The expansion of the World Wide Web and the increasing popularity of microblogging websites such as Twitter and Facebook has created massive stores of textual data that is short in length. Although traditional topic models have proven to work successfully on collections of long texts such as books and news articles, they tend to produce results that are less coherent when applied to short text, such as status messages and product reviews. Over the last few decades it has become of greater relevance to analyse short texts due to the realisation that such bodies of text could potentially hold useful information. Latent Dirichlet allocation (LDA) is one of the most popular topic models and it makes the generative assumption that each document contains multiple topics in varying proportions, which is a sensible assumption about long text. On the other hand, the Dirichlet multinomial mixture model (GSDMM), a seemingly less popular topic model, assumes that a document can only belong to a single topic, which seems to be a more appropriate assumption for short text. The objective of this paper is to investigate the hypothesis that GSDMM will outperform LDA on short text, using topic coherence and stability as performance measures.

Most topic models are constructed under the assumption that documents follow a multinomial distribution. The Poisson distribution is an alternative distribution to describe the probability of count data. For topic modelling, the Poisson distribution describes the number of occurrences of a word in documents of fixed length. The Poisson distribution has been successfully applied in text classification, but its application to topic modelling is not well documented, specifically in the context of a generative probabilistic model. Furthermore, the few Poisson topic models in the literature are admixture models, making the assumption that a document is generated from a mixture of topics. In this study, we focus on short text. Many studies have shown that the simpler assumption of a mixture model fits short text better. With mixture models, as opposed to admixture models, the generative assumption is that a document is generated from a single topic. One topic model, which makes this one-topic-per-document assumption, is the Dirichlet-multinomial mixture model. The main contributions of this work are a new Gamma-Poisson mixture model, as well as a collapsed Gibbs sampler for the model. The benefit of the collapsed Gibbs sampler derivation is that the model is able to automatically select the number of topics contained in the corpus. The results show that the Gamma-Poisson mixture model performs better than the Dirichlet-multinomial mixture model at selecting the number of topics in labelled corpora. Furthermore, the Gamma-Poisson mixture produces better topic coherence scores than the Dirichlet-multinomial mixture model, thus making it a viable option for the challenging task of topic modelling of short text.

Bimodal distributions have rarely been studied although they appear frequently in datasets. We develop a novel bimodal distribution based on the triangular distribution and then expand it to the multivariate case using a Gaussian copula. To determine the goodness of fit of the univariate model, we use the Kolmogorov–Smirnov (KS) and Cramér–von Mises (CVM) tests. The contributions of this work are that a simplistic yet robust distribution was developed to deal with bimodality in data, a multivariate distribution was developed as a generalisation of this univariate distribution using a Gaussian copula, a comparison between parametric and semi-parametric approaches to modelling bimodality is given, and an R package called btld is developed from the workings of this paper.

Most topic models are constructed under the assumption that documents follow a multinomial distribution. The Poisson distribution is an alternative distribution to describe the probability of count data. For topic modelling, the Poisson distribution describes the number of occurrences of a word in documents of fixed length. The Poisson distribution has been successfully applied in text classification, but its application to topic modelling is not well documented, specifically in the context of a generative probabilistic model. Furthermore, the few Poisson topic models in literature are admixture models, making the assumption that a document is generated from a mixture of topics. In this study, we focus on short text. Many studies have shown that the simpler assumption of a mixture model fits short text better. With mixture models, as opposed to admixture models, the generative assumption is that a document is generated from a single topic. One topic model, which makes this one-topic-per-document assumption, is the Dirichlet-multinomial mixture model. The main contributions of this work are a new Gamma-Poisson mixture model, as well as a collapsed Gibbs sampler for the model. The benefit of the collapsed Gibbs sampler derivation is that the model is able to automatically select the number of topics contained in the corpus. The results show that the Gamma-Poisson mixture model performs better than the Dirichlet-multinomial mixture model at selecting the number of topics in labelled corpora. Furthermore, the Gamma-Poisson mixture produces better topic coherence scores than the Dirichlet-multinomial mixture model, thus making it a viable option for the challenging task of topic modelling of short text.

Appropriate training data is a prerequisite for the development of natural language processing (NLP) techniques. Vast amounts of language data are typically required to develop NLP tools that perform at state-of-the-art level. Such abundant resources are currently only available in a few languages. The remaining languages have to find alternative ways to become ``NLP-enabled''. The aim of the study reported on here is to make more language data available to support NLP development in the official languages of South Africa. In this paper we present the idea of generating text data by means of translation. We also propose the use of topic modelling to identify text in a highly resourced source language that will yield meaningful translations in under-resourced target languages. More specifically, the paper describes how topic modelling was used to identify English Wikipedia articles that should be suitable for translation into South Africa's 10 other official languages.