# US 2013/0159310 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 2013/0159310 A1
- Document ID: US20130159310A1
- Status: Published application
- Kind: A1
- Application number: 13/753,653
- Filed: 2013-01-30
- 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/US20130159310A1

## 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 comprises 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. By way of example, one of a criminal activity and a fraudulent activity event, an intrusion event and a fire event and residual objects may be predicted and located and qualified such that, for example, properties of the residual objects may be qualified, for example, via black body radiation and micro-body databases including charcoal assemblages.

## Claims

### Claim 1

A method of detecting a pattern of one of a fraudulent activity and a criminal activity by recording an object in a transaction database, the object comprising at least one suspicious transaction in a transaction database, the transaction database being coupled to processor search manager apparatus, the processor search manager apparatus comprising a processor and memory, the memory coupled to the transaction database, the method comprising
testing by said processor of hypotheses representing a possible organizational structure of a criminal or fraudulent conspiracy, said transaction database comprising one of a similarity-based and a nearest neighbor search; and
applying at least one model to results returned from the database in response to a query, the query requesting objects in said transaction database most similar to a target object, the queried objects, responsive to the query, being most similar to each other, the testing including testing at least one hypothesis and detecting the pattern of one of the fraudulent activity and the criminal activity; the pattern detection comprising a prediction of attributes including two of gender, marital status, occupation and age of an offender of the suspicious transaction and member of the criminal or fraudulent conspiracy.

### Claim 2 (depends on 1)

The method of claim 1, the transaction database comprising purchasing patterns of a plurality of consumers, the method further comprising applying the model to consumer members of a cluster of similar consumer behavior to obtain a relationship comprising a flow of one of capital, goods and services between consumer members of the cluster.

### Claim 3 (depends on 1)

The method of claim 1 an organizational structure hypothesis comprising interactions among a plurality of communication devices forming nodes of a communications network.

### Claim 4 (depends on 1)

The method of claim 1 wherein the transaction comprises a medical record of medical data of a medical claim of a consumer of medical services to detect a pattern of fraudulent activity of the consumer.

### Claim 5 (depends on 1)

The method of claim 1, the transaction database comprising a financial transaction database, the method comprising predicting an instance of medical insurance fraud by one of a consumer and an enterprise responsive to a query of the financial transaction database.

### Claim 6 (depends on 1)

The method of claim 1 wherein the pattern comprises equities that move as a group in a first transaction or move as a plurality of transactions delayed in time with respect to one another.

### Claim 7 (depends on 1)

The method of claim 1 further comprising mining transaction information stored in the transaction database to identify a cluster representing a similar behavior including a purchasing pattern of one of a consumer and an enterprise.

### Claim 8 (depends on 1)

The method of claim 1, the model comprising one of a Bayesian model and a belief network, the model comprising a graph of hypothetical relationships between objects responsive to the query.

### Claim 9 (depends on 1)

The method of claim 1 further comprising performing a search of the transaction database against itself to determine a list of most similar clusters.

### Claim 10 (depends on 9)

The method of claim 9 further comprising adding data to the transaction database comprising one of metadata of the most similar clusters and identification of the list of most similar clusters.

### Claim 11 (depends on 1)

The method of claim 1 further comprising predicting one of an individual perpetrator and an entity engaged in an illicit enterprise.

### Claim 12 (depends on 1)

The method of claim 1 further comprising predicting one of a type of material used to accelerate a fire and a type of carried weapon.

### Claim 13 (depends on 1)

The method of claim 1, the hypotheses comprising first and second hypothesis comprising first and second likelihoods, the set of first and second likelihoods comprising a likelihood ratio.

### Claim 14 (depends on 1)

The method of claim 1 further comprising predicting a property responsive to the query of the transaction database, the predicted property being one of a likelihood, a probability and a belief that a target object an retrieved objects responsive to the query satisfy a postulated relationship across alternative hypotheses.

### Claim 15 (depends on 14)

The method of claim 14, the predicted property being a likelihood, the method further comprising expressing the alternative hypotheses as a likelihood ratio.

### Claim 16 (depends on 1)

The method of claim 1, the nearest neighbor search comprising a similarity measure and a search by said similarity measure of the transaction database.

