17945663. MACHINE-DERIVED INSIGHTS FROM TIME SERIES DATA simplified abstract (International Business Machines Corporation)
Contents
- 1 MACHINE-DERIVED INSIGHTS FROM TIME SERIES DATA
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 MACHINE-DERIVED INSIGHTS FROM TIME SERIES DATA - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
MACHINE-DERIVED INSIGHTS FROM TIME SERIES DATA
Organization Name
International Business Machines Corporation
Inventor(s)
Sumanta Mukherjee of Bangalore (IN)
Bhanukiran Vinzamuri of Long Island City NY (US)
Vikas C. Raykar of Bangalore (IN)
Arindam Jati of Bengaluru (IN)
Nupur Aggarwal of Bangalore (IN)
MACHINE-DERIVED INSIGHTS FROM TIME SERIES DATA - A simplified explanation of the abstract
This abstract first appeared for US patent application 17945663 titled 'MACHINE-DERIVED INSIGHTS FROM TIME SERIES DATA
Simplified Explanation
The abstract of the patent application describes a method for deriving insights from time series data using a machine learning model based on subject matter expert input and a rule-based translator. The model template generated from the SME input is used to create a neural network that extracts components from the time series data and determines their contribution to a decision.
- Method for deriving insights from time series data:
* Receive subject matter expert input characterizing aspects of the time series * Generate a model template using a rule-based translator * Create a machine learning model based on the model template, such as a multilayer neural network * Extract components from the time series data using component definition layers * Determine the contribution of each component to a decision * Generate an output including the component-wise contribution of the components
Potential Applications
This technology can be applied in various fields such as finance, healthcare, and manufacturing for predictive analytics, anomaly detection, and forecasting.
Problems Solved
This technology helps in efficiently analyzing time series data, extracting meaningful insights, and making informed decisions based on the contributions of different components.
Benefits
The benefits of this technology include improved accuracy in decision-making, enhanced understanding of time series data, and automation of insights generation.
Potential Commercial Applications
- Predictive maintenance in manufacturing
- Stock market trend analysis
- Disease outbreak prediction in healthcare
Possible Prior Art
One possible prior art for this technology could be the use of machine learning models for time series analysis in various industries.
Unanswered Questions
How does this technology handle noisy time series data?
This article does not address how the machine learning model deals with noisy data and whether any preprocessing techniques are used to clean the data before analysis.
Can this technology be applied to streaming time series data?
The article does not mention whether the method described can handle real-time streaming data or if it is limited to static time series datasets.
Original Abstract Submitted
Deriving insights from time series data can include receiving subject matter expert (SME) input characterizing one or more aspects of a time series. A model template that specifies one or more components of the time series can be generated by translating the SME input using a rule-based translator. A machine learning model based on the model template can be a multilayer neural network having one or more component definition layers, each configured to extract one of the one or more components from time series data input corresponding to an instantiation of the time series. With respect to a decision generated by the machine learning model based on the time series data input, a component-wise contribution of each of the one or more components to the decision can be determined. An output can be generated, the output including the component-wise contribution of at least one of the one or more components.