Most organizations struggle with all manners of decision making, but I think the most complex one relates to planning for the future. These decisions not only require high quality, complete, and governed data sets but they also demand the combination of predictive and prescriptive capabilities that enables organizations to respond rapidly to changing requirements and constraints. In the last few years, we have seen many AI and ML-based predictive solutions popping up in the market. But the adoption of these solutions is still low, and more importantly usage of their output in decision making to compel action is not happening. According to Gartner, as of November 2020, “Only about 1 in 5 CIOs who thought they would employ AI within the next 12 months actually achieved that, especially in the 2019-2021 timeframe.”
We believe our platform for Active Intelligence, when combined with Big Squid’s no-code automated machine learning, is going to change this paradigm.
Here are five areas where I think we’ll see significant momentum in fueling actions with AutoML.
- Lowering barriers for predictive model creation. ML-based prediction requires expert skill sets. A common ML pipeline is made up of pre-processing, feature extraction, feature selection, feature engineering, algorithm selection, and hyper-parameter tuning. These are the activities that data scientists conduct across the world, and these skill sets in short supply. Big Squid’s no code AutoML solution removes some of these constraints by significantly reducing the time it takes to typically implement an ML process under human supervision, but also improving the accuracy of the model when compared to those trained and deployed by humans.
- It all starts with Data: While AutoML lowers the barriers for the creation of predictive models, ML pipelines start with data. Just like any other analytics artifact development, data plays a key role in any ML use case. 60% of the work of a data scientist lies in collecting the data. According to Gartner, data quality problems and data accessibility challenges are among top reasons holding back the adoption of ML. This is where Qlik Cloud’s catalog capabilities will play an important role. Citizen data scientists can start right from the catalog, where they can shop and see the profile of the data, understand possible data quality issues, any security attributes, and popularity of the data. Qlik Cloud catalog, paired with Big Squid’s data science techniques to pre-process data and to make it model ready is a unique user value.
- Governance and Explainable AI: Another particular challenge that stands out for ML adoption is governance and explainability. A lack of robust governance doesn't just limit the potential success of an organization AI/ML initiative; it could put the entire business in peril as well. By integrating Qlik Cloud catalog, and our recent NodeGraph data lineage capabilities with Big Squid’s explainable AI approach, our users will gain governance on the manageability and access of the models and gain full visibility on the data flows, model performance metrics, feature importance, and will have full feature explanability to increase trust in both the data and the models.
- What If scenarios and Collaboration: Big Squid’s What-if decision planning feature lets users test their ideas for actions based on the predictive results. With Scenario Builder, users can play out predictions and compare scenarios side-by-side. By combining this unique What-if decision planning capability with our Collaborative notes, our users will make the best decisions possible with confidence and while also leveraging the wisdom of crowds as needed.
- Taking AI consumption to the next level: With the exciting release of Qlik Application Automation, we see a great opportunity for taking the AI consumption to the next level and triggering AutoML-based business processes on cloud with Big Squid’s API deployable AI models. Most businesses now plan to increase their automation investment to help increase their agility and improve their ability to quickly handle changing conditions. Being able to trigger actions and processes, based on AutoML generated predictions, will be a game changer.
Most organizations struggle in going beyond core analytics efforts to generate the understanding from their data of what might happen and why it could happen. With Big Squid’s SaaS no-code AutoML capabilities, our users will be able to seamlessly and easily augment their analytics efforts directly within Qlik with AI-driven models. This value gets even more exciting when it is combined with Qlik’s platform, taking advanced analytics to the next level with a robust data fabric, governance, collaboration, and automation. Stay tuned!