No description available.

There may be several mechanisms that drive observed interactions between plants and pollinators in an ecosystem, many of which may involve trait matching or trait complementarity. Hence a model of insect species activity on plant species should be represented as a mixture of these linkage rules. Unfortunately, ecologists do not always know how many, or even which, traits are the main contributors to the observed interactions. This thesis proposes the Latent Dirichlet Allocation (LDA) model from artificial intelligence for modelling the observed interactions in an ecosystem as a finite mixture of (latent) interaction groups in which plant and pollinator pairs that share common linkage rules are placed in the same interaction group. Several model selection criteria are explored for estimating how many interaction groups best describe the observed interactions. This thesis also introduces a new model selection score called ``penalized perplexity". The performance of the model selection criteria, and of LDA in general, are evaluated through a comprehensive simulation study that consider networks of various size along with varying levels of nesting and numbers of interaction groups. Results of the simulation study suggest that LDA works well on networks with mild-to-no nesting, but loses accuracy with increased nestedness. Further, the penalized perplexity tended to outperform the other model selection criteria in identifying the correct number of interaction groups used to simulate the data. Finally, LDA was demonstrated on a real network, the results of which provided insights into the functional roles of pollinator species in the study region.

In the big data context, the growing volume of textual data presents challenges for traditional structured data-based decision support systems (DSS). DSS based on structured data is difficult to process the semantic information of text data. To meet the challenge, this thesis proposes a solution for the Decision Support System (DSS) based on machine Learning and qualitative analysis, namely TLE-DSS. TLE-DSS refers to three critical analytical modules: Thematic Analysis (TA), Latent Dirichlet Allocation (LDA)and Evolutionary Grounded Theory (EGT). To better understand the operation mechanism of TLE-DSS, this thesis used an experimental case to explain how to make decisions through TLE-DSS. Additionally, during the data analysis of the experimental case, by calculating the difference of perplexity of different models to compare similarities, this thesis proposed a solution to determine the optimal number of topics in LDA. Meanwhile, by using LDAvis, a model with the optimal number of topics was visualized. Moreover, the thesis also expounded the principle and application value of EGT. In the last part, this thesis discussed the challenges and potential ethical issues that TLE-DSS still faces.

The thesis describes a new database of uzun havas, a non-metered structured improvisation form in Turkish folk music, and a system, which uses Variable-Length Markov Models (VLMMs) to predict the melody in the uzun hava form. The database consists of 77 songs, encompassing 10849 notes, and it is used to train multiple viewpoints, where each event in a musical sequence are represented by parallel descriptors such as Durations and Notes. The thesis also introduces pitch-related viewpoints that are specifically aimed to model the unique melodic properties of makam music. The predictability of the system is quantitatively evaluated by an entropy based scheme. In the experiments, the results from the pitch-related viewpoints mapping 12-tone-scale of Western classical theory and 17 tone-scale of Turkish folk music are compared. It is shown that VLMMs are highly predictive in the note progressions of the transcriptions of uzun havas. This suggests that VLMMs may be applied to makam-based and non-metered musical forms, in addition to Western musical styles. To the best of knowledge, the work presents the first symbolic, machine-readable database and the first application of computational modeling in Turkish folk music.

People read for various purposes like learning specific skills, acquiring foreign languages, and enjoying the pure reading experience, etc. This kind of pure enjoyment may credit to many aspects, such as the aesthetics of languages, the beauty of rhyme, and the entertainment of being surprised by what will happen next, the last of which is typically featured in fictional narratives and is also the main topic of this project. In other words, “good” fiction may be better at entertaining readers by baffling and eluding their expectations whereas “normal” narratives may contain more cliches and ready-made sentences that are easy to predict. Therefore, this project examines whether “good” fiction is less predictable than “normal” fiction, the two of which are predefined as canonized and non-canonized.  The predictability can be statistically reflected by the probability of the next words being correctly predicted given the previous content, which is then further measured in the metric of perplexity. Thanks to recent advances in deep learning, language models based on neural networks with billions of parameters can now be trained on terabytes of text to improve their performance in predicting the next unseen texts. Therefore, the generative pre-trained modeling and the text generator are combined to estimate the perplexities of canonized literature and non-canonized literature.  Due to the potential risk that the terabytes of text on which the advanced models have been trained may contain book content within the corpus, two series of models are designed to yield non-biased perplexity results, namely the self-trained models and the generative pre-trained Transformer-2 models. The comparisons of these two groups of results set up the final hierarchy of architecture constituted by five models for further experiments.  Over the process of perplexity estimation, the perplexity variance can also be generated at the same time, which is then used to denote how predictability varies across sequences with a certain length within each piece of literature. Evaluated by the perplexity variance, the literature property of homogeneity can also be examined between these two groups of literature.  The ultimate results from the five models imply that there lie distinctions in both perplexity values and variances between the canonized literature and non-canonized literature. Besides, the canonized literature shows higher perplexity values and variances measured in both median and mean metrics, which denotes that it is less predictable and homogeneous than the non-canonized literature.  Obviously, the perplexity values and variances cannot be used to define the literary quality directly. However, they offer some signals that the metric of perplexity can be insightful in the literary quality analysis using natural language processing techniques.

The availability of the stenographic manuscripts of Astrid Lindgren have sparked an interest in the creation of a language model for stenography. By its very nature stenography is low-resource and the unavailability of data requires a tool for using normal data. The tool presented in this thesis is to create stenographic data from manipulating orthographic data. Stenographic data is distinct from orthographic data through three different types manipulations that can be carried out. Firstly stenography is based on a phonetic version of language, secondly it used its own alphabet that is distinct from normal orthographic data, and thirdly it used several techniques to compress the data. The first type of manipulation is done by using a grapheme-to-phoneme converter. The second type is done by using an orthographic representation of a stenographic alphabet. The third type of manipulation is done by manipulating based on subword level, word level and phrase level. With these manipulations different datasets are created with different combinations of these manipulations. Results are measured for both perplexity on a GPT-2 language model and for compression rate on the different datasets. These results show a general decrease of perplexity scores and a slight compression rate across the board. We see that the lower perplexity scores are possibly due to the growth of ambiguity.

Prvotním cílem tohoto projektu bylo prostudovat problematiku jazykového modelování pro rozpoznávání řeči a techniky pro získávání textových dat z Webu. Text představuje základní techniky rozpoznávání řeči a detailněji popisuje jazykové modely založené na statistických metodách. Zvláště se práce zabývá kriterii pro vyhodnocení kvality jazykových modelů a systémů pro rozpoznávání řeči. Text dále popisuje modely a techniky dolování dat, zvláště vyhledávání informací. Dále jsou představeny problémy spojené se získávání dat z webu, a v kontrastu s tím je představen vyhledávač Google. Součástí projektu byl návrh a implementace systému pro získávání textu z webu, jehož detailnímu popisu je věnována náležitá pozornost. Nicméně, hlavním cílem práce bylo ověřit, zda data získaná z Webu mohou mít nějaký přínos pro rozpoznávání řeči. Popsané techniky se tak snaží najít optimální způsob, jak data získaná z Webu použít pro zlepšení ukázkových jazykových modelů, ale i modelů nasazených v reálných rozpoznávacích systémech.