Data analysis apparatus and method

• US 9,454,595 B2
• Filed: 05/30/2014
• Issued: 09/27/2016
• Est. Priority Date: 05/31/2013
• Status: Active Grant

• Current Assignee: Samsung SDS (China) Co., Ltd. (Samsung Group)
• Original Assignee: Samsung SDS (China) Co., Ltd. (Samsung Group)
• Inventors: Seo, Bum Joon, Kwon, Soon Hwan, Oh, Min Hwan, Kim, Hyung Chan, Oh, Kyu Sam
• Primary Examiner(s): Le, Hung
• Application Number:
  US14/291,855
• Publication Number:
  US 20140358924A1
• Time in Patent Office: 851 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2456 Join operations
• CPC Class Codes: G06F 16/285 Clustering or classification

Abstract

The present invention relates to a heterogeneous data cluster generation apparatus and method and a data clustering method and apparatus, and more particularly, to a data clustering method and apparatus which cluster data measured by different sensors into a number of groups. Aspects of the present invention provide an apparatus and method for generating clusters by putting together heterogeneous data which are values measured by different types of sensors. Aspects of the present invention also provide an apparatus and method for generating clusters by setting indices in order to effectively cluster multi-dimensional data, massive data, or scattered data.

Claims

1. A data analysis apparatus comprising:

a receiver configured to receive first time series data measured by a first sensor and second time series data measured by a second sensor;

a data merging unit configured to generate an ordered pair set using the first time series data and the second time series data; and

a cluster generation unit configured to generate a cluster, using the ordered pair set, the cluster being generated in accordance with one or more locations in a coordinate system having a first axis indicating measured values of the first time series data and a second axis indicating measured values of the second time series data,

wherein the ordered pair set comprises one or more ordered pairs, each including a measured value of the first time series data corresponding to a specific time, and a measured value of the second time series data corresponding to the specific time.

2. The apparatus of claim 1, wherein the second sensor is configured to measure a unit of data different from a unit of data measured by the first sensor.

3. The apparatus of claim 1, wherein the cluster generation unit comprises an index setting unit configured to:

set a minimum value and a maximum value of the first time series data, as a minimum value and a maximum value of the first axis;

set a minimum value and a maximum value, of the second time series data, as a minimum value and a maximum value of the second axis;

generate a grid coordinate system by dividing the coordinate system into grid sections according to a preset sensitivity level, and set a different index for each of the grid sections.

4. The apparatus of claim 1, further comprising:

a normalization unit configured;

to calculate the mean and variance of the first time series data and the mean and variance of the second time series data;

to normalize the first time series data using the calculated mean and variance of the first time series data to provide normalized first time series data; and

to normalize the second time series data using the calculated mean and variance of the second time series data to provide normalized second time series data; and

an index setting unit configured to;

set a minimum value and a maximum value, of the normalized first time series data, as the minimum value and the maximum value of the first axis;

set a minimum value and a maximum value, of the normalized second time series data, as the minimum value and the maximum value of the second axis;

generate the grid coordinate system by dividing the coordinate system into grid sections according to a preset sensitivity level; and

set a different index for each of the grid sections.

5. The apparatus of claim 4, wherein:

the data merging unit is further configured to generate a normalized ordered pair set using the normalized first time series data and the normalized second time series data;

the cluster generation unit is further configured to generate a normalized cluster, using the normalized ordered pair set, the normalized cluster being generated in accordance with one or more locations in the coordinate system with the first axis indicating measured values of the normalized first time series data and the second axis indicating measured values of the normalized second time series data,

wherein the normalized ordered pair set comprises one or more normalized ordered pairs, each including a value of the normalized first time series data which corresponds to the specific time, and a value of the normalized second time series data which corresponds to the specific time.

6. The apparatus of claim 5, wherein the cluster generation unit comprises a representative value setting unit configured to generate a region having the normalized ordered pairs corresponding to the location in the grid coordinate system as the normalized cluster, and to set a representative value, of the normalized cluster, using the normalized ordered pairs existing in the generated cluster.

7. The apparatus of claim 6, wherein the representative value setting unit is further configured to set the normalized ordered pair as the representative value when the cluster comprises only one normalized ordered pair, and to seta mean of the two or more normalized ordered pairs as the representative value when the cluster comprises two or more normalized ordered pairs.

8. The apparatus of claim 1, wherein:

the measured value corresponding to the specific time is data measured at the specific time among the time series data received by the receiver, and

when the data measured does not exist for the specific time, the measured value corresponding to the specific time is taken as a value measured at a preceding time closest to the specific time, among the measured time series data.

9. The apparatus of claim 1, wherein:

the measured value corresponding to the specific time is data measured at the specific time among the time series data received by the receiver, and

when the data measured does not exist for the specific time, the measured value corresponding to the specific time is taken as a value measured at a time closest to the specific time, among the time series data.

10. The apparatus of claim 1, wherein the specific time is based on at least one of:

a measurement time of the first sensor, a measurement time of the second sensor, and a time arriving at specific intervals.

11. The apparatus of claim 1, wherein:

the receiver is further configured to receive third time series data measured by a third sensor, and

the data merging unit is further configured to generate each of the one or more ordered pairs so as to include a measured value of the third time series data corresponding to the specific time.

12. A data analysis method comprising:

receiving first time series data measured by a first sensor and second time series data measured by a second sensor;

generating an ordered pair set using the first time series data and the second time series data; and

generating a cluster, using the ordered pair set, the cluster being generated in accordance with one or more locations in a coordinate system having a first axis indicating measured values of the first time series data and a second axis indicating measured values of the second time series data,

wherein the ordered pair set comprises one or more ordered pairs, each including a measured value of the first time series data corresponding to a specific time, and a measured value of the second time series data corresponding to the specific time;

wherein at least one of the receiving, the generating of the ordered pair set, and the generating of the cluster is implemented using one or more hardware processors.

13. The method of claim 12, further comprising:

measuring the data with the second sensor using a unit of data different from a unit of data measured by the first sensor.

14. The method of claim 12, further comprising:

calculating the mean and variance of the first time series data and the mean and variance of the second time series data;

normalizing the first time series data using the calculated mean and variance of the first time series data to provide normalized first time series data;

normalizing the second time series data using the calculated mean and variance of the second time series data to provide normalized second time series data;

setting a minimum value and a maximum value of the normalized first time series data as a minimum value and a maximum value of the first axis;

setting a minimum value and a maximum value of the normalized second time series data as a minimum value and a maximum value of the second axis;

generating a grid coordinate system by dividing a coordinate system into grid sections according to a preset sensitivity level; and

setting a different index for each of the grid sections.