Features & Roadmap

GovText Web Portal - Topic modelling with interactive presentation of results

Topic modelling is a statistical method used to uncover hidden themes within a set of documents by identifying collections of related words and phrases. GovText provides visualisations that allow users to explore topic models from both high-level overviews and detailed perspectives. Users can view the various topics and their associated words across the entire dataset, as well as the distribution of these topics within individual documents.

The GovText Web Portal offers two options for Topic Modelling, using the following algorithms:

• Latent Dirichlet Allocation (LDA)
• Correlated Topic Model (CTM)

GovText Web Portal - Text summarisation

Text summarisation is the process of generating a shorter version of a document that remains concise while preserving the essential information from the original.

The GovText Web Portal offers two types of summarisation:

• Abstractive summarization (Normal mode)
  • The key points of an article are paraphrased into a short and coherent paragraph, by a language model pre-trained on news articles and open datasets.
• Extractive summarization (Quick mode)
  • The most important sentences from an article are compiled to form a summary. While this approach does not involve paraphrasing, it offers faster processing speed.

GovText Model-hosting Platform - Development, deployment, and hosting services of customised AI models

The GovText team can assist agencies and central products in developing, deploying, and hosting customised text analytics models on the GovText Model-hosting Platform, enabling client systems to access text analytics services through APIs.

GovText Model-hosting Platform - Prediction of use case results by customised AI models

The models are capable of providing specific predictions for each use case. Examples of currently deployed live models include:

• A feedback case classifier that automatically predicts case categories and sub-categories.
• An agency and case owner classifier that automatically predicts the responding agency or department responsible for handling the feedback.
• An information extractor that automatically retrieves details from feedback.
• A summariser that condenses lengthy feedback email chains into 1-2 sentences for quick reading and inclusion in management reports.