Project description
A pioneering predictive model has numerous use cases
Predicting what might happen in the future and with what certainty or margin of error is not only intriguing on a personal level but also highly beneficial when it comes to business. The Italian SME HPA has developed a predictive analytics service to be marketed as a forecast-as-a-service model, which outperforms most competitors in speed and accuracy. Based on cutting-edge deep learning algorithms, its flexibility will be highlighted by application in the energy, technology and media forecasting domains. The EU-funded BEYOND project is helping the team optimise the technology and prepare it for commercialisation.
Objective
HPA S.r.l. has reimagined the way predictive analytics services are made today, with a new Forecast-as-a Service model, based on cutting-edge deep learning algorithms that proved wide applicability in multiple vertical industries.
After validating the problem/solution fit, HPA has engaged 3 prospects in the Energy, Technology and Media forecasting domains to prepare some Proofs of Concept and benchmark the solution. It’s also worth pointing out that no similar tools exist in the case of the second and the third verticalizations.
The novelty of this solution consists of the following unique selling points:
1.BEYOND predictive models outperforms most used competitor solutions in terms of accuracy (on average, 5% of error against 12-38% of errors by competitors) and speed. An early adopter of BEYOND operating in the energy trading market, demonstrated that increased accuracy provided by our model can result in a yearly operating margin increase of at least €120.000.
2.BEYOND will overcome adoption and usage barriers for many businesses providing easy-to-use models that require little knowledge from users with different skill levels from beginners to experienced analysts.
3.As a Forecast-as-a Service model, BEYOND will be offered with a flexible pricing strategy that will allow installation costs drop in a range between 50 to 90 % and long term costs decrease by a minimum of 10% up to more than 70%.
Thanks to the SME Phase1 funding, HPA will be able to: a) finalize the Minimum Viable Product (i.e. development of programming language custom libraries, output formats and integration of third-party datasets); b) design the web application; c) assessment of more industry verticalizations and the search of corresponding prospects; d) set-up a massive promotional/communication plan to reach the global market during the Phase 2.
With a SME Phase 2 funding we will be able to massively market BEYOND, reaching a € 10MLN turnover and 26 jobs by 2023.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
- natural sciencescomputer and information sciencesdata sciencebusiness intelligence
- natural sciencescomputer and information sciencesartificial intelligencemachine learningdeep learning
You need to log in or register to use this function
Programme(s)
Funding Scheme
SME-1 - SME instrument phase 1Coordinator
38122 TRENTO
Italy
The organization defined itself as SME (small and medium-sized enterprise) at the time the Grant Agreement was signed.