# US 8,280,899 B2 · Abstracting events for data mining

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

- Number: US 8,280,899 B2
- Document ID: US8280899B2
- Status: Granted
- Kind: B2
- Application number: 12/578,603
- Filed: 2009-10-14
- Published: 2011-04-14
- Granted: 2012-10-02
- Earliest claimed: 2009-10-14
- Assignees: Microsoft Corporation
- Inventors: David Lo; Ganesan Ramalingam; Venkatesh-Prasad Ranganath; Kapil Vaswani
- Web page: https://patentgrep.com/patent/US8280899B2

## CPC

- G06F 16/2465: {Query processing support for facilitating data mining operations in structured databases}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An event is described herein as being representable by a quantified abstraction of the event. The event includes at least one predicate, and the at least one predicate has at least one constant symbol corresponding thereto. An instance of the constant symbol corresponding to the event is identified, and the instance of the constant symbol is replaced by a free variable to obtain an abstracted predicate. Thus, a quantified abstraction of the event is composed as a pair: the abstracted predicate and a mapping between the free variable and an instance of the constant symbol that corresponds to the predicate. A data mining algorithm is executed over abstracted, quantified events to ascertain a correlation between the event and another event.

## Claims

### Claim 1

A method configured for execution by at least one computer processor over a dataset in a data repository, the method comprising:
accessing a raw data record in the data repository, wherein the raw data record comprises a first event, wherein the first event comprises at least one predicate and at least one constant symbol that corresponds to the at least one predicate;
generating a quantified, abstracted event that is representative of the first event, wherein generating the quantified, abstracted event comprises:
using at least one free variable to abstract the predicate; and
generating a mapping between the at least one free variable and at least one instance of the at least one constant symbol to quantify the at least one predicate;
causing the quantified, abstracted event to be stored as a second data record in the data repository;
executing, via the at least one computer processor, a data mining algorithm over the second data record and at least one other data record to determine a correlation between the second data record and the at least one other data record, wherein executing the data mining algorithm comprises comparing the mapping between the at least one free variable and at least one instance of the at least one constant symbol and a second mapping that corresponds to the at least one other data record; and
outputting an indication of the correlation between the second data record and the at least one other data record based at least in part upon the comparing between mappings.

### Claim 2 (depends on 1)

The method of claim 1, wherein the correlation between the second data record and the at least one other data record is a transactional correlation.

### Claim 3 (depends on 1)

The method of claim 1, wherein the correlation between the second data record and the at least one other data record is a temporal correlation.

### Claim 4 (depends on 1)

The method of claim 1, wherein the first event comprises a plurality of predicates, and wherein generating the quantified, abstracted event further comprises:
selecting a subset of the plurality of predicates;
representing instances of constant symbols corresponding to the subset of the plurality of predicates with a set of free variables; and
generating mappings between the set of free variables and the instances of constant symbols that correspond to the subset of predicates.

### Claim 5 (depends on 1)

The method of claim 1, wherein the raw data record comprises an event pertaining to an application programming interface.

### Claim 6 (depends on 1)

The method of claim 1, wherein the first event corresponds to an execution path of a program executed by the computer processor.

### Claim 7 (depends on 1)

The method of claim 1, wherein the second data record is composed as a pair, wherein the pair comprises free variables that represent attributes corresponding to the first event and a mapping between the free variables and instances of constant symbols corresponding to the at least one predicate.

### Claim 8 (depends on 1)

The method of claim 1, further comprising generating the first event by encoding an execution path of an application executed by the computer processor.

### Claim 9 (depends on 1)

The method of claim 1, wherein the data mining algorithm is configured to mine a specification of a software application.

### Claim 10 (depends on 1)

The method of claim 1, wherein the data mining algorithm outputs a rule, wherein the rule comprises an antecedent and a consequent, and wherein a free variable is included in the antecedent and the consequent of the rule.

### Claim 11 (depends on 1)

The method of claim 1, wherein the first event is a tuple of primitive values, wherein a primitive value is a string or an integer.

### Claim 12 (depends on 1)

The method of claim 1, wherein the first event represents a call to an application programming interface method by a client computing device.

### Claim 13

A system, comprising:
a processor; and
a memory that comprises a plurality of components that are executed by the processor, the plurality of components comprising:
an accessor component that accesses a data repository in a computer and retrieves a raw data record from the data repository, wherein the raw data record comprises a first event that has at least one predicate corresponding thereto, wherein the at least one predicate has a constant symbol corresponding thereto;
a representor component that represents the at least one constant symbol in the at least one predicate with at least one free variable;
an identifier component that identifies at least one instance of the at least one constant symbol that corresponds to the at least one predicate;
a mapper component that generates a mapping between the at least one free variable and the at least one instance of the at least one constant symbol; and
a generator component that generates a second data record, wherein the second data record comprises an abstracted, quantified representation of the first event and causes the abstracted, quantified representation of the first event to be stored as the second data record in a data repository, wherein the second data record comprises the at least one free variable and the mapping.

### Claim 14 (depends on 13)

The system of claim 13, wherein the at least one event pertains to a function call in a software program.

### Claim 15 (depends on 13)

The system of claim 13, wherein the first event represents a call to an application programming interface method by a client computing device.

### Claim 16 (depends on 13)

The system of claim 13, wherein the plurality of components further comprises a data mining component that executes a data mining algorithm over the second data record and at least one other data record, wherein the data mining component outputs an indication of correlation between the second data record and the at least one other data record.

### Claim 17 (depends on 16)

The system of claim 16, wherein the indication of correlation comprises an indication of temporal correlation between the first event and a second event represented by the at least one other data record.

### Claim 18 (depends on 16)

The system of claim 16, wherein the indication of correlation comprises an indication of transactional correlation between the first event and a second event represented by the at least one other data record.

### Claim 19 (depends on 16)

The system of claim 16, wherein the at least one predicate is an attribute/value pair of the first event.

### Claim 20

A computer readable data storage device comprising instructions that, when executed by a processor, cause the processor to perform acts comprising:
accessing a first data record in a data repository, wherein the first data record comprises a first event that occurred during execution of a software program, wherein the first event has at least one predicate corresponding thereto, and wherein the at least one predicate has a constant symbol corresponding thereto;
identifying at least one instance of the constant symbol;
representing the at least one constant symbol in the at least one predicate with at least one free variable;
for the at least one identified instance of the constant symbol, generating a mapping between the free variable and the at least one instance of the constant symbol;
generating a second data record in the data repository, wherein the second data record comprises the at least one free variable and the mapping;
executing a data mining algorithm over the second data record and at least one other data record to determine a temporal correlation between the second data record and the at least one other data record; and
outputting the temporal correlation.
