North-West University, Potchefstroom Campus, 2007.

An important core technology in the development of human language technology applications is an automatic morphological analyser. Such a morphological analyser consists of various modules, one of which is a tokeniser. At present no tokeniser exists for Afrikaans and it has therefore been impossible to develop a morphological analyser for Afrikaans. Thus, in this research project such a tokeniser is being developed, and the project therefore has two objectives: i)to postulate a tag set for integrated tokenisation, and ii) to develop an algorithm for integrated tokenisation. In order to achieve the first object, a tag set for the tagging of sentences, named-entities, words, abbreviations and punctuation is proposed specifically for the annotation of Afrikaans texts. It consists of 51 tags, which can be expanded in future in order to establish a larger, more specific tag set. The postulated tag set can also be simplified according to the level of specificity required by the user. It is subsequently shown that an effective tokeniser cannot be developed using only linguistic, or only statistical methods. This is due to the complexity of the task: rule-based modules should be used for certain processes (for example sentence recognition), while other processes (for example named-entity recognition) can only be executed successfully by means of a machine-learning module. It is argued that a hybrid system (a system where rule-based and statistical components are integrated) would achieve the best results on Afrikaans tokenisation. Various rule-based and statistical techniques, including a TiMBL-based classifier, are then employed to develop such a hybrid tokeniser for Afrikaans. The final tokeniser achieves an ∫-score of 97.25% when the complete set of tags is used. For sentence recognition an ∫-score of 100% is achieved. The tokeniser also recognises 81.39% of named entities. When a simplified tag set (consisting of only 12 tags) is used to annotate named entities, the ∫-score rises to 94.74%. The conclusion of the study is that a hybrid approach is indeed suitable for Afrikaans sentencisation, named-entity recognition and tokenisation. The tokeniser will improve if it is trained with more data, while the expansion of gazetteers as well as the tag set will also lead to a more accurate system / Thesis (M.A. (Applied Language and Literary Studies))--North-West University, Potchefstroom Campus, 2006.

Text -to-speech systems currently are of great importance in the community. One core technology in this human language technology resource is stress assignment which plays an important role in any text-to-speech system. At present no automatic stress assigner for Afrikaans exists. For these reasons, the two most important aims of this project will be: a) to develop a complete and accurate set of stress rules for Afrikaans that can be implemented in an automatic stress assigner, and b) to develop an effective and highly accurate stress assigner in order to assign Afrikaans stress to words quickly and effectively. A set of stress rules for Afrikaans was developed in order to reach the first goal. It consists of 18 rules that are divided into groups for words that contain a schwa, derivations, and disyllabic, tri-syllabic and polysyllabic simplex words. Next, different approaches that can be used to develop a stress assigner were examined, and the rule-based approach was used to implement the developed stress rules within the stress assigner. The programming language, Perl, was chosen for the implementation of the rules. The chosen algorithm was used to generate a stress assigner for Afrikaans by implementing the stress rules developed. The hyphenator, Calomo and the compound analyser, CKarma was used to hyphenate all the test data and detect word boundaries within compounds. A dataset of 10 000 correctly annotated tokens was developed during the testing process. The evaluation of the stress assigner consists of four phases. During the first phase, the stress assigner was evaluated with the 10 000 tokens and achieved an accuracy of 92.09%. The grapheme - to-phoneme converter was evaluated with the same data and scored 91.9%. The influence of various factors on stress assignment was determined, and it was established that stress assignment is an essential component of rule-based grapheme-to-phoneme conversion. In conclusion, it can be said that the stress assigner achieved satisfactory results, and that the stress assigner can be successfully utilized in future projects to develop training data for further experiments with stress assignment and grapheme-to-phoneme conversion for Afrikaans. Experiments can be conducted in future with data-driven approaches that possibly may lead to better results in Afrikaans stress assignment and grapheme-to-phoneme conversion. / Thesis (M.A. (Applied Language and Literary Studies))--North-West University, Potchefstroom Campus, 2010.