### Claim 17 (depends on 1)

The method of claim 1, the possible organizational structure of the criminal conspiracy comprising a single criminal perpetrator.

### Claim 18 (depends on 1)

The method of claim 1 further comprising clustering of the data of the transaction database to recognize clustering of fire combustion products and a man-made fire.

### Claim 19 (depends on 1)

The method of claim 1 further comprising a plurality of databases, the plurality of databases including the transaction database, a spectral property database, a micro-body assemblage database and an image database, the method further comprising predicting one of a source, an origin and a location of one of an object responsive to the query with reference to the spectral database, the micro-body assemblage database or the image database.

### Claim 20 (depends on 1)

The method of claim 1 wherein the plurality of databases comprises one of time-series data and frequency-series data.

### Claim 21 (depends on 20)

The method of claim 20 wherein the time-series data comprises time-of-day, day-of-week and instance of one of a new moon and a full moon.

### Claim 22

A method of detecting a pattern of a fraudulent activity by recording an object in a transaction database, the object comprising at least one suspicious transaction in a transaction database, the transaction database being coupled to processor search manager apparatus, the processor search manager apparatus comprising a processor and memory, the memory coupled to the transaction database, the method comprising
testing by said processor of hypotheses representing possible organizational structures of fraudulent activity, said transaction database comprising one of a similarity-based and a nearest neighbor search; and
applying at least one model to results returned from the database in response to a query, the query requesting objects in said transaction database most similar to a target object, the queried objects, responsive to the query, being most similar to each other, the testing including testing at least one hypothesis and detecting the pattern of fraudulent activity.

### Claim 23 (depends on 22)

The method of claim 22, the transaction database comprising purchasing patterns of a plurality of consumers, the method further comprising applying the model to consumer members of a cluster of similar consumer behavior to obtain a relationship comprising a flow of one of capital, goods and services between consumer members of the cluster.

### Claim 24 (depends on 22)

The method of claim 22 an organizational structure hypothesis comprising interactions among a plurality of communication devices forming nodes of a communications network.

### Claim 25 (depends on 22)

The method of claim 22 wherein the transaction comprises a medical record of medical data of a medical claim of a consumer of medical services to detect a pattern of fraudulent activity of one of the consumer and an enterprise.

### Claim 26 (depends on 22)

The method of claim 22, the transaction database comprising a medical transaction database, the method comprising predicting an instance of medical insurance fraud by one of a consumer and an enterprise responsive to a query of the transaction database.

### Claim 27 (depends on 22)

The method of claim 22 wherein the pattern comprises one of equities and money that move as a group in a first transaction or the one of equities and money move as a plurality of transactions delayed in time with respect to one another and the possible organizational structure.

### Claim 28 (depends on 22)

The method of claim 22 further comprising mining transaction information stored in the transaction database to identify a cluster representing a similar behavior including a purchasing pattern of one of a consumer and an enterprise.

### Claim 29 (depends on 22)

The method of claim 22, the model comprising one of a Bayesian model and a belief network, the model comprising a graph of hypothetical relationships between objects responsive to the query.

### Claim 30

A method of detecting a pattern of a criminal activity by recording an object in a criminal transaction database, the object comprising at least one suspicious criminal transaction in the criminal transaction database, the criminal transaction database being coupled to processor search manager apparatus, the processor search manager apparatus comprising a processor and memory, the memory coupled to the criminal transaction database, the method comprising
testing by said processor of hypotheses representing a possible organizational structure of a criminal conspiracy, said criminal transaction database comprising one of a similarity-based and a nearest neighbor search; and
applying at least one model to results returned from the database in response to a query, the query requesting objects in said criminal transaction database most similar to a target object, the queried objects, responsive to the query, being most similar to each other, the testing including testing at least one hypothesis and detecting the pattern of the criminal activity; the pattern detection comprising a prediction of attributes including two of gender, marital status, occupation and age of an offender of the criminal transaction and member of the criminal conspiracy.

### Claim 31 (depends on 30)

The method of claim 30, the model comprising one of a Bayesian model and a belief network, the model comprising a graph of hypothetical relationships between objects responsive to the query.

