In the past decade, the study of disease dynamics on complex networks has at­tracted great attention from both theoretical and empirical viewpoints. Under such a framework, people try to predict the outbreak of disease and propose im­munization mechanisms. However, this framework possesses a limitation, which makes it inconsistent with realistic cases. That is, this framework does not con­sider the impact of human behavior or decision-making progress on disease dy­namic characters and prevention measures. To further resolve this problem, we in this thesis propose behavioral epidemiology based on game theory, which in­volves the interactions between disease dynamics and human behavior in complex networks. Motivated by realistic cases, we proceed with the research from theo­retical models and consider the following aspects. We .rst re-construct a scheme of risk perception incorporating local and global information and show that this new evaluation scenario not only promotes vaccination uptake, but also eliminates the disease spreading. This interesting .nding could be attributed to the positive feedback mechanism between vaccination uptake and disease spreading. Then, we introduce a self-protection measure, which, due to low cost, can only provide tem­porary protection. By simulations and analysis we show that this measure leads to multiple e.ects: contrary with cases of low (high) e.ciency and cost of the self-protection measure, middle values drive more infection and larger cost, which is related to the loss of positive feedback between prevention measures and disease propagation. Subsequently, another scheme of adaptive protection is proposed, where a healthy agent can cut the connection with infected ones. We .nd that adaptive protection can e.ectively eradicate the disease and result in an optimal level of pruning infected links. Di.erent from these proposals focusing on indi­vidual interest, we lastly study a subsidy policy from the viewpoint of population bene.t. We .nd that disease can be well controlled with an increase of the vac­cination level, while the total expense reduces. Taken together, these .ndings of the thesis further demonstrate that the interplay between disease dynamics and human behavior plays an important role in the control of diseases. The models presented in this thesis, especially combining with empirical data, may serve as a foundation for further investigation of the subject in the future.

Semantics is the study of meaning and here we explore it through three major representations: brain, image and text. Researchers in the past have performed various studies to understand the similarities between semantic features across all the three representations. Distributional Semantic (DS) models or word vectors that are trained on text corpora have been widely used to study the convergence of semantic information in the human brain. Moreover, they have been incorporated into various NLP applications such as document categorization, speech to text and machine translation. Due to their widespread adoption by researchers and industry alike, it becomes imperative to test and evaluate the performance of di erent word vectors models. In this thesis, we publish the second iteration of BrainBench: a system designed to evaluate and benchmark word vectors using brain data by incorporating two new Italian brain datasets collected using fMRI and EEG technology. In the second half of the thesis, we explore semantics in Convolutional Neural Network (CNN). CNN is a computational model that is the state of the art technology for object recognition from images. However, these networks are currently considered a black-box and there is an apparent lack of understanding on why various CNN architectures perform better than the other. In this thesis, we also propose a novel method to understand CNNs by studying the semantic representation through its hierarchical layers. The convergence of semantic information in these networks is studied with the help of DS models following similar methodologies used to study semantics in the human brain. Our results provide substantial evidence that Convolutional Neural Networks do learn semantics from the images, and the features learned by the CNNs correlate to the semantics of the object in the image. Our methodology and results could potentially pave the way for improved design and debugging of CNNs. / Graduate

abstract: Automating aspects of biocuration through biomedical information extraction could significantly impact biomedical research by enabling greater biocuration throughput and improving the feasibility of a wider scope. An important step in biomedical information extraction systems is named entity recognition (NER), where mentions of entities such as proteins and diseases are located within natural-language text and their semantic type is determined. This step is critical for later tasks in an information extraction pipeline, including normalization and relationship extraction. BANNER is a benchmark biomedical NER system using linear-chain conditional random fields and the rich feature set approach. A case study with BANNER locating genes and proteins in biomedical literature is described. The first corpus for disease NER adequate for use as training data is introduced, and employed in a case study of disease NER. The first corpus locating adverse drug reactions (ADRs) in user posts to a health-related social website is also described, and a system to locate and identify ADRs in social media text is created and evaluated. The rich feature set approach to creating NER feature sets is argued to be subject to diminishing returns, implying that additional improvements may require more sophisticated methods for creating the feature set. This motivates the first application of multivariate feature selection with filters and false discovery rate analysis to biomedical NER, resulting in a feature set at least 3 orders of magnitude smaller than the set created by the rich feature set approach. Finally, two novel approaches to NER by modeling the semantics of token sequences are introduced. The first method focuses on the sequence content by using language models to determine whether a sequence resembles entries in a lexicon of entity names or text from an unlabeled corpus more closely. The second method models the distributional semantics of token sequences, determining the similarity between a potential mention and the token sequences from the training data by analyzing the contexts where each sequence appears in a large unlabeled corpus. The second method is shown to improve the performance of BANNER on multiple data sets. / Dissertation/Thesis / Ph.D. Computer Science 2013

The purpose of this thesis was the further development of a rule set used in an automatic text simplification system, and the exploration of whether it is possible to improve the performance of a rule based text simplification system by manual training. A first rule set was developed from a thor- ough literature review, and the rule refinement was performed by manually adapting the first rule set to a set of training texts. When there was no more change added to the set of rules, the training was considered to be completed, and the two sets were applied to a test set, for evaluation. This thesis evaluated the performance of a text simplification system as a clas- sification task, by the use of objective metrics: precision and recall. The comparison of the rule sets revealed a clear improvement of the system, since precision increased from 45% to 82%, and recall increased from 37% to 53%. Both recall and precision was improved after training for the ma- jority of the rules, with a few exceptions. All rule types resulted in a higher score on correctness for R2. Automatic text simplification systems target- ing real life readers need to account for qualitative aspects, which has not been considered in this thesis. Future evaluation should, in addition to quantitative metrics such as precision, recall, and complexity metrics, also account for the experience of the reader.

