PredicSis.ai
PredicSis | 4.0.0Linux/Unix, Amazon Linux 2017.03.1.20170623 - 64-bit Amazon Machine Image (AMI)
External reviews
External reviews are not included in the AWS star rating for the product.
Highly Recommended
We already used ML / AI as part of our sales marketing processes but we had a packaged solution that was helping us increase our email marketing efficiency. Whilst in one hand that allowed us to understand what we could do with ML, we realised the limitation of what is essentially a black box (eg: limited use case, just a score with no driver, etc...). We were unsure "going solo" was a possibility as we thought it would be too complex and lengthy processes. But how wrong were we! A simple drag and drop process quickly highlighted the key drivers of churn and success our email campaigns. This seriously opens up opportunities for introduction of more ML optimised processes!
- Leave a Comment |
- Mark review as helpful
Questions answered. Simple as that.
At Loyalty Lab, predictive analysis is essential in getting to know our customers' customers. What indicators are key in knowing what theyâÂÂll do next? Until now, the almost mystic process of building our machine learning models was a time consuming job for highly specialised analysts and mathematicians.
Predicsis.ai has changed all this completely. Now, basically anyone who can use a spreadsheet, can also build advanced and highly accurate AI models and put them into action.
Wondering what parameters drive your business and where to improve? Just upload some data, sit back and Predicsis will get your questions answered.
Remarkable tool for data discovery & predictive analytics
Context: we wanted to understand our customers' behaviours but couldn't afford a data scientist.
PredicSis.ai allows us to explain invisible behaviours without the need for any long & complex data science training.
Pros:
- Setup done in a few minutes and then directly operational
- Very easy to use (a few clicks and you have results)
- Models are created very quickly
- Nice rendering with reports and charts
- Customizable (selection of variables, aggregates)
- SDKs available if you want to avoid the UI
A few cons:
- Only supports 2-class classification (for now?)
- User interface could be more friendly