Information solutions you can trust
Cognitive Systems
The founders of DSSlab have more than twenty years of experience providing advanced technology solutions for cognitive/semantic information processing problems through ground-up software development and through the innovative use of existing technologies. We specialize in solving tough problems where there exists both definitional (or rule or model) complexity and significant data complexity, as well as complex data usage requirements. Especially where the solution requires a solid logical and mathematical foundation for representation, inference and/or interpretation. To see relevant papers explaining (and patents following) our approach to architecting cognitive systems pls see our LC Type Logic page under resources, or read this draft of a new paradigm for logic based on cognitive processing with extensible types that have both logical and physical characteristics.
LC Type Logic has proven to be a powerful base upon which to construct cognitive/semantic technologies and applications. Using it, we were first in the industry to achieve a number of technical breakthroughs including:
In the area of representation
+Creating a truly multi level information architecture for which we were awarded one of the fundamental patents in the BI space.+Designing and implementing cell level dependency tracking in a multi dimensional environment
+Algorithms and data structures for mapping heterogeneous data sources into a unified semantic representation
In the area of inference
+Providing computational controls over the
processing of missing data and null intersections (illegitimate
formulas in mathematics and logic) without resorting to multi
valued logics that complicate inferencing
+The use of data mining to create aggregation
dimensions for analytical reporting
+Algorithms for calculating semi orthogonal partial
bases in N dimensional categorical data sets - critical for
visualizing high dimensional data and for which we were awarded a
patent
+Algorithms and structures that provide for multi
level reasoning under conditions of uncertainty, of varying
believabilities of different kinds of information as well as
inconsistencies
In the area of interpretation
+The use of a single logical/semantic
representation into which we interpret the contents of spreadsheets
or visual images
+The use of logical type analysis to map data
queries into the same underlying logical/semantic representation as
are mapped visualizations thus providing for the automatic
selection of appropriate visualizations for data
queries
+The use of logical type analysis to map words and
numbers into a common typed representation