Comparison of Topic Modelling Approaches in the Banking Context

Ogunleye, Bayode; Maswera, Tonderai; Hirsch, Laurence; Gaudoin, Jotham; Brunsdon, Teresa

doi:10.3390/app13020797

Cited by 25 publications

(29 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, there has been development of large language models (LLMs), which have taken the world by surprise. The LLMs have been applied to several NLP tasks like topic modelling [40], sentiment analysis [41], a recommendation system [42], and harmful news detection [43]. In the context of CB detection, Paul and Saha [44] compared bidirectional encoder representations from transformers (BERT) to BiLSTM, SVM, LR, CNN, and a hybrid of RNN and LSTM, using three real-life CB datasets.…”

Section: Related Workmentioning

confidence: 99%

The Use of a Large Language Model for Cyberbullying Detection

Ogunleye,

Dharmaraj

2023

Analytics

Self Cite

View full text Add to dashboard Cite

The dominance of social media has added to the channels of bullying for perpetrators. Unfortunately, cyberbullying (CB) is the most prevalent phenomenon in today’s cyber world, and is a severe threat to the mental and physical health of citizens. This opens the need to develop a robust system to prevent bullying content from online forums, blogs, and social media platforms to manage the impact in our society. Several machine learning (ML) algorithms have been proposed for this purpose. However, their performances are not consistent due to high class imbalance and generalisation issues. In recent years, large language models (LLMs) like BERT and RoBERTa have achieved state-of-the-art (SOTA) results in several natural language processing (NLP) tasks. Unfortunately, the LLMs have not been applied extensively for CB detection. In our paper, we explored the use of these models for cyberbullying (CB) detection. We have prepared a new dataset (D2) from existing studies (Formspring and Twitter). Our experimental results for dataset D1 and D2 showed that RoBERTa outperformed other models.

show abstract

Section: Related Workmentioning

confidence: 99%

The Use of a Large Language Model for Cyberbullying Detection

Ogunleye,

Dharmaraj

2023

Analytics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Based on this, scholars have conducted rich research using BERTopic, for example, [22] used tweets from Nigerian bank customers as a dataset and topic modelling. In addition, [23] used clinical records from the Canadian Primary Care Electronic Medical Record for describing and monitoring the primary care system.…”

Section: Bertopic Modelmentioning

confidence: 99%

Identification of Product Innovation Path Incorporating the FOS and BERTopic Model from the Perspective of Invalid Patents

Zhang,

Wu,

Liu

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Under the premise of resource constraint, it is crucial to identify the product innovation opportunities contained in failed patents through external search in order to compensate for the shortcomings of enterprises’ own technology. Due to the cost of patent research and development and the risk of infringement, this paper constructs a product innovation identification path that integrates the FOS and BERTopic model from the perspective of invalid patents. The path consists of three stages, including the identification of the problem to be solved by the product based on functional analysis, the extraction of the subject matter elements based on the core failed patent technology, and the generation and evaluation of innovative solutions based on TRIZ theory and the best- worst method (BWM). Finally, the feasibility of the path constructed in this paper is verified by taking a slurry pump as an example. The application results show that the product innovation identification path constructed in this paper can provide theoretical support for enterprises to carry out technological innovation activities efficiently.

show abstract

“…Cluster analysis is an iterative process that divides data into clusters based on similar attributes [23]. Techniques include density-based, partition-based, and hierarchical-based algorithms.…”

Section: Clustering Techniquesmentioning

confidence: 99%

“…The K-means algorithm enables the identification of homogenous groups within a consumer base [23]. It splits a dataset into discrete clusters based on similarity, enabling personalised marketing activities that cater to individual tastes and requirements [24].…”

Section: K-means Algorithmmentioning

confidence: 99%

An Exploration of Clustering Algorithms for Customer Segmentation in the UK Retail Market

John,

Shobayo,

Ogunleye

2023

Analytics

Self Cite

View full text Add to dashboard Cite

Recently, peoples’ awareness of online purchases has significantly risen. This has given rise to online retail platforms and the need for a better understanding of customer purchasing behaviour. Retail companies are pressed with the need to deal with a high volume of customer purchases, which requires sophisticated approaches to perform more accurate and efficient customer segmentation. Customer segmentation is a marketing analytical tool that aids customer-centric service and thus enhances profitability. In this paper, we aim to develop a customer segmentation model to improve decision-making processes in the retail market industry. To achieve this, we employed a UK-based online retail dataset obtained from the UCI machine learning repository. The retail dataset consists of 541,909 customer records and eight features. Our study adopted the RFM (recency, frequency, and monetary) framework to quantify customer values. Thereafter, we compared several state-of-the-art (SOTA) clustering algorithms, namely, K-means clustering, the Gaussian mixture model (GMM), density-based spatial clustering of applications with noise (DBSCAN), agglomerative clustering, and balanced iterative reducing and clustering using hierarchies (BIRCH). The results showed the GMM outperformed other approaches, with a Silhouette Score of 0.80.

show abstract

Comparison of Topic Modelling Approaches in the Banking Context

Cited by 25 publications

References 62 publications

The Use of a Large Language Model for Cyberbullying Detection

The Use of a Large Language Model for Cyberbullying Detection

Identification of Product Innovation Path Incorporating the FOS and BERTopic Model from the Perspective of Invalid Patents

An Exploration of Clustering Algorithms for Customer Segmentation in the UK Retail Market

Contact Info

Product

Resources

About