Text Classification

Text Classification

Focus

Development of AI models to classify scraped content from documents and web pages.

Approach

The client was an online platform that collected, stored and analysed a vast quantity of media data from a significant number of webpages and sources.

The analysis and filtering of this information required a sophisticated classification model for characteristics such as industry and content tags for documents, media releases and social media posts.

The solution involved the experimentation then development of a natural language processing techniques utilising neural networks and support vector machines with word embeddings.

Outcome

• Increased accuracy of classification
• Enhanced platform experience