The Outcomes and Publication Standards of Research Descriptions in Document Classification: A Systematic Review

Leave a Reply

Abstract

Document classification, a critical area of research, employs machine and deep learning methods to solve real-world problems. This study attempts to highlight the qualitative and quantitative outcomes of the literature review from a broad range of scopes, including machine and deep learning methods, as well as solutions based on nature, biological, or quantum physics-inspired methods. A rigorous synthesis was conducted using a systematic literature review of 102 papers published between 2003 and 2023. The 20 Newsgroups (bydate version) were used as a reference point of benchmarks to ensure fair comparisons of methods. Qualitative analysis revealed that recent studies utilize Graph Neural Networks (GNNs) combined with models based on the transformer architecture and propose end-to-end solutions. Quantitative analysis demonstrated state-of-the-art results, with accuracy, micro and macro F1-scores of 90.38%, 88.28%, and 89.38%, respectively. However, the reproducibility of many studies may need to be revised for the scientific community. The resulting overview covers a wide range of document classification methods and can contribute to a better understanding of this field. Additionally, the systematic review approach reduces systematic error, making it useful for researchers in the document classification community.

Pages: 1 2

Leveraging spiking neural networks for topic modeling

Leave a Reply

Abstract

This article investigates the application of spiking neural networks (SNNs) to the problem of topic modeling (TM): the identification of significant groups of words that represent human-understandable topics in large sets of documents. Our research is based on the hypothesis that an SNN that implements the Hebbian learning paradigm is capable of becoming specialized in the detection of statistically significant word patterns in the presence of adequately tailored sequential input. To support this hypothesis, we propose a novel spiking topic model (STM) that transforms text into a sequence of spikes and uses that sequence to train single-layer SNNs. In STM, each SNN neuron represents one topic, and each of the neuron’s weights corresponds to one word. STM synaptic connections are modified according to spike-timing-dependent plasticity; after training, the neurons’ strongest weights are interpreted as the words that represent topics. We compare the performance of STM with four other TM methods Latent Dirichlet Allocation (LDA), Biterm Topic Model (BTM), Embedding Topic Model (ETM) and BERTopic on three datasets: 20Newsgroups, BBC news, and AG news. The results demonstrate that STM can discover high-quality topics and successfully compete with comparative classical methods. This sheds new light on the possibility of the adaptation of SNN models in unsupervised natural language processing.

Pages: 1 2

Biologically Plausible Learning of Text Representation with Spiking Neural Networks

Leave a Reply

Abstract

This study proposes a novel biologically plausible mechanism for generating low-dimensional spike-based text representation. First, we demonstrate how to transform documents into series of spikes (spike trains) which are subsequently used as input in the training process of a spiking neural network (SNN). The network is composed of biologically plausible elements, and trained according to the unsupervised Hebbian learning rule, Spike-Timing-Dependent Plasticity (STDP). After training, the SNN can be used to generate low-dimensional spike-based text representation suitable for text/document classification. Empirical results demonstrate that the generated text representation may be effectively used in text classification leading to an accuracy of 80.19% on the bydate version of the 20 newsgroups data set, which is a leading result amongst approaches that rely on low-dimensional text representations.

Pages: 1 2

Recognising innovative companies by using a diversified stacked generalisation method for website classification

Leave a Reply

Abstract

In this paper, we propose a classification system which is able to decide whether a company is innovative or not, based only on its public website available on the internet. As innovativeness plays a crucial role in the development of myriad branches of the modern economy, an increasing number of entities are expending effort to be innovative. Thus, a new issue has appeared: how can we recognise them? Not only is grasping the idea of innovativeness challenging for humans, but also impossible for any known machine learning algorithm. Therefore, we propose a new indirect technique: a diversified stacked generalisation method, which is based on a combination of a multi-view approach and a genetic algorithm. The proposed approach achieves better performance than all other classification methods which include: (i) models trained on single datasets; or (ii) a simple voting method on these models. Furthermore, in this study, we check if unaligned feature space improves classification results. The proposed solution has been extensively evaluated on real data collected from companies’ websites. The experimental results verify that the proposed method improves the classification quality of websites which might represent innovative companies.

Pages: 1 2