eCommerce Product Categorisation Automation


Infocomm - Video/Image Analysis & Computer Vision
Infocomm - Video/Image Processing
Show more >
New Zealand


The company has designed and implemented a product categorisation solution to help eCommerce retailers automatically categorise products and add product attribute information. 

The technology's artificial intelligence (AI) model has been trained on over 1 million products and has already classified over 116,000 unlabelled products into over 1250 sub-categories. The technology have demonstrated comparable accuracy to human labelling and have saved the company's clients thousands of hours of manual product categorisation.

Additionally, the technology provides potential clients with powerful data for recommendation engines such as the ‘age’ recommendations for children’s toys and baby products.

The technology also support clients in meeting regulatory requirements by identifying missing labels/information such as DVD and video game ratings.

The technology is already available as a working solution in the market, used and loved by real enterprise customers.  


The technology is available as Software as a Service, which is capable of automatically:

• Grouping similar products and assigning them to the relevant product categories

• Labelling product attributes e.g., age, size, colour

• Identification of missing labels and defects

The technology utilises the latest artificial intelligence and machine learning to categorise products based on attributes such as the product image, labels and description.


The technology can be applied for use-cases related to eCommerce / Retail sectors.

Given an eCommerce business owner whose platform may contain over thousands and millions of product types.

Manual tasks such as product categorisation, labelling and identification of certain defects of individual products will be laborious, inefficient and not scalable in terms of manpower in the long run.

The technology’s value creation comes into play in such a setting, by creating a standard way of grouping, categorising, labelling and identification of defects with AI technologies.

Market Trends & Opportunities

The company is a New Zealand AI technology company open to collaboration projects with potential partners who are seeking AI automation technologies in eCommerce and retail sectors to:

• streamline existing process for product categorisation, labelling and identifications of defects

• provide a productivity boost and reduce human errors

• enhance overall work efficiency


Potential clients can save thousands of hours related to manual product inspection and categorisation works by adopting/integrating such technology into their existing infrastructure.

The company is also able to map potential clients’ existing product structure to their pre-existing product ontology tree, thus allowing for potential reduction of development time in a collaboration project.

The company is also able to cater for different levels of predictive granularities e.g., product attributes prediction/labelling based on a potential client’s needs.

The company also possesses existing capabilities and have experience in application development and recommendation engines which will compliment customisation and personalisation of their core AI automation technology based on their partners’ technology needs.

Non-iterative Simultaneous Localization and Mapping (NI-SLAM)
Optimise Utility and Industrial Asset Management with Predictive Analytics Technology
Computer Vision and Video Analytics System for Plant Growth and Disease Detection
Digital Hygiene Map System for Indoor Space
Analysis of Gene Expression Data
Non-invasive Diagnosis of Internal Combustion Engine Performance
Remote Photoplethysmography for Contactless Vital Signs Monitoring
Advanced Analytics and Sensor Platform for Workspace Management
Unified Platform with AI Modules for Management of Smart Estate
Operational Intelligence Platform For Optimization of Water Treatment