Real-Time Cognitive AI

Cognitive Analytics is a subset of cognitive computing AI that mimics human judgement and perception. Making decisions often involves a Real-Time understanding of context, intent, environmental factors and how they relate to past experiences. Cognitive computing teaches machines how to perform the same fact gathering, semantic analysis and historic data search we use in decision making. Most importantly, a congitive AI platform can identify which facts are relevant. It brings together Machine Learning, Thematic Analysis, Language Processing and traditional BI techniques to improve and automate problem solving.

Cognitive computing systems augment human intelligence, improving our decision making with data-driven insights and machine-guided recommendations. Results are often presented as Knowledge Graphs used to discover patterns, understand meaning based on context and learn about data relationships. The coolest thing about cognitive systems is that they can learn from their own computations. The more data we add to the system, the more connections it forms allowing the AI to constantly adapt to new information.

Classification + Thematic Analysis

Thematic analysis and classification AI teaches machines how to identify and interpret patterns of meaning ("topics" or "themes") in text and documents. Unlike content search which simply counts words or looks for phrases, theme analysis and classification lets you discover explicit and implicit meanings within data.

StreamScape's Application Dataspaces™ are the only data fabric that supports thematic insights with tightly integrated Cognitive AI. Users can easily identify critical topics, relevant terms or features in text and documents, linking them with a abroad variety of systems to unlock the hidden knowledge in enterprise data.

Querying Data by Inference

Inferences are made when a person (or machine) goes beyond available facts or evidence to form a conclusion. The result is inferred knowledge that expands initial information, adding new data into the output. A critical part of accurate inference is being able to describe missing information using familiar terms that are already part of a user's general knowledge.

Dataspaces make use of Inference Types to annotate source data with additional information, filling the knowledge gap. The annotations may be populated by looking up related terms or concepts in a domain-specific semantic graph or resolved using a classification service, allowing the original information to be tagged with synonyms or enriched with relevant features. This new knowledge can be used in data fabric queries to Join or Lookup related information, simulating Pragmatic Inference of human decision making.

Cognitive AI Improves
Quality and Consistency of Human Decisions with Machine Guided Outcomes

Intuitive + Easy to Use

Using graph query languages and ontology tools is hard! In addition to learning new new vocabulary users also need to understand vendor-specific details and syntax. StreamScape data fabric requires no specialized knowledge. Make use of your existing, skills and knowledge of SQL, JSON, XML and scripting languages to work with graph data and drive Machine Learning and AI flows.

Real-Time Intelligence at Scale

Engineered for performance, StreamScape's platform is the only Data Fabric built on a multi-model data engine, capable of massive parallel processing at in-memory speeds. AI powered automation lets you apply intelligence at scale and easily handle complex graph models with expanding data volumes, diverse data sources and types. Dataspace Query Language simplifies integration and data preparation with SQL-like syntax, making it easy to answer complex questions.

Feature Engineering

If you cant identify it, you cant measure it ! The success of any Machine Learning algorithm depends on our ability to identify and analyze meaningful data points or measurements, often referred to as a model's features. Features describe critical things about data that best characterize a problem you are trying to solve or a question you are trying to answer. Errors in Predictive Analytics typically occur because we pick the wrong things to measure.

Data scientists spend most of their time on feature engineering, which can be a slow and labor-intensive process. Training data is often gathered from disparate sources and may require significant preparation before it can be used. But identifying the right features makes a big difference in model accuracy. Feature Engineering is about finding out what measurements are meaningful for predicting outcomes. The Data Fabric lets you identify, store and share model features accelerating the machine learning process.

Cognitive AI Services

StreamScape offers a variety of services for Clustering and Classification, Semantic Graphs, Text Search, Topic and Key extraction as well as Text Summarization and Entity Identification and Extraction. Service results can be combined with inference query and graph analytics to seamlessly blend cognitive AI with traditional BI techniques.