RESEARCH

Bidirectional Encoder Representations from Transformers (BERT)

Comparing Bidirectional Encoder Representations from Transformers (BERT) with DistilBERT and Bidirectional Gated Recurrent Unit (BGRU) for anti-social online behavior detection

Continue reading

Economic Uncertainty Identification

Identification of Economic Uncertainty from Newspaper Articles Using State of the Art Models.

Continue reading

Generating Synthetic Comments to Balance Data for Text Classification

Generating toxic comment text using GPT-2 to improve classification when data for one class is sparse

Continue reading

SHOPPER: A Probabalistic Consumer Choice Model

Estimating shopper preferences and price elasticities in a retail grocery setting.

Continue reading

Big Peer Review Challenge

Application of state-of-the-art text classification techniques ELMo and ULMFiT to A Dataset of Peer Reviews (PeerRead)’

Continue reading

State Of The Art Text Summarisation Techniques

Developing a Sequence-to-Sequence model to generate news headlines – trained on real-world articles from US news publications – and building a text classifier utilising these headlines.

Continue reading

Causal Neural Networks

Individual Treatment Effect Estimation using a Residual Neural Network Architecture

Continue reading

Convolutional Neural Networks - sales forecast

This blog post deals with convolutional neural networks applied to a structured dataset with the aim to forecast sales.

Continue reading

Generative Models

This blog post deals with generative models applied to an imbalanced dataset of credit ratings.

Continue reading

Opening the black box of machine learning

Opening the black-box: state-of-the-art in explaining opaque ML/DL models

Continue reading

Crime and Neural Nets

Introducing Recurrent Neural Networks with Long-Short-Term Memory and Gated Recurrent Unit to predict reported Crime Incident

Continue reading

ULMFiT: State-of-the-Art in Text Analysis

Application of state-of-the-art text analysis technique ULMFiT to a Twitter Dataset

Continue reading

Wide and Deep Learning Model for Grocery Product Recommendations

Exploring and applying current trends in machine learning to a large scale product recommendation based on implicit feedback.

Continue reading

Image Captioning

The goal of this blog is an introduction to image captioning, an explanation of a comprehensible model structure and an implementation of that model.

Continue reading

Neural Network Fundamentals

This blog post is a guide to help readers build a neural network from the very basics.

Continue reading