Back to Newsroom
Back to Newsroom

Applied AI: Neural Networks For Short-Form Text Classification On Kubro(TM) Platform By Robotic Online Intelligence

Friday, 27 November 2020 10:30 AM

HONG KONG SAR, CHINA / ACCESSWIRE / November 27, 2020 / Robotic Online Intelligence Ltd (ROI), an AI-powered business intelligence and analytics startup based in Hong Kong, introduces a neural network it developed and deployed on Kubro(TM) platform, optimised for classification of short-form text, available to clients on a SaaS basis.

In filtering external information for the human consumption in finance and business, it can be very beneficial to be able to classify the information based only on a limited text such news headlines, company announcements or tweets, without the need to access or process the full content.  

The use cases in this context range from e.g. identifying signals of events that might affect a company share price, to the search for particular data points, to determining what an analyst or an executive should spend time reading or be alerted to specific developments.

Automation of such filters can save time and give an information edge.

On Kubro(TM) platform from Robotic Online Intelligence (ROI), in addition to the keywords, expressions and Regex (regular expressions), clients can now also apply the AI to classify and score text in the short form.

ROI's strategy has been to integrate multiple neural network techniques to leverage the power of each of them. The resulting network has a few layers, two of which are a convolutional neural network (CNN) and long short-term memory (LSTM).

A high-level architecture:

Robotic Online Intelligence Ltd, Friday, November 27, 2020, Press release picture

Examples of actual applications include a model that helps determine what the ‘top reads’ should be in a specific domain of China Property or central banks disclosure. In another case, clients can use a model to eliminate irrelevant content from the larger content set to be processed by the analysts looking for particular data points, where it can save 30-50% of the analysts’ time by removing the noise, while losing only 0.9-3% of ‘good items’. In a different example, a model helps determine whether the content meets a certain threshold of relevance to be included in distribution to clients. 

ROI takes a very practical approach to the application of AI.

"The AI model itself is one thing and the practical use of it another. It's important to consider the trade-offs between accuracy and the computing or time resources needed to achieve certain accuracy", commented Robert Ciemniak, Founder-CEO of Robotic Online Intelligence (ROI).

How the results of the AI’s work integrate into a workflow of an analyst or an executive matter too, along with the practical aspects of labelling the data, including the ability to review and test consistency of data labelling done by the analysts.

ROI’s Kubro(TM) platform provides on a SaaS basis an intuitive and easy to use web interface for users to label and train their own models and apply the models as filters on content collections:

Robotic Online Intelligence Ltd, Friday, November 27, 2020, Press release picture

The models can then be used throughout the system to apply to specific content sets, incl. for example robo-reports:

Robotic Online Intelligence Ltd, Friday, November 27, 2020, Press release picture

The models can serve as a filter for picking the ‘top items’ in the Signallium™ market intelligence platform, an example for China Property:

Robotic Online Intelligence Ltd, Friday, November 27, 2020, Press release picture

That very practical aspect of deploying neural networks remains ROI's focus, seeking cost-efficient yet substantial improvements rather than chasing the 99.9% perfection.

The tools and the neural network are fully ready for use on ROI's Kubro™ platform as well as via customized solutions.

ABOUT ROBOTIC ONLINE INTELLIGENCE LTD

Robotic Online Intelligence (ROI), founded in 2017 and based in Hong Kong, provides AI-powered tools for data, research and investment firms, helping them save a significant amount of time in research of publicly available intelligence - through automation and workflow tools for data teams - in the search, collection, filtering, and communication of information.

Robotic Online Intelligence was selected to Cyberport Incubation Programme in Hong Kong in 2019 and is based in Cyberport, Hong Kong SAR.

MORE INFORMATION

Request info or contact: [email protected]
Website: http://www.roboticonline.ai
Follow on Twitter: @ROI_infobesity
Connect on LinkedIn: Connect with Robert Ciemniak / ROI
Ms Lijun Cao
+852 817 01 421

SOURCE: Robotic Online Intelligence Ltd

Topic:
Product Announcements
Back to newsroom
Back to Newsroom
Share by: