Introduction ( 1 / 4 )

• Text mining, also known as

  • Text data mining

  • Knowledge discovery

• Text mining

  • Extracting interesting

    and non-trivial patterns

  • Knowledge from unstructured documents

• Extension of data mining

Introduction ( 2 / 4 )

• Text

• Complex task

  • Inherently unstructured and fuzzy

• Commercial potential higher

• 80% of a company's information

Introduction ( 3 / 4 )

• Text Mining is a multidisciplinary field

  • Involving information retrieval

  • Tex analysis

  • Information extraction

  • Clustering, Categorization

  • Visualization

  • Database technology

  • Machine learning

  • Data mining

Introduction ( 4 / 4 )

• Text Mining consisting of two componets

  • Text refining

  • Knowledge distillation

• Products and Applications

  • Text refining

  • Knowledge distillation

  • Intermediate form

A Framework of text mining ( 1 / 3 )

• Consisting of two phases

  • Text refining

  • Knowledge distillation

• Intermediate form

  • Semi-structured

  • Structured

• Intermediate form

  • Document-based

  • Concept-based

A Framework of text mining ( 2 / 3 )

• Document-based

  • Patterns and relationship
    across documents

  • Clustering/visualization and categorization

• Concept-based

  • Derives pattern and relationship across objects or concepts

  • Predictive modeling and associative discovery

A Framework of text mining ( 3 / 3 )

A Survey of text mining products ( 1 / 3 )

• Information retrieval
• Information Extraction
• Categorization
• Summarization

• Document organization
• Visualization
• Navigation

A Survey of text mining products ( 2 / 3 )

3.1. Document visualization

A Survey of text mining products ( 3 / 3 )

3.2. Text analysis and understanding

Open problems and future directions ( 1 / 4 )

4.1. Intermediate form

• Varying degrees of complexity

• Sufficiently rich representation to capture the relationship

• Semantic analysis methods are computationally expensive
  and often operate

Open problems and future directions ( 2 / 4 )

4.2. Multilingual text refining

• Data mining is largely language independent, text
  mining involves a significant language component.

• Text refining algorithms

• Most text mining tools focus on processing English documents

Open problems and future directions ( 3 / 4 )

4.3. Domain knowledge intergration

• Domain knowledge

• Text mining

• Knowledge distillation

Open problems and future directions ( 4 / 4 )

4.4. Personalized autonomous mining

• Future text mining tools, as part of the
  knowledge management systems

• Text mining tools

• Interpret natural language queries and automatically
  perform the appropriate mining operations.