SYSTEMS AND METHODS FOR PREDICTING OUTCOMES USING A PREDICTION LEARNING MODEL

• US 20160267397A1
• Filed: 03/11/2016
• Published: 09/15/2016
• Est. Priority Date: 03/11/2015
• Status: Active Application

• Current Assignee: SymphonyAI Sensa LLC (Fortive Corp.)
• Original Assignee: Ayasdi, Inc. (Fortive Corp.)
• Inventors: Carlsson, Gunnar
• Granted Patent: US 11,868,851 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G05B 23/0281 Quantitative, e.g. mathematical distance; Clustering; Neural networks; Statistical analysis
• CPC Class Codes: G06F 11/3048 where the topology of the computing system or computing system component explicitly influences the monitoring activity, e.g. serial, hierarchical systems
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/45 Clustering; Classification
• CPC Class Codes: G06F 16/55 Clustering; Classification
• CPC Class Codes: G06F 16/65 Clustering; Classification
• CPC Class Codes: G06F 16/75 Clustering; Classification
• CPC Class Codes: G06F 16/906 Clustering; Classification
• CPC Class Codes: G06F 18/2137 based on criteria of topology preservation, e.g. multidimensional scaling or self-organising maps
• CPC Class Codes: G06F 18/23 Clustering techniques
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

A method comprises receiving a network of a plurality of nodes and a plurality of edges, each of the nodes comprising members representative of at least one subset of training data points, each of the edges connecting nodes that share at least one data point, grouping the data points into a plurality of groups, each data point being a member of at least one group, creating a first transformation data set, the first transformation data set including the training data set as well as a plurality of feature subsets associated with at least one group, values of a particular data point for a particular feature subset for a particular group being based on values of the particular data point if the particular data point is a member of the particular group, and applying a machine learning model to the first transformation data set to generate a prediction model.

Claims

1. A non-transitory computer readable medium including executable instructions, the instructions being executable by a processor to perform a method, the method comprising:

receiving a network of a plurality of nodes and a plurality of edges, each of the nodes of the plurality of nodes comprising members representative of at least one subset of training data points, each of the edges of the plurality of edges connecting nodes that share at least one data point of the training data points, the training data set including rows and columns, each row defining a data point of the training data set and each column defining a feature, the training data set including an initial number of columns, each column including values associated with a feature of a plurality of features;

grouping the data points of the training data set into a plurality of groups, each group of the plurality of groups including a different subset of data points of the training data set, each data point of the training data set being a member of at least one group of the plurality of groups;

creating a first transformation data set, the first transformation data set including the training data set as well as a plurality of feature subsets, each of the plurality of feature subsets being associated with at least one group of the plurality of groups, values of a particular data point for a particular feature subset for a particular group being based on values of the particular data point in the training data set if the particular data point is a member of the particular group; and

applying a machine learning model to the first transformation data set to generate a prediction model.

2. The non-transitory computer readable medium of claim 1, the method further comprising:

receiving an analysis data set;

grouping data points of the analysis data set into one or more groups of the plurality of groups;

creating a second transformation data set, the second transformation data set including the analysis data set as well as the plurality of feature subsets, each of the plurality of feature subsets being associated with the at least one group of the plurality of groups, values of a particular data point of the analysis data set for a particular feature subset for a particular group being based on values of the particular data point in the analysis data set if the particular data point is a member of the particular group;

applying the prediction model to the second transformation data set to generate predicted outcomes; and

generating a report indicating one or more of the predicted outcomes.

3. The non-transitory computer readable medium of claim 2, the method further comprising comparing the predicted outcomes to known outcomes to assess the quality of the prediction model.

4. The non-transitory computer readable medium of claim 1, wherein the network of the plurality of nodes and the plurality of edges are a result of topological data analysis applied to the training data set.

5. The non-transitory computer readable medium of claim 1, wherein the network of the plurality of nodes and the plurality of edges are generated by:

receiving the training data set;

generating a reference space;

mapping the data points of the training data into the reference space using at least one filter;

generating a cover based on a resolution;

clustering data in the cover based on a metric and data points of the training data set;

identifying nodes based on the clustered data; and

identifying edges between nodes if nodes share member data points from the training data set.

6. The non-transitory computer readable medium of claim 1, wherein values of a particular data point for a particular feature subset for a particular group are zero if the particular data point of the training data set is not a member of the particular group.

7. The non-transitory computer readable medium of claim 1, wherein values of a particular data point for a particular feature subset for a particular group are null if the particular data point of the training data set is not a member of the particular group.