Text -to-speech systems currently are of great importance in the community. One core technology in this human language technology resource is stress assignment which plays an important role in any text-to-speech system. At present no automatic stress assigner for Afrikaans exists. For these reasons, the two most important aims of this project will be: a) to develop a complete and accurate set of stress rules for Afrikaans that can be implemented in an automatic stress assigner, and b) to develop an effective and highly accurate stress assigner in order to assign Afrikaans stress to words quickly and effectively. A set of stress rules for Afrikaans was developed in order to reach the first goal. It consists of 18 rules that are divided into groups for words that contain a schwa, derivations, and disyllabic, tri-syllabic and polysyllabic simplex words. Next, different approaches that can be used to develop a stress assigner were examined, and the rule-based approach was used to implement the developed stress rules within the stress assigner. The programming language, Perl, was chosen for the implementation of the rules. The chosen algorithm was used to generate a stress assigner for Afrikaans by implementing the stress rules developed. The hyphenator, Calomo and the compound analyser, CKarma was used to hyphenate all the test data and detect word boundaries within compounds. A dataset of 10 000 correctly annotated tokens was developed during the testing process. The evaluation of the stress assigner consists of four phases. During the first phase, the stress assigner was evaluated with the 10 000 tokens and achieved an accuracy of 92.09%. The grapheme - to-phoneme converter was evaluated with the same data and scored 91.9%. The influence of various factors on stress assignment was determined, and it was established that stress assignment is an essential component of rule-based grapheme-to-phoneme conversion. In conclusion, it can be said that the stress assigner achieved satisfactory results, and that the stress assigner can be successfully utilized in future projects to develop training data for further experiments with stress assignment and grapheme-to-phoneme conversion for Afrikaans. Experiments can be conducted in future with data-driven approaches that possibly may lead to better results in Afrikaans stress assignment and grapheme-to-phoneme conversion. / Thesis (M.A. (Applied Language and Literary Studies))--North-West University, Potchefstroom Campus, 2010.

An approach to biomedical language processing is to apply existing natural language processing (NLP) solutions to biomedical texts. Often, existing NLP solutions are less successful in the biomedical domain relative to their non-biomedical domain performance (e.g., relative to newspaper text). Biomedical NLP is likely best served by methods, information and tools that account for its particular challenges. In this thesis, I describe an NLP system specifically engineered for sentinel event extraction from clinical documents. The NLP system's design accounts for several biomedical NLP challenges. The specific contributions are as follows. - Biomedical tokenizers differ, lack consensus over output tokens and are difficult to extend. I developed an extensible tokenizer, providing a tokenizer design pattern and implementation guidelines. It evaluated as equivalent to a leading biomedical tokenizer (MedPost). - Biomedical part-of-speech (POS) taggers are often trained on non-biomedical corpora and applied to biomedical corpora. This results in a decrease in tagging accuracy. I built a token centric POS tagger, TcT, that is more accurate than three existing POS taggers (mxpost, TnT and Brill) when trained on a non-biomedical corpus and evaluated on biomedical corpora. TcT achieves this increase in tagging accuracy by ignoring previously assigned POS tags and restricting the tagger's scope to the current token, previous token and following token. - Two parsers, MST and Malt, have been evaluated using perfect POS tag input. Given that perfect input is unlikely in biomedical NLP tasks, I evaluated these two parsers on imperfect POS tag input and compared their results. MST was most affected by imperfectly POS tagged biomedical text. I attributed MST's drop in performance to verbs and adjectives where MST had more potential for performance loss than Malt. I attributed Malt's resilience to POS tagging errors to its use of a rich feature set and a local scope in decision making. - Previous automated clinical coding (ACC) research focuses on mapping narrative phrases to terminological descriptions (e.g., concept descriptions). These methods make little or no use of the additional semantic information available through topology. I developed a token-based ACC approach that encodes tokens and manipulates token-level encodings by mapping linguistic structures to topological operations in SNOMED CT. My ACC method recalled most concepts given their descriptions and performed significantly better than MetaMap. I extended my contributions for the purpose of sentinel event extraction from clinical letters. The extensions account for negation in text, use medication brand names during ACC and model (coarse) temporal information. My software system's performance is similar to state-of-the-art results. Given all of the above, my thesis is a blueprint for building a biomedical NLP system. Furthermore, my contributions likely apply to NLP systems in general. / Graduate