# US 2013/0159309 A1 · Method and apparatus for predicting object properties and events using similarity-based information retrieval and modeling

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2013/0159309 A1
- Document ID: US20130159309A1
- Status: Published application
- Kind: A1
- Application number: 13/740,400
- Filed: 2013-01-14
- Published: 2013-06-20
- Earliest claimed: 2009-06-25
- Assignees: University of Tennessee Research Foundation
- Inventors: J. Douglas Birdwell; Tse-Wei Wang; David J. Icove; Sally P. Horn
- Web page: https://patentgrep.com/patent/US20130159309A1

## CPC

- G06Q 30/0185: {Product, service or business identity fraud}
- G06F 16/2228: {Indexing structures}
- G06F 16/245: Query processing
- G06F 16/2453: Query optimisation
- G06F 16/285: {Clustering or classification}
- G06F 16/3331: Query processing
- G06F 16/355: Creation or modification of classes or clusters
- G06F 18/2323: based on graph theory, e.g. minimum spanning trees [MST] or graph cuts
- G06F 18/40: Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
- G06Q 50/265: {Personal security, identity or safety}
- G06V 10/7635: {based on graphs, e.g. graph cuts or spectral clustering}
- G06V 10/945: {User interactive design; Environments; Toolboxes}
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Method and apparatus for predicting properties of a target object comprise application of a search manager for analyzing parameters of a plurality of databases for a plurality of objects, the databases comprising an electrical, electromagnetic, acoustic spectral database (ESD), a micro-body assemblage database (MAD) and a database of image data whereby the databases store data objects containing identifying features, source information and information on site properties and context including time and frequency varying data. The method utilizes a model comprising application of multivariate statistical analysis and principal component analysis in combination with content-based image retrieval for providing two-dimensional attributes of three dimensional objects, for example, via preferential image segmentation using a tree of shapes and to predict further properties of objects by means of k-means clustering and related methods.

## Claims

### Claim 1

A method of storing and organizing information stored in at least one database by using a tree-structured index to find and retrieve stored objects having attributes similar to attributes of a target object to mine the at least one database for data leading to a prediction of a property of the target object, the at least one database being coupled to processor search manager apparatus, the processor search manager apparatus comprising a processor and memory, the method comprising
inputting data describing attributes associated with at least two objects to the at least one database;
storing said attribute data in at least one database and referencing this attribute data using a tree-structured index determined by multivariate statistical analysis;
applying multivariate statistical analysis and clustering to said stored data and stored attribute data to determine an indexing and search method for indexing and searching said at least one database using said determined tree-structured index;
using said indexing and search method to find and retrieve stored objects having attribute data similar to the attribute data of a target object; and
applying at least one model to the attribute data of the retrieved stored objects to obtain a predicted property of the target object for output to an output device.

### Claim 2 (depends on 1)

The method of claim 1 wherein the multivariate statistical analysis comprises principal component analysis.

### Claim 3 (depends on 1)

The method of claim 1, the clustering being performed on projections of said stored information and stored information to an attribute subspace determined by at least two principal component vectors.

### Claim 4 (depends on 1)

The method of claim 1 further comprising applying at least two models corresponding to alternative hypotheses and selecting one of the alternative hypotheses to obtain said predicted property.

### Claim 5 (depends on 1)

The method of claim 1 wherein clustering is performed using a k-means algorithm.

### Claim 6 (depends on 1)

The method of claim 1 wherein an attribute of some of the stored objects and the target object is one of a time series and of an electrical or acoustic spectral attribute.

### Claim 7 (depends on 5)

The method of claim 5 where an R2 correlation is used to measure a similarity between attributes that are one of the time series and of the electrical or acoustic spectral attribute.

### Claim 8 (depends on 1)

The method of claim 1 where an attribute of some of the stored objects and the target object is an image, the model comprising an image segmentation model.

### Claim 9 (depends on 1)

The method of claim 1 further comprising describing each object as a vector of attributes, each attribute having a real value responsive to data inputting.

### Claim 10 (depends on 2)

The method of claim 2 further comprising data reduction, the data reduction comprising using principal component analysis to determine a smaller subspace of an attribute space.

### Claim 11 (depends on 10)

The method of claim 10 wherein said data reduction comprises attribute space dimension reduction.

### Claim 12 (depends on 1)

The method of claim 1 further comprising adding data for a first target object to the database and acquiring a further target object for analysis in accordance with the database including the added data for the first target object.

### Claim 13 (depends on 1)

The method of claim 1, the database comprising a plurality of databases, further comprising allocating separate processor search manager resources to each database.

### Claim 14 (depends on 1)

The method of claim 1, the database comprising a plurality of databases, further comprising allocating a search manager utilizing multiple threads to at least one database.

### Claim 15

Processor search manager apparatus for storing and organizing information stored in at least one database of a memory, the processor search manager apparatus using a tree-structured index to find and retrieve stored objects having stored attributes similar to attributes of a target object to mine the at least one database for data leading to a prediction of a property of the target object, the processor search manager apparatus comprising a processor and the memory, the processor search manager apparatus further comprising
an input device for receiving data describing attributes associated with at least two objects to the at least one database;
the processor for applying multivariate statistical analysis and clustering to said stored information and received attribute data to determine an indexing and search method for indexing and searching said at least one database;
the memory for storing a tree-structured index determined by said multivariate statistical analysis and clustering and referencing said attribute data in at least one database;
the processor using said determined tree-structured index of the determined indexing and search method to find and retrieve stored objects having attribute data similar to the attribute data of a target object; and
the processor applying at least one model to the attribute data of the retrieved stored objects to obtain a predicted property of the target object for output to an output device.

### Claim 16 (depends on 15)

The apparatus of claim 15 wherein the multivariate statistical analysis comprises principal component analysis.

### Claim 17 (depends on 15)

The apparatus of claim 15, the processor performing clustering on projections of said stored information and stored information to an attribute subspace determined by at least two principal component vectors.

### Claim 18 (depends on 15)

The apparatus of claim 15, the processor applying at least two models corresponding to alternative hypotheses and selecting one of the alternative hypotheses to obtain said predicted property.

### Claim 19 (depends on 17)

The apparatus of claim 17, the processor performing clustering using a k-means algorithm.

### Claim 20 (depends on 15)

The apparatus of claim 15, the at least one database comprising a plurality of databases, further comprising allocating separate processor search manager resources to each database.

### Claim 21 (depends on 15)

The apparatus of claim 15, the at least one database comprising a plurality of databases, further comprising the processor allocating multiple threads to at least one database.