8. The non-transitory computer readable medium of claim 1, wherein the values of a particular data point for a particular feature subset for a particular group of which the particular data point is a member are weighted.

9. The non-transitory computer readable medium of claim 8, wherein weighting of the values for the particular data point at least partially depend on how many the plurality of groups the particular data point is a member of.

10. The non-transitory computer readable medium of claim 1, wherein the machine learning model is selected from a group consisting of a linear regression machine learning model, a polynomial regression machine learning model, a logistic regression machine learning model, and a random forest machine learning model.

11. A method comprising:

receiving a network of a plurality of nodes and a plurality of edges, each of the nodes of the plurality of nodes comprising members representative of at least one subset of training data points, each of the edges of the plurality of edges connecting nodes that share at least one data point of the training data points, the training data set including rows and columns, each row defining a data point of the training data set and each column defining a feature, the training data set including an initial number of columns, each column including values associated with a feature of a plurality of features;

grouping the data points of the training data set into a plurality of groups, each group of the plurality of groups including a different subset of data points of the training data set, each data point of the training data set being a member of at least one group of the plurality of groups;

creating a first transformation data set, the first transformation data set including the training data set as well as a plurality of feature subsets, each of the plurality of feature subsets being associated with at least one group of the plurality of groups, values of a particular data point for a particular feature subset for a particular group being based on values of the particular data point in the training data set if the particular data point is a member of the particular group; and

applying a machine learning model to the first transformation data set to generate a prediction model.

12. The method of claim 11, further comprising:

receiving an analysis data set;

grouping data points of the analysis data set into one or more groups of the plurality of groups;

creating a second transformation data set, the second transformation data set including the analysis data set as well as the plurality of feature subsets, each of the plurality of feature subsets being associated with the at least one group of the plurality of groups, values of a particular data point of the analysis data set for a particular feature subset for a particular group being based on values of the particular data point in the analysis data set if the particular data point is a member of the particular group;

applying the prediction model to the second transformation data set to generate predicted outcomes; and

generating a report indicating one or more of the predicted outcomes.

13. The method of claim 12, further comprising comparing the predicted outcomes to known outcomes to assess the quality of the prediction model.

14. The method of claim 11, wherein the network of the plurality of nodes and the plurality of edges are a result of topological data analysis applied to the training data set.

15. The method of claim 11, wherein the network of the plurality of nodes and the plurality of edges are generated by:

receiving the training data set;

generating a reference space;

mapping the data points of the training data into the reference space using at least one filter;

generating a cover based on a resolution;

clustering data in the cover based on a metric and data points of the training data set;

identifying nodes based on the clustered data; and

identifying edges between nodes if nodes share member data points from the training data set.

16. The method of claim 11, wherein values of a particular data point for a particular feature subset for a particular group are zero if the particular data point of the training data set is not a member of the particular group.

17. The method of claim 11, wherein values of a particular data point for a particular feature subset for a particular group are null if the particular data point of the training data set is not a member of the particular group.

18. The method of claim 11, wherein the values of a particular data point for a particular feature subset for a particular group of which the particular data point is a member are weighted.

19. The method of claim 18, wherein weighting of the values for the particular data point at least partially depend on how many the plurality of groups the particular data point is a member of.

20. The method of claim 11, wherein the machine learning model is selected from a group consisting of a linear regression machine learning model, a polynomial regression machine learning model, a logistic regression machine learning model, and a random forest machine learning model.

21. A system comprising:

a processor; and

a memory, the memory comprising instructions executable by the processor to perform the steps of;

receiving a network of a plurality of nodes and a plurality of edges, each of the nodes of the plurality of nodes comprising members representative of at least one subset of training data points, each of the edges of the plurality of edges connecting nodes that share at least one data point of the training data points, the training data set including rows and columns, each row defining a data point of the training data set and each column defining a feature, the training data set including an initial number of columns, each column including values associated with a feature of a plurality of features;

grouping the data points of the training data set into a plurality of groups, each group of the plurality of groups including a different subset of data points of the training data set, each data point of the training data set being a member of at least one group of the plurality of groups;

creating a first transformation data set, the first transformation data set including the training data set as well as a plurality of feature subsets, each of the plurality of feature subsets being associated with at least one group of the plurality of groups, values of a particular data point for a particular feature subset for a particular group being based on values of the particular data point in the training data set if the particular data point is a member of the particular group; and

applying a machine learning model to the first transformation data set to generate a prediction model.