The newest version of the RapidMiner platform solves a series of issues that prevent enterprises from operationalizing AI at scale
September 26, 2019 – Boston, MA – Worldwide spend on AI is expected to grow to $37.5 billion dollars this year (IDC), however there’s been increased awareness of a set of issues that are preventing many organizations from operationalizing the models they’re building and maximizing the investments that are being made. RapidMiner announces new research that analyzes the root causes underlying this trend, as well as a new set of enhancements to its platform that can help enterprises eliminate many of these root cause issues.
“We estimate that businesses are collectively producing, at a minimum, 5 billion models every year,” says Dr. Ingo Mierswa, founder of RapidMiner, “and yet as we speak with new organizations every day, we’re discovering an alarming number of businesses that have a really hard time deploying the worthy models into production where they can have an impact. It’s getting to the point that we view it as an ‘epidemic.’”
According to Gartner1, “even within organizations benefiting from the expertise of mature data science teams, less than half of data science projects end up being fully deployed.”
“The democratization of machine learning platforms is proliferating analytical assets and models. The challenge now is to deploy and operationalize at scale. Data and analytics leaders must establish operational tactics and strategies to secure and systematically monetize data science efforts,” wrote Erick Brethenoux of Gartner, Sr Director Analyst and Research Director on the AI team, with Shubhangi Vashisth and Jim Hare.
Some of the most frequently cited challenges include:
- A shortage of skilled data scientists stretches the resources that are available very thin
- A lack of DevOps or technical skills to manage the deployment of finished models
- Difficulty overcoming political pushback and driving through change management issues
- An inability to demonstrate proven ROI of a particular model before operationalizing it
- Growing concerns over model governance and compliance issues such as bias detection
“In order to stop any epidemic, you must identify and eradicate the source of the outbreak,” says Mierswa. “We worked very hard to identify and analyze all of the issues that are causing this unhealthy situation and we built capabilities that address those root causes. We also published our findings as a research report to help everyone understand the epidemic.” The new RapidMiner Model Impact Epidemic report is available now.
The new product release, RapidMiner 9.4, includes the following features that aim to fix the issues causing the epidemic:
- RapidMiner Go (previously Auto Model Web) is a new browser-based version of the proprietary RapidMiner Auto Model technology, built for business users who know their data and use case, but don’t have advanced data science background. The best way to produce high quality initial models at volume and scale is to democratize data science so that business users can produce accurate and trustworthy models.
- Automatic Model Ops offers an easy way for business users to put models into production. Users can automatically create robust scoring processes, integrate with IT systems, manage and monitor performance on a model leaderboard, and prevent concept drift and bias. Model Ops helps close the feedback loop, so users can create a prediction, act, see impact and improve models over time.
- Profit-Sensitive Scoring is a unique capability which allows business users to input cost and revenue variables in order for the model to self-optimize for profitability. Identifying the ROI of models facilitates better business buy-in to deploy models into production.
- Managed Offerings in the RapidMiner AI Cloud allow users to deploy models into production without acquiring and managing infrastructure. This is important for models built by users without direct access to IT resources, who can now obtain elastic data science services on-demand.
- New Visualizations and Geographic Charts which help to tell a compelling and intuitive story about data and models to facilitate better cross-functional buy-in to deploy a model into production.
RapidMiner has been named a Leader in the Magic Quadrant for Data Science and Machine Learning Platforms for six years.2
Sources: 1Gartner, How to Operationalize Machine Learning and Data Science Projects, Erick Brethenoux, Shubhangi Vashisth, et al., 3 July 2018. 2Gartner, Magic Quadrant for Data Science and Machine Learning Platforms, Carlie Idoine, Peter Krensky, et al., 28 January 2019.
Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner’s research organization and should not be construed as statements of fact. Gartner disclaims all warranties, express or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.
RapidMiner is reinventing enterprise AI so that anyone has the power to positively shape the future. We’re doing this by enabling ‘data loving’ people of all skill levels, across the enterprise, to rapidly create and operate AI solutions to drive immediate business impact. We offer an end-to-end platform that unifies data prep, machine learning, and model operations with a user experience that provides depth for data scientists and simplifies complex tasks for everyone else. The RapidMiner Center of Excellence methodology and the RapidMiner Academy ensures customers are successful, no matter their experience or resource levels. More than 30,000 organizations in over 150 countries rely on RapidMiner to increase revenue, cut costs, and reduce risk.