### Claim 32 (depends on 30)

The method of claim 30 further comprising performing a search of the transaction database against itself to determine a list of most similar clusters.

### Claim 33 (depends on 32)

The method of claim 32 further comprising adding data to the criminal transaction database comprising one of metadata of the most similar clusters and identification of the list of most similar clusters.

### Claim 34 (depends on 30)

The method of claim 30 further comprising predicting one of an individual offender and an entity engaged in an illicit enterprise.

### Claim 35 (depends on 30)

The method of claim 30 further comprising predicting one of a type of material used to accelerate a fire and a type of carried weapon.

### Claim 36 (depends on 30)

The method of claim 30, the hypotheses comprising first and second hypothesis comprising first and second likelihoods, the set of first and second likelihoods comprising a likelihood ratio.

### Claim 37 (depends on 30)

The method of claim 30 further comprising predicting a property responsive to the query of the transaction database, the predicted property being one of a likelihood, a probability and a belief that a target object an retrieved objects responsive to the query satisfy a postulated relationship across alternative hypotheses.

### Claim 38 (depends on 37)

The method of claim 37, the predicted property being a likelihood, the method further comprising expressing the alternative hypotheses as a likelihood ratio.

### Claim 39 (depends on 30)

The method of claim 30, the nearest neighbor search comprising a similarity measure and a search by said similarity measure of the transaction database.

### Claim 40 (depends on 30)

The method of claim 30, the possible organizational structure of the criminal conspiracy comprising a single criminal perpetrator.

### Claim 41 (depends on 30)

The method of claim 30 further comprising clustering of the data of the transaction database to recognize clustering of fire combustion products and a man-made fire.

### Claim 42 (depends on 30)

The method of claim 30 further comprising a plurality of databases, the plurality of databases including the transaction database, a spectral property database, a micro-body assemblage database and an image database, the method further comprising predicting one of a source, an origin and a location of one of an object responsive to the query with reference to the spectral database, the micro-body assemblage database or the image database.

### Claim 43 (depends on 30)

The method of claim 30 wherein the plurality of databases comprises one of time-series data and frequency-series data.

### Claim 44 (depends on 43)

The method of claim 43 wherein the time-series data comprises time-of-day, day-of-week and instance of one of a new moon and a full moon.

### Claim 45

A method of predicting one of a fraudulent activity and a criminal activity event via receiving a query to search manager apparatus coupled to a plurality of databases comprising a core database of property data for objects, the processor search manager apparatus comprising a processor and memory, the method comprising:
determining a set of objects most similar to a target object via said processor search manager apparatus,
applying at least one model to predict a property of the target object,
comparing hypothetical relationships involving the target object and a subset of a set of similar objects to the target object using one of likelihoods, probabilities and beliefs,
determining a probable relationship and a property value of the target object in relation to the set of objects, and
predicting an occurrence the one of the fraudulent activity and the criminal activity event responsive to said determined probable relationship and property value of the target object.

### Claim 46 (depends on 45)

The method as recited in claim 45 wherein the core database comprises a financial transaction database.

### Claim 47 (depends on 46)

The method as recited in claim 46, predicting an occurrence of fraudulent activity comprising predicting an instance of medical insurance fraud by one of a consumer and an enterprise responsive to a query of the financial transaction database.

### Claim 48 (depends on 46)

The method as recited in claim 46 wherein a financial transaction of the financial transaction database comprises a medical record of a consumer of medical services for predicting a pattern of fraudulent activity by one of the consumer and an enterprise.

### Claim 49 (depends on 46)

The method as recited in claim 46 further comprising mining financial transaction information stored in the financial transaction database to identify a cluster representing a similar behavior including a purchasing pattern of one of a consumer and an enterprise.

### Claim 50 (depends on 45)

The method as recited in claim 45 wherein the core database comprises a criminal transaction database, the criminal transaction database comprising crime and offender records.

### Claim 51 (depends on 50)

The method as recited in claim 50 further comprising adding data to the criminal transaction database comprising one of metadata of most similar clusters and identification of a list of most similar clusters.

### Claim 52 (depends on 45)

The method as recited in claim 45, a hypothetical relationship being an organizational structure hypothesis comprising interactions among a plurality of communications devices forming nodes of a communication network.
