Oracle Analytics embeds AI/ML throughout the platform, catering to users of all skill levels, from clickers to coders. Expand beyond built-in AI/ML capabilities with Oracle Database Machine Learning and OCI AI Services to cover a broader range of use cases.
The Oracle Analytics platform has machine learning embedded into every aspect of the analytics process, from data to decision, including recommending insights and supports every role, from clickers to coders. Expand on built-in capabilities with additional OCI services, such as AI services, to address even more AI/ML use cases.
One-click, advanced analytics displays quick forecasts, trend lines, clusters, and reference lines. Users can customize the prediction interval and model type of the built-in algorithms to better fit the data and business use case.
The Explain capability examines the data set to identify meaningful business drivers, contextual insights, and data anomalies with only a few clicks and no coding. Choose visuals and findings from Explain to start a new dashboard and story.
The Auto-Insights capability examines data sets and uses ML to automatically create visual insights with all available metrics and attributes. This can lead to undiscovered connections and patterns in the data that may otherwise not have been considered. With one click, Oracle Analytics will display a range of visualizations with detail descriptions that can be easily added to your project canvas. All calculations used to derive the insights are transparent and editable.
Use the Explain and Auto-Insights capabilities to start your projects with ML–powered, analytics-driven insights and avoid the blank canvas syndrome and biased outcomes.
Use AI-powered natural language processing (NLP) and natural language generation (NLG) along with generative AI–created responses to better interact and understand analytics. Simply use natural language—spoken or search keywords—to query any information from Oracle Analytics data sets and semantic layer, without the need to understand where the data resides or the composition of the data set. Automatically render visualizations in context as the query is being built.
NLG creates smart textual narratives of visualizations that, by default, are connected live to the data source and interact with other data objects on the canvas, such as visualizations and filters. The narrative detail has seven selectable levels, and the description can be set to either “trend” or “breakdown.” Textual narratives are available in multiple languages. Using the mobile app, analytics workbooks can be converted into spoken narratives, such as podcasts.
During data preparation, you can use the data flow editor to choose from training numeric prediction, multi-classifier, binary classifier, or clustering with different built-in algorithms. These ML algorithms can be customized, trained, tuned, and then published to the wider analytics user community. Once models are published, they can be applied to new corporate or personal data sets.
Models trained within Oracle Analytics Cloud can be checked for quality and accuracy. For example, this Naïve Bayes binary classifier has been trained with attrition data, and the quality of the algorithm is rated against the known true values from the test data.
Access the depth and sophistication of Machine Learning in Oracle Database, part of Oracle Autonomous Database. Machine Learning in Oracle Database provides a centrally governed platform to develop, test, and publish ML models using SQL, R, Python, REST, and AutoML. It gives business users the flexibility of self-service when preparing their data. Published models can then be registered into Oracle Analytics Cloud for wider business populations to access and execute with their own data sets.
Oracle Analytics Cloud integrates with OCI AI Services, including OCI Vision and Document Understanding. These integrations extend OAC's existing embedded machine learning capabilities to include an even broader range of business use cases. Use pretrained models or design, refine, and deploy custom models, registering them within OAC for direct access by business professionals. With Document Understanding service, you can apply AI models to documents—such as JPEG and PDF files—and extract key values and their context. This helps organizations unlock information from documents to generate additional insights, even if the information has not been recorded in a central database. These dynamic, self-service approaches reduce business users’ reliance on your data science team for routine, repetitive model execution and results delivery. Business users can independently schedule and execute their models, enabling data science experts to focus on more strategic tasks.