Now showing 1 - 3 of 3
  • Publication
    Synthetic Dataset Generation for Online Topic Modeling
    Online topic modeling allows for the discovery of the underlying latent structure in a real time stream of data. In the evaluation of such approaches it is common that a static value for the number of topics is chosen. However, we would expect the number of topics to vary over time due to changes in the underlying structure of the data, known as concept drift and concept shift. We propose a semi-synthetic dataset generator, which can introduce concept drift and concept shift into existing annotated non-temporal datasets, via user-controlled paramaterization. This allows for the creation of multiple different artificial streams of data, where the “correct” number and composition of the topics is known at each point in time. We demonstrate how these generated datasets can be used as an evaluation strategy for online topic modeling approaches.
      267
  • Publication
    Ensemble Topic Modeling via Matrix Factorization
    (CEUR Workshop Proceedings, 2016-09-21) ; ;
    Topic models can provide us with an insight into the underlying latent structure of a large corpus of documents, facilitating knowledge discovery and information summarization. A range of methods have been proposed in the literature, including probabilistic topic models and techniques based on matrix factorization. However, these methods tend to have stochastic elements in their initialization, which can lead to their output being unstable. That is, if a topic modeling algorithm is applied to the same data multiple times, the output will not necessarily always be the same. With this idea of stability in mind we ask the question – how can we produce a definitive topic model that is both stable and accurate? To address this, we propose a new ensemble topic modeling method, based on Non-negative Matrix Factorization (NMF), which combines a collection of unstable topic models to produce a definitive output. We evaluate this method on an annotated tweet corpus, where we show that this new approach is more accurate and stable than traditional NMF.
      352
  • Publication
    Tweeting Europe: A text-analytic approach to unveiling the content of political actors' Twitter activities in the European Parliament
    Twitter is an important platform for communication and is frequently used by Members of the European Parliament (MEPs) to campaign and en- gage in discussion with constituents and colleagues in the parliament. Ex- amining the issues that MEPs talk about on Twitter can thus inform us about their political priorities. Topic modelling aims to summarise a corpus of documents by capturing the underlying hidden structure of the data and pre- senting the user with an overview of the key subjects and themes discussed in the corpus, known as topics. This paper aims to quantify and explore the content that MEPs pay attention to on Twitter by applying a new en- semble approach for topic modelling which involves applying two layers of Non-Negative Matrix Factorisation (NMF). The resulting set of issues paid attention to by MEPs are explained by considering the effects of events, is- sue characteristics, and MEP characteristics.
      763