Contextbased Data Mining Using Ontologies Sachin Singh, Pravin Vajirkar, and Yugyung Lee University

1
Context-based Data Mining Using OntologiesSachin
Singh, Pravin Vajirkar, and Yugyung
LeeUniversity of MissouriKansas City.

• Himani Tidke
• CS 586

2
Data mining

• Discovers useful interesting information
• From large collections of data
• Widely used as an active decision making tool.
• Eg Link Discoveries, WALMART

3
Real-world applications of data mining

• Require a dynamic and resilient model.
• Wide variety of diverse and unpredictable
  contexts.
• Huge amount of data .
• Encompass entities that evolve over time.
• Due to dynamic nature of environment, data must
  be interpreted differently depending upon
  situation (context).
• For instance, the meaning of a cold patients
  high fever might be different from the fever of a
  pneumonia patient.

4
Context Based Data Mining

• Consist of circumstantial aspects of the user and
  domain that may affect the data mining process.
• May improve performance and efficacy of data
  mining .
• Context-aware data mining is related to how the
  attributes should be interpreted under specific
  request criteria.

5
Context-aware Data Mining Framework

• Context will be represented in an ontology.
• Context will be automatically captured during
  data mining process.
• Context will allow the adaptive behavior to carry
  over to powerful data mining.

6
Ontologies

• Represent information or knowledge that is
  machine processable and can be communicated
  between different agents.
• We can differentiate the context aware data
  mining into two parts
• Actual representation of the context factor for a
  domain in a corresponding ontology.
• A generic framework which can query this ontology
  and invoke the mining processes and coordinate
  them according to the ontology design.

7
Context-Awareness

• Information has to be conveyed from one element
  to another we need to let the receiving element
  know the reference of our discussion.
• Such as location, environment, identity of people
  and time.
• Determines an application behavior or describes
  where the event occurs.
• Lack of context-awareness leads to missing a lot
  of critical and useful information that would
  affect the data mining process and results.

8
Types of Context

• Domain Context- Target (Patient) Context
• Location context
• Data Context-combination of datasets
• User Context-
• User Identity Context
• User History Context

9
Example of Context-Aware Data Mining

• Doctor wants to know the likelihood of a patient
  having the major blood vessels lt 50 or gt 50
  narrowing as a measure of heart attack risk.
• Attributes of the dataset
• Age, Sex, Symptoms of smoke disease ,
  resting blood pressure, Serum cholesterol,
  Location of the person where he lives, Diagnosis
  of heart (angiographic disease status), etc.

10
Contd.

• Diagnosis of heart is the pivot element (class
  attribute)Rest are query parameters.
• Location Context- not a significant node in the
  classification tree. It can be used to cluster
  data records based on that.Eg Zones, States.
• Domain context(Patient context)- Historical
  patient repository.
• Data Context-Smoke Disease
• The system picks up a dataset referring to
  Smoking Effects, mines it and builds the
  classification tree, which predicts if the person
  has smoking problems.

11
Context-Aware Data Mining Framework

• A set of context factors which may affect the
  behavior of data mining C c1, c2,. . ., cn.
• Context factor takes values c1, c2, . . ck.
• D a dataset composed of a set of tuples, T t1,
  t2, . . ., tn,
• A , a set of attributes a1, a2, . . ., am,
  
• V a set of values for a given attribute aj ,
  v1, v2, . . ., vl.

12
The Model

• Phase 1.-Preprocessing.
• Datasets to be mined are prepared using
  different schemas against tuples (T), attributes
  (A) or values (V ) of available datasets (D)
• Pick
• Join
• Trim

13
The Model

• Phase 2. Data Mining.
• Types of mining processes
• Cascading Mining Process
• Sequential Mining Process
• Iterative Mining Process
• Parallel fork process
• Aggregating Mining Process

14
Architecture
15
Ontology Design
16
Questions?

• Thank You