Aligning parallell terms in a parallell corpus can be done by aligning all words and phrases in the corpus and then performing term extraction on the aligned set of word pairs. Alternatively, term extraction in the source and target text can be made separately and then the resulting term candidates can be aligned, forming aligned parallell terms. This thesis describes an implementation of a word aligner that is applied on extracted term candidates in both the source and the target texts. The term aligner uses statistical measures, the tool Giza++ and heuristics in the search for alignments. The evaluation reveals that the best results are obtained when the term alignment relies heavily on the Giza++ tool and Levenshtein heuristic.

Due to the lack of short vowels or diacritics in Persian orthography, many Natural Language Processing applications for this language, including information retrieval, machine translation, text-to-speech, and automatic speech recognition systems need to disambiguate the input first, in order to be able to do further processing. In machine translation, for example, the whole text should be correctly diacritized first so that the correct words, parts of speech and meanings are matched and retrieved from the lexicon. This is primarily because of Persian’s ambiguous orthography. In fact, the core engine of any Persian language processor should utilize a diacritizer and a lexical disambiguator. This dissertation describes the design and implementation of an automatic diacritizer for Persian based on the state-of-the-art Finite State Transducer technology developed at Xerox by Beesley & Karttunen (2003). The result of morphological analysis and generation on a test corpus is shown, including the insertion of diacritics. This study will also look at issues that are raised by phonological and semantic ambiguities as a result of short vowels in Persian being absent in the writing system. It suggests a hybrid model (rule-based & inductive) that is inspired by psycholinguistic experiments on the human mental lexicon for the disambiguation of heterophonic homographs in Persian using frequency and collocation information. A syntactic parser can be developed based on the proposed model to discover Ezafe (the linking short vowel /e/ within a noun phrase) or disambiguate homographs, but its implementation is left for future work.

This thesis investigates the evaluation of automatic and manual text segmentation. Text segmentation is the process of placing boundaries within text to create segments according to some task-dependent criterion. An example of text segmentation is topical segmentation, which aims to segment a text according to the subjective definition of what constitutes a topic. A number of automatic segmenters have been created to perform this task, and the question that this thesis answers is how to select the best automatic segmenter for such a task. This requires choosing an appropriate segmentation evaluation metric, confirming the reliability of a manual solution, and then finally employing an evaluation methodology that can select the automatic segmenter that best approximates human performance. A variety of comparison methods and metrics exist for comparing segmentations (e.g., WindowDiff, Pk), and all save a few are able to award partial credit for nearly missing a boundary. Those comparison methods that can award partial credit unfortunately lack consistency, symmetricity, intuition, and a host of other desirable qualities. This work proposes a new comparison method named boundary similarity (B) which is based upon a new minimal boundary edit distance to compare two segmentations. Near misses are frequent, even among manual segmenters (as is exemplified by the low inter-coder agreement reported by many segmentation studies). This work adapts some inter-coder agreement coefficients to award partial credit for near misses using the new metric proposed herein, B. The methodologies employed by many works introducing automatic segmenters evaluate them simply in terms of a comparison of their output to one manual segmentation of a text, and often only by presenting nothing other than a series of mean performance values (along with no standard deviation, standard error, or little if any statistical hypothesis testing). This work asserts that one segmentation of a text cannot constitute a “true” segmentation; specifically, one manual segmentation is simply one sample of the population of all possible segmentations of a text and of that subset of desirable segmentations. This work further asserts that an adapted inter-coder agreement statistics proposed herein should be used to determine the reproducibility and reliability of a coding scheme and set of manual codings, and then statistical hypothesis testing using the specific comparison methods and methodologies demonstrated herein should be used to select the best automatic segmenter. This work proposes new segmentation evaluation metrics, adapted inter-coder agreement coefficients, and methodologies. Most important, this work experimentally compares the state-or-the-art comparison methods to those proposed herein upon artificial data that simulates a variety of scenarios and chooses the best one (B). The ability of adapted inter-coder agreement coefficients, based upon B, to discern between various levels of agreement in artificial and natural data sets is then demonstrated. Finally, a contextual evaluation of three automatic segmenters is performed using the state-of-the art comparison methods and B using the methodology proposed herein to demonstrate the benefits and versatility of B as opposed to its counterparts.

In this study, I constructed a named-entity linking system that maps between contextual word embeddings and knowledge graph embeddings to predict correct entities. To establish a named-entity linking system, I first applied named-entity recognition to identify the entities of interest. I then performed candidate generation via locality sensitivity hashing (LSH), where a candidate group of potential entities were created for each identified entity. Afterwards, my named-entity disambiguation component was performed to select the most probable candidate. By concatenating contextual word embeddings and knowledge graph embeddings in my disambiguation component, I present a novel approach to named-entity linking. I conducted the experiments with the Kensho-Derived Wikimedia Dataset and the AIDA CoNLL-YAGO Dataset; the former dataset was used for deployment and the later is a benchmark dataset for entity linking tasks. Three deep learning models were evaluated on the named-entity disambiguation component with different context embeddings. The evaluation was treated as a classification task, where I trained my models to select the correct entity from a list of candidates. By optimizing the named-entity linking through this methodology, this entire system can be used in recommendation engines with high F1 of 86% using the former dataset. With the benchmark dataset, the proposed method is able to achieve F1 of 79%.

<p>Statistički jezički model, u teoriji, predstavlja raspodelu verovatnoća nad skupom svih mogućih sekvenci reči nekog jezika. U praksi, to je mehanizam kojim se estimiraju verovatnoće sekvenci, koje su od interesa. Matematički aparat vezan za modele jezika je uglavnom nezavisan od jezika. Međutim, kvalitet obučenih modela ne zavisi samo od algoritama obuke, već prvenstveno od količine i kvaliteta podataka koji su na raspolaganju za obuku. Za jezike sa kompleksnom morfologijom, kao &scaron;to je srpski, tekstualni korpus za obuku modela mora biti daleko obimniji od korpusa koji bi se koristio kod nekog od jezika sa relativno jednostavnom morfologijom, poput engleskog. Ovo istraživanje obuhvata razvoj jezičkih modela za srpski jezik, počev&scaron;i od prikupljanja i inicijalne obrade tekstualnih sadržaja, preko adaptacije algoritama i razvoja metoda za re&scaron;avanje problema nedovoljne količine podataka za obuku, pa do prilagođavanja i primene modela u različitim tehnologijama, kao &scaron;to su sinteza govora na osnovu teksta, automatsko prepoznavanje govora, automatska detekcija i korekcija gramatičkih i semantičkih gre&scaron;aka u tekstovima, a postavljaju se i osnove za primenu jezičkih modela u automatskoj klasifikaciji dokumenata i drugim tehnologijama. Jezgro razvoja jezičkih modela za srpski predstavlja definisanje morfolo&scaron;kih klasa reči na osnovu informacija koje su sadržane u morfolo&scaron;kom rečniku, koji je nastao kao rezultat jednog od ranijih istraživanja.</p> / <p>A statistical language model, in theory, represents a probability distribution over sequences of words of a language. In practice, it is a tool for estimating probabilities of word sequences of interest. Mathematical basis related to language models is mostly language independent. However, the quality of trained models depends not only on training algorithms, but on the amount and quality of available training data as well. For languages with complex morphology, such as Serbian, textual corpora for training language models need to be significantly larger than the corpora needed for training language models for languages with relatively simple morphology, such as English. This research represents the entire process of developing language models for Serbian, starting with collecting and preprocessing of textual contents, extending to adaptation of algorithms and development of methods for addressing the problem of insufficient training data, and finally to adaptation and application of the models in different technologies, such as text-to-speech synthesis, automatic speech recognition, automatic detection and correction of grammar and semantic errors in texts, and determining basics for the application of the models in automatic document classification and other tasks. The core of the development of language models for Serbian is defining morphologic classes of words, based on the information contained within the morphologic dictionary of Serbian, which was one of the results of a previous research.</p>

Thesis (M. Sc. (Computer Science)) -- University of Limpopo, 2019 / The pronunciation of words and phrases in any language involves careful manipulation of linguistic features. Factors such as age, motivation, accent, phonetics, stress and intonation sometimes cause a problem of inappropriate or incorrect pronunciation of words from non-native languages. Pronunciation of words using different phonological rules has a tendency of changing the meaning of those words. This study presents the development of an automatic pronunciation assistant system for under-resourced languages of Limpopo Province, namely, Sepedi, Xitsonga, Tshivenda and isiNdebele. The aim of the proposed system is to help non-native speakers to learn appropriate and correct pronunciation of words/phrases in these under-resourced languages. The system is composed of a language identification module on the front-end side and a speech synthesis module on the back-end side. A support vector machine was compared to the baseline multinomial naive Bayes to build the language identification module. The language identification phase performs supervised multiclass text classification to predict a person’s first language based on input text before the speech synthesis phase continues with pronunciation issues using the identified language. The speech synthesis on the back-end phase is composed of four baseline text-to-speech synthesis systems in selected target languages. These text-to-speech synthesis systems were based on the hidden Markov model method of development. Subjective listening tests were conducted to evaluate the performance of the quality of the synthesised speech using a mean opinion score test. The mean opinion score test obtained good performance results on all targeted languages for naturalness, pronunciation, pleasantness, understandability, intelligibility, overall quality of the system and user acceptance. The developed system has been implemented on a “real-live” production web-server for performance evaluation and stability testing using live data.