Artificial Intelligence for market moving news identification

AI has come a long way – it can spot cancer, recognize people and translate foreign languages with a click.

PreMarketCall reads news from multiple sources to identify what’s relevant and what’s not. Using enhanced modern techniques developed by experts in deep learning, financial markets and news, we identify what others miss.

SOTA

 

State of the art performance

Our developers aim to use the latest algorithms in deep learning and natural language processing to create the best insights possible.

Natural Language

Reads, parses and really understands the news using published and unpublished alrogithms

Python

Developed using Python, CUDA, Theano, Tensorflow and Pytorch with the latest NLP algorithms and word distributions

Identify insights

Using latent space analysis we predict the events behind the news and their market making impact

Network analysis

Network analysis between stocks, topics and entities.

Deep Learning

Using the latest in generative and adversarial training, combined with reinforcement and transfer learning.

Topic and trends

Dirichlet multinomial generative models combine with distributed representations to get the best insights

Published works

Some of our developers have published in top tier journals and conferences and bring unpublished enhancements

CUDA

Developed using CUDA 1080 graphics cards for faster learning and predictions

Scour the news

Information overload is common among portfolio managers and traders. There is always more research to be done than hours in the day.

AI can solve this. By reading more news and analyzing its impact using statistically relevant models, we can target and filter what news is important and what news is junk.

This saves time and money, and helps increase returns.

Modular analysis

PreMarketCall brings in data from diverse sources and processes them through modular stages

This modular design allows us to continually update and improve our analysis as new algorithms and techniques become available and are developed in-house.

Multiple devices

PreMarketCall works on multiple devices including PC, laptop, Mac, Ipad and Android with a fantastic API

This allows news and analysis to be used by portfolio managers, traders, in meetings or for discussions over lunch. The API allows for easy integration with dynamic systems and risk management.

Achieve your goals

Analyze more data, read more news and don’t miss a thing

PreMarketCall simply reads faster than you can. Achieve your goals by spending less time doing what a computer can now do. You don’t cut your lawn with nail scissors so let technology work for you.


Get the results you want

Don’t be left behind the AI wave. Developed by PhD graduates in deep learning and natural language processing, PreMarketCall will help keep you ahead of the technology cuve

FAQ

Who Is PreMarketCall ?

PreMarketCall is a small team of developers and market experts who combined to produce and deliver the latest in Artificial Intelligence solutions for the financial market

Does it Work ?

Yes, PreMarketCall uses proven algorithms in deep learning and natural language processing

How much does it cost ?

Surprisingly little. Give us a call.

How does it work

We scour the news using dozens of machine learning algorithms to predict which news is important for you

Do you need my data ?

No, the system can work from a local server in your office so your data never leaves your place

How old is it ?

We started working on PreMarketCall in 2010 when deep learning was still new. Since then we have replaced many algorithms many times.

Why not use it yourself ?

We do use it ourselves, but gain greater leverage letting others use it too.

Free Courses

Duis egestas aliquet aliquet. Maecenas erat eros, fringilla et leo eget, viverra pretium nulla. Quisque sed augue tincidunt, posuere dui tempor.

Premium Courses

Duis egestas aliquet aliquet. Maecenas erat eros, fringilla et leo eget, viverra pretium nulla. Quisque sed augue tincidunt, posuere dui tempor.

Ready to get started?

Get in touch, or